Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
APETT Engineering
Magazine June 2019
June 2016 Edition
June2019
Edition
The Association of
Professional Engineers of
Trinidad and Tobago
APETT’s Mission:
The Association of Professional Engi-neers of Trinidad and
Tobago is a learned society of profession-al engineers dedicat-
ed to the develop-ment of engineers and the engineering
profession. The asso-ciation promotes the highest standards of professional practice
and stimulates awareness of tech-nology and the role
of the engineer in society.
ISSUE 6
Dec 2018/ Jan 2019 Edition
June 2019 Edition
Issue 7
Page 2 APETT Engineering Magazine June 2019
TABLE OF CONTENTS
DISCLAIMER: Statements made and information presented by contributors to this Magazine does not necessarily reflect the views
of APETT, and no responsibility can be assumed for them by APETT or its Executive Members and Editors.
Page 5
Page 8
Page 11
Page 17
Page 22
Page 26 Page 26 Page 29
Page 26
ARTICLE
What's the Difference Between a 'Proof of Concept' and a 'Prototype'? By: Arshad Mohammed
Make the Most Out of Your Internship
By: Brenton Chatoo
Design and Selection of Choke Valves Part II By: Avinash Babwah
Automated Identification of Vehicular Accidents from Acoustic Signals
Using Artificial Neural Networks By: Aaron D’Arbasie and Renique Murray
Bioplastics and Environmental Sustainability: Some Thoughts
By: Trishel Gokool
Assessing The Sensitivity Of The East Coast of Trinidad To Oil Spills By: Nyoka Sinanan
A Sustainable Solution for Red Mud in the Caribbean
By: Gino Hosein and Professor Indrajit Ray
KMA In-Plant Cold Recycling Technology In Road Construction and Rehabilitation In Trinidad ad Tobago
By: Laurence Bridgemohan
The “Human” case for Energy Efficiency in Trinidad and Tobago By: Sheena Gosine
Hand-Held NPK Sensor
By: Dillon Boodoo
Article References
PAGE 5 8
10
16
25
28
35
37
40
45
53
Editor’s Message
Eng. Vicard Gibbings Page 3
Eng. Vicard Gibbings is a
Process Engineer with ap-
proximately 2 years’ experi-
ence in onshore and offshore
Oil & Gas and Petrochemical
Facilities. Experience ranges
from Conceptual Study,
Front End Engineering and
Design (FEED) through to
Detailed Engineering Design.
Experience includes: devel-
opment/review of Process
Engineering documents,
equ ipm ent/ va lve/ l ine/
restriction orifice sizing, de-
v e lo pm ent / upda t ing / a s -
building P&IDs; process safe-
ty support for PHAST conse-
quence modeling and subse-
quent firewater system de-
sign utilizing Pipenet. He is
also experience in Aspen
HYSYS software modeling a
range of process unit opera-
tions. In addition, he handles
project engineering duties if
required.
Apart from his normal day-
job, Vicard is a part-time
tutor, actively involved both
in IChemE TTMG and
APETT enjoys learning dif-
ferent things through online
courses.
GREETINGS! It has been my great pleasure working with the Chemical Division Team to present you the
very seventh edition of the APETT’s Bi-annual Engineering Magazine publication. Since I took over as the
Editor-in-Chief in January 2019, we have successfully made the smooth transition without too much disrup-
tion of the review flow, thanks to a few individuals who have helped me during this process. They are Julio
Bissessar (previous Editor-in Chief), Janelle Ramdahin, Therese Lee Chan, Nishawn Hanif. Sara Bernard
(Chemical Division Chair). Our former Editor-in-Chief, Julio Bissessar, continued behind the scenes to assist
whenever I needed help so that I could familiarize myself with the editorial procedure.
One of my goals as editor is dedicated to the rapid dissemination of high quality articles / papers on how
advances in STEM can help us meet the challenges of the 21st century, and to capitalize on the promises
ahead. We welcome contributions that can demonstrate near-term practical usefulness, particularly contri-
butions that take a multidisciplinary / convergent approach because many real world problems are complex
in nature.
In the past couple decades, we have witnessed significant issues in the real world. There are also the on-
going issues, such as energy, food due to rapid urbanization, management of resources, biological diseases
and environmental issues like the current debate about climate change. As engineers, scientists and re-
searchers, we aim to seek ways to harness the power of STEM to meet some of these real world challeng-
es, and to provide substance for making informed judgments on important matters.
Innovation and technological development will be essential for a global human benefits. In fact, historical
efforts in such sectors have already brought important results. Basic and applied research and development
(R&D) in science and technologies have made today. Innovation and R&D will be needed to continue im-
proving their efficiency, but greater efforts will be required to commercialize breakthroughs in all fields of
applications.
APETT provides an ideal forum for exchange of information on all of the above topics and more, in various
formats: full length and letter length research papers, survey papers, work-in-progress reports on promising
developments, case studies / best practice articles written by industry experts, and tutorials on up-and-
coming technological breakthroughs. Thus the APETT Engineering Magazine is a platform used to showcase
any breakthrough development or solutions to current and future problems and will be published two times
a year.
In STEM, as in most human endeavors, quality is more important than quantity. As stewards of APETT, the
editors have a fiduciary responsibility to the readership to ensure that only the very best STEM appears in
the journal. In a very real sense, the editors work for the readers; their charge is to select papers rigorous-
ly, publishing only truly new or novel information that constitutes an important conceptual advance in rela-
tion to existing knowledge, so that the readers’ time is spent wisely. In an increasingly busy and competitive
environment, the readers’ decision to look at our journal must be worth the effort.
The current “Professional Development” section alludes to the views of the real world. I am encouraging
different sections in the Engineering Magazine for the readership such as important events and STEM ad-
vances using a fresh, informal, conversational style. The content is miscellaneous: news of broad general
interest to the STEM community (including international news); major recent discoveries in the STEM field
as well as ground-breaking discoveries; highlights of the most exciting basic and translational science pre-
sented at STEM meetings; and commentaries by nominated STEM leaders. I wish to stress that this sec-
tion (like all sections of the journal) will not be used to push a political or ideological agenda. I believe that
journals / magazines ought to respect intellectual diversity, not only because of the obvious reason that it is
fair but also because exposure to different ideas is enriching. At APETT, we will not push political views or
ideologies of any kind; instead, we will strive to be inclusive, to respect diversity of opinions, and to the
extent possible and present all sides of an issue fairly.
Finally, we wish to encourage more contributions from the STEM community to ensure a continued success
of the journal / magazine. Authors, reviewers and guest editors are always welcome. We also welcome
comments and suggestions that could improve the quality of the journal.
Thank you and I hope that this article is enjoyable and informative!
Message from APETT’s President
Page 4
I was a member of the Executive Council that made the decision to pub-
lish the APETT Engineering Magazine. As I recall, the Chemical Division had successfully published a magazine focused on their specific discipline.
The feedback from that publication was extremely positive.
A proposal was then put forward for the magazine to be made open to all the remaining divisions. I am pleased with the success of the magazine to
date as it continued to grow from strength to strength.
This biannual publication has proven to be a significant mechanism utilised by the association in its efforts to fulfil its mission to stimulate “awareness
of technology and the role of the engineer in society.”
We are truly thankful to our contributing writers who have provided in-
teresting articles based on their research and experiences in the field. It is
also noteworthy to mention the number of young engineers who have
provided articles for the magazine. It is my hope that their articles not
only serve to inform you, but also to inspire more of us to submit articles
for consideration.
I take this opportunity to thank our former Magazine Editor, Eng. Julio
Bissessar for his contribution to the magazine. I have no doubt that our
newly appointed editor Eng. Vicard Gibbings will continue in a similar
strain and build on the foundation set by Eng. Bissessar.
I would like to remind our members and supporter that APETT will be hosting our annual Honours and Awards function later this year. Infor-
mation about the event shall be disseminated to the membership in due
course.
We look forward to seeing you, our valued members, at this prestigious
event.
Eng. Vince Ramlochan possesses
over 18 years' experience as a prac-
tising Civil/Structural Engineer and
Project Manager. He is an Associate
of C.E.P. Limited, a Civil and Struc-
tural Design and Project Manage-
ment firm.
He has been involved in the civil and
structural design of several promi-
nent commercial and residential
projects as well as structural assess-
ments of existing structures. He has
also acted as Project Manager/
FIDIC Engineer for several promi-
nent construction projects.
Vince graduated from the University
of the West Indies (UWI) with a
BSc. (Hons) in Civil Engineering. He
is a Registered Engineer with the
Board of Engineering of Trinidad and
Tobago (BOETT) and a Member of
the Association of Professional Engi-
neers of Trinidad and Tobago, the
American Society of Civil Engineers
and the Project Management Insti-
tute. He is an Associate Member of
The Chartered Institute of Arbitra-
tors. Vince previously served on the
Executive Council of APETT as
Assistant-Secretary, Honorary Sec-
retary and Vice-President.
Vince is the current President of the
Association of Professional Engi-
neers of Trinidad and Tobago.
Page 4
Page 5 APETT Engineering Magazine June 2019
Page 6 APETT Engineering Magazine June 2019
What's the Difference Between a 'Proof
of Concept' and a 'Prototype'?
By: Arshad Mohammed, Graduate Research Assistant
When designing, I hear people using these terms, proof of concept and prototype, loosely when they are not
the same thing, thus I am writing this short and informal piece to explain the differences.
Proof of concept:
Proof of concept is usually done or built before any prototype. It is used to validate that a function of the envisioned
product can work. Let's say you are designing a new type of wind turbine to generate electricity to power homes in a
certain area, a proof of concept could just be using a rough shape or jury-rigging another small turbine, not necessarily
your own, to see if it is possible. Proof of concept could be held together by rubber bands and paperclips; it is your idea
broken down to its simplest and most fundamental form to see how well it holds up.
Problems arising in this stage is usually not design related but rather more to do with the practicality of your idea within
the context you would like to implement it. For example, you may find that during certain times of the day there is no
wind or that there are a lot of birds in the area that fly close to the turbine. Many times, issues arise due to the human
element; the community may not like the sight of the wind turbine in the area.
The information you gather here, problems arising, observations and success of certain functions can be used to refine
your idea before creating a prototype.
Prototype:
Types of prototypes include physical and non-physical. A physical prototype is the classical ‘prototype’ that you build and
put in place to test. Non-physical can be your Computer-Aided Design and Drafting (CAD) models and only exist in the
computer or even in mathematical equations. Do not underestimate the power of a CAD prototype though. You can
test pretty much all your working conditions in there such as wind speeds and directions or even how well your design
would hold up against a bird strike.
The design can be changed and retested in a matter of minutes before spending time and money on building the physical
version. There are two sub-types of physical and non-physical prototypes: prototype for form and prototype for function.
A prototype for form is one that does not work but is there to make sure the shape and size are good. For example, you
can use it to make sure your turbine fits in the required space or even use it to gauge how receptive people are to how
your design looks. A prototype for form does not have to work, it just needs to look the part.
On the other hand, a functional prototype does not need to look exactly like the finished product, it just needs to work.
In the case of the turbine, the blade shape is critical so that may remain the same, but mounting brackets and other hard-
ware need not look like the final intended product. It’s important to note that prototypes for form or function can fall
into both the physical and non-physical categories.
Page 7
Bonus Round
Concept:
The concept is the idea that you came up with but have not developed as yet. This could just be a drawing scribbled hast-
ily on a napkin. It’s one of the first steps when designing new products.
Teams of designers are encouraged to come up with as many concepts as they can. During the concept generation phase
no concept is turned away, instead they all go through a process to narrow down the most appropriate. Innovative ideas
from concepts that would otherwise be discarded can be combined into other concepts.
It is for this reason at the early stages, no concept is rejected.
Pretotype:
No, I didn't misspell that. It is essentially ‘faking it til you make it’. It's a term coined by Alberto Savoia in 2009. I love it
for its ability to gather useful customer information before investing significant time and resources. The most famous
example is IBM having customers try out speech to text before the technology capability even existed. They put persons
in a room with a computer monitor and a microphone but no keyboard.
As the person spoke, the words would magically appear on the screen. These persons did not know that someone in
another room was listening and typing for them. It allowed IBM to discover problems such as how to delete something
once it was spoken.
This meant that as soon as the technology advanced to the point where this was possible, IBM had a massive head start
over their competitors who would have been finding those problems for the first time. Using a pretotype is a great way
to get funding for your idea; by showing potential investors what your product can do without actually having the prod-
uct. We see this used a lot in crowd funded products.
My advice when using this strategy is to ensure it’s made clear at some point that the product does not yet exist. I know
I’ve left out a lot of stuff so don’t take this to be a complete resource. I encourage you to look up more information
online if you are designing a product. Doing the proper steps builds a solid foundation and would enable you to have a
much clearer sense of the direction in which you want to take your idea.
Arshad Mohammed has been teaching
courses related to product design and product realization for the last five years and is cur-rently pursuing his PhD in manufacturing at the
University of the West Indies This, coupled with working with people for his design and 3D printing business have revealed many mis-
understandings of the field among persons looking to create their own inventions.
Page 8 APETT Engineering Magazine June 2019
Make the Most Out of Your Internship
By: Brenton Chatoo, Junior Mechanical Engineer
Acknowledgments: I would like to extend a special thanks to Andy Newton, Kiran Rampat, Nicholas Burke, Adesh
Jaikaran and the entire staff of Nutrien Trinidad.
Congratulations! You’ve just been hired for an Internship at Nutrien. “You must be very excited to work alongside the
top professional engineers in your area of study”, were my first thoughts. On the 4th day of June, 2018, I started working
as an Intern attached to the Inspection Department located at Mediterranean Drive, Point Lisas where I worked closely
with Inspection Specialists and Reliability Specialists. To my mind I thought it to be a typical Internship with the usual du-
ties such as backlogging, filing and if I’m lucky enough maybe data entry and that’s what happened.
My very first task at Nutrien was to rearrange a library situated in the Drawing office which was shared with the Inspec-
tion office. The task was very tedious and cumbersome and I would have loved to do something else. Within the first
week I realized this internship was going to be different, not because I was receiving satisfactory work but because every-
one, and I stress, everyone was always busy. This was slightly intimidating as I really wanted to give my best impression.
This high level of productivity kept me motivated and really impacted my work ethic which made me appreciate all the
tedious work given to me and it made me realize one thing. As an intern, the work is not as exciting for the most but
someone has to do it. I could not imagine any of the other employees sparing 10 minutes to attempt the tasks given to
me. At this point I drew a line between Intern jobs vs Professional jobs. As an intern you are supposed to do these jobs
so that the professionals’ jobs can be easier and always remember, someone has to do it. It helps build character and you
will have a greater appreciation for all the mechanisms involved at Nutrien. The job has to be done.
Throughout the course of my Internship, I saw it as the beginning of my career and expected to be afforded upward mo-
bility within the structures and corporate ladder of Nutrien. I saw myself having a realistic prospect of becoming perma-
nent, achieving promotions and opportunities for advancement within my Department. At this point, I tried my best to
get the work done as quick and efficient as possible. Your attitude and approach to tasks as an intern should align itself
with maximizing productivity and efficiency while understanding that the reward will often times be another opportunity
to prove yourself, to yourself and to your company. Maintain an open mind to assisting in different disciplines as this pro-
ject a helpful spirit onto your employers and it will be commended. So interns, get as involved as possible.
Page 9
Up to this point, you should be very busy. Soon enough, you’ll be given some unreasonable tasks to get done that you
may not think you’re ready for. For example, I was asked to present on a new program developed by the IT Department
that they would use to manage data from the Inspection Department. This program was very new and only 3 Interns
were trained to use it, including me. When I was asked to present initially, I assumed I had a day or two to prepare how-
ever I had a humbling 10 minutes. What was even more humbling was the audience I had to present to, which included
many managers and superintendents. I could have rejected it but where would that have gotten me?
As an intern, I really had nothing to lose so I weighed my options and decided to leave my comfort zone and proceed
with the presentation. The first five minutes did not go as horrible as I thought it would go but midway the System Ana-
lyst was able to come in and take over the presentation. She then left me to deal with the questions. Even though I
couldn’t present to everyone, I was able to have a conversation afterwards with a room full of professionals and their
concerns on the program. Interns, do not stay in your comfort zone. It may sound cliché but don’t let the fear of failure
hold you back. Weeks after, I was given my first engineering project alongside another intern and only since then I’ve
truly appreciated how crucial that very same library we sorted as interns, is to the staff of Nutrien.
I was fortunate to have good timing, great mentors, supervisors and bosses who were key in my development as an engi-
neer. Sometimes we may take for granted the people we work with and not realize that they may be one of the very
best in their area of study, so always show them the respect they deserve. If they are difficult to work with or even strict,
embrace it because it just means they have high standards and expectations. Also keep in mind to be very professional
and do not get involved in office politics at all. Create clear boundaries between your personal and professional life.
In conclusion, take your job seriously and have it done to the best of your ability, always remember the job has to be
done, no two ways about it. By doing your job, the professionals’ job becomes easier allowing them to get the more criti-
cal jobs done. Also, I advise to get involved as much as you possibly can, never shy away from stepping outside your com-
fort zone. You’ll only know you can’t do something until you do it and remember to stay professional. Most importantly
stay humble and respectful to the people you work with on every level. An internship does not always promise a job but
give it its best chance to be. It provides an opportunity for growth, learning and advancement in your career goals. Make
the most of it.
Brenton Chatoo is currently a Junior Mechani-cal Engineer stationed at Nutrien Trinidad with
growing experience in the Oil and Gas Industry. He provides specialized engineering support for the Reliability and Inspection of plant assets to
ultimately aid in production of Ammonia and Urea. Brenton aspires to become a well-accomplished Reliability Engineer and one of the
best in his field. He has special interests in Metal-lurgy and Plant Processes.
Page 10 APETT Engineering Magazine June 2019
Page 11 APETT Engineering Magazine June 2019
Design and Selection of Choke Valves
Part 1I
By: Avinash Babwah, Process Engineer
1.0 Results
Case 1
Table 1.0: Application data for datasheet 1 and simulat-
ed results
Figure 1.0: Cv vs Percentage Opening of Choke valve, Lin-
ear 1.5'' Trim
Page 12
Case 1I
Table 2.0 : Application data for datasheet 2 and simulated
result
Flow Characteristic curves of choke valves were generated
based from the model of NOV’s choke valve selected and
information from the data sheets. The NOV Varco MPC60S
1.50’’ Linear and MPC60S 2.5’’ equal percentage were both
well suited per datasheets 1 and 2 respectively with the life
of the well in mind. The required Cv falls well within the
designed sizing ranges of 40-75% of choke valves. NOV’s
MPC60S model nominal designation body size is 6’’ which
allows various trim sizes to be used ranging from: 1’’, 1.50’’,
2.00’’, 2.50’’, 3.00’’, 3.50’’, 4.50’’ and 5.00’’. The
MPC60S model was selected for its 6’’ nominal body size in
keeping with the specified 6’’ ANSI 2500 outlet flanges as
per both datasheets/applications. What required the use of
discretion, was the selection of trim size or orifice size. The
linear curve, figure 6 shows two cases, the minimum and
maximum Cv for the 1.5’’ trim. Using a trim/orifice size of
1.5’’, as the choke is initially opened to 20%, there is no
flow with a constant Cv of approximately 0.
Figure 2.0: Cv vs Percentage Opening of Choke valve,
Equal % 2.5'' Trim
This is considered a dead band due to the plug travel not
exposing the flow area until 12.5% and greater. Looking
further, from 20-89% there is a sharp but linear increase in
Cv as the choke is opened further to maximum of 50. This
is because the orifice pattern is similar throughout the en-
tire trim with travel. With the desired gas flow of 30mmscf/
day, associated condensate of 60 bbl/day and produced wa-
ter of 15bbl/day, requires an average Cv of 16.5 and
Page 13
42.0% opening of the MPC60S 1.5’’choke generating a
maximum exit velocity of 48.09ft/s and predicted noise
less than 85dBA. The exit velocity is well below the sonic
velocity or mach 1 and industrial standard of 85dBA which
decreases the risk of cavitation and erosion. ChokeSizer
does not show the predicted noise levels less than 85dBA
but do when greater than 85 dBA. This is a clear indica-
tion that vibration and cavitation would not occur within
this 1.5’’ choke trim and choke body.
For this application the choke is operating less than half its
capacity which gives the flexibility of opening the choke
further to provide a greater flow area when well pressures
have decreased below initial production. This would help
maintain production levels of at least 30MMSCF/day or
more if desired. Hence, the choke valve was sized using
the MPC60S 1.5’’ trim keeping in mind the life of the well.
Application two, only one case was provided with a gas
flow rate of 45.00mmscf/day, associated condensate of
90.00bbl/day and produced water of 22.50bbl/day with a
pressure drop of 225.00psig. The flow requirements for
this application is greater than application one by 15mm-
scf/day of gas, 7.5bbl/day of water and 30bbl/day of con-
densate. Thus, from simulations the best design to achieve
these flow is the MPC60S 2.5’’ equal percentage trim with
an exit velocity of 72.13ft/sec and predicted noise of less
than 85dBA (0 dBA as per results). This is a clear indica-
tion that vibration and cavitation would not occur.
The MPC60S 2.5’’ equal percentage trim design, a Cv of
49.87 and 48.06% travel is required to produce 45mmscf/
day and its associates. A maximum Cv of 150 is achievable
with approximately 96% of travel allowing the flexibility to
open the choke valve more later in the well life to main-
tain gas production. This is the ideal choke valve trim
bearing the well life in mind. Interpretation of the flow
with percentage travel (figure 6) shows a relatively linear
pattern except the lower and upper boundaries of the
flow curve. This is due to the flow geometry being smaller
at these regions of the trim (smaller slots), where the slots
size increased and remains uniform in the linear region.
The dead band at the bottom of curve is due to the plug
not exposing the flow area to fluid flow while the top dead
band occurs since the total flow area is already exposed.
This choke would produce negligible/no flow when
opened up to 11.9%. Opening the valve further above
21%, the Cv increases gradually up to 26% choke travel, as
it is opened further, and a larger flow area is exposed and
more fluid flows, there is then a faster increase between
26% and 90%. The fastest rate of production in terms of %
travel would be seen in the range of 30% and 90% where
the maximum Cv is attained and entire orifice area is ex-
posed to flow. The plug exposes all of the available holes
to flow at 96% open and further opening won’t affect any
increase in production.
For both application, the choke trims would be construct-
ed from tungsten carbide steel because of its erosive re-
sistant properties to cater for any solid or sand produc-
tion while the body and bonnet would be made of AISI
4130 low alloy steel. Typically the AISI 4130 low alloy
steel is cheaper than the tungsten carbide steel creating a
more economically affordable valve.
The choke would be automated with a stepping actuator,
National Oilwell Varco (NOV) VBS100. This would be
operated via instrument air signal ranging from 80-120psig
to the actuator via solenoid valves and controlled via the
control room. The stepper choke control would be used
such that when the step count is increased or decreased
the choke valve opens or closes respectively. Each pulse
signal (instrument air) to the internal piston, moves the
pawl in contact with the drive spool and produces a 36o
rotation of the stem coupling. This gives the operator ac-
curate controllability of the flow and prevention of instan-
taneous increases. The valve is fail-safe last position mean-
ing that if the solenoids fails to fire the choke would re-
main in the same position unless manually override. The
chokes can be override manually via an external hex nut
coupled to the drive spool and choke valve stem.
The travel of the choke would be monitored via a Top-
worx DXP position transmitter and be relayed to the con-
trol room for monitoring. It consists of a 5KΩ and a 4-
20mA transmitter for remote feedback of the choke posi-
tion. This allows the operator to determine the theoreti-
cal flow out of the choke compared to actual. If the flow
is more than the theoretical then it can be possible that
the choke plug is eroded and exposing greater area to the
flow area.
Page 14
Thus, the well needs to be shut in and choke valve removed
for inspection and repairs.
2.0 Conclusion
Chokes valves were designed and selected based on choke
valve datasheets utilizing the service conditions (Fluid flow
rates and properties, ΔP) and actuator parameters into Na-
tional Oilwell Varco Chokesizer! Program. Flow characteris-
tic curves (Cv) were developed for two cases or wells in
Trinidad’s East Coast where the Cv’s were less than half the
total percentage of choke opening. This provides the end
user with the ability to produce more gas without any design
modifications to the choke. This would save the end user
money and downtime long term in future modifications and
expansion.
The trims of both valves would be constructed from tung-
sten carbide steel due to its erosive resistant properties
whereas the valve body and trim would be made of AISI
4130 low alloy steel. Both cases were of a typical choke
valve, where the pressure drop was relatively small, noise
and vibration were within permissible limits. Cases where
the pressure drop i.e. differential pressures between the
inlet and outlet of the choke valve is greater than 1500psig
should be investigated for future work. Typically with high
pressure drops in choke valve, noise, vibration and cavitation
are usually associated and unwanted. This causes severe
valve body wall loss and pipe/material fatigue in downstream
piping, posing a serious process safety hazard. The industry
have moved to choke valve having multistage trims in an ef-
fort in reducing noise, vibration, and cavitation.
Multistage trims can be recommended to be used in lieu of
the existing single stage trims on both choke valves if cavita-
tion and vibration were to occur. Additionally, the trim wear
monitor port should be tied into an alarm system whether
audible or electronic to alert operators or tied to an emer-
gency shut down system to prevent damage to seat body.
Sand production should also be monitored and special atten-
tion and precautions should be paid to choke valves where
there are abnormal increases of sand. As with any piece of
equipment, routine maintenance programmes should be de-
veloped and executed to maintain equipment integrity and
functionality.
Figure 3.0: Choke Valve Datasheet 1
Page 15
Figure 4.0: Choke Valve Datasheet 1I
Avinash Babwah, a National Scholar, having placed 8th in Environmental Studies at the CAPE 2010 Examinations, and graduated in 2015 with a BSc. In
Chemical and Process Engineering from the University of West Indies.
Avinash is currently employed at Process Components Limited as a Process Engineer in the Projects Department, focusing on Technical Sales and Services,
Project Management and After Market Support. Throughout his tenure at ProCom, Avinash has worked with a number of Up-
stream Operators and has been involved in major Brownfield Projects with companies such as British Petroleum Trinidad & Tobago, Shell Trinidad and Tobago and EOG Resources. His exposure to the oil and gas industry and his
yearning for knowledge has led him to excel in his career and to further his studies with the pursuit of a MSc. In Petroleum Engineering.
Avinash has a passion for the game of cricket and is known for not missing a match. He appre-
ciates a great movie or TV-Show, and has a love for travelling, exploring beaches, hikes and snorkeling.
Page 16 APETT Engineering Magazine June 2019
Page 17
Abstract
As a consequence of its critical impact upon societies, the
occurrence of vehicular traffic accidents is a globally studied
phenomenon. Much effort has been directed towards the
understanding and identification of causal factors, with the
intention of minimising the occurrence. In a related area, the
development of methods for the identification and classifica-
tion of vehicles has also received necessary attention. How-
ever, little work has been done on the development of
methods for the identification of motor vehicle accident
occurrences. Thus, this work sought to develop an automat-
ed system for the identification of motor vehicular acci-
dents. It utilises an artificial neural network approach to
estimate the probability of occurrence, based on recorded
acoustic signals. More specifically, it first characterises acci-
dent acoustic signals by 9 selected signal features, in both
the time and frequency domains. It then develops a dual
layer artificial neural network, which accepts as its input the
9 characterising signal features and as its output calculates
the probability of occurrence. The system was built and
tested in the MATLAB environment, utilising 22 sample sig-
nals in the design phase and a further 53 for testing. An eval-
uation of the system found it have an accuracy of 86% and a
precision of 76%, with a 100% identification of actual acci-
dents. Additionally, it was found that the system prioritises
the time domain signal features over those of the frequency
domain, in the identification process. These results validate
the structure of the system used and demonstrate its poten-
tial for real-world applications.
1.0 Introduction
Road safety is a global concern. The World Health Organi-
sation reports that there were 1.25 million road traffic
deaths in 2013 alone (WHO, 2017). The impact of this phe-
nomenon is far reaching and many countries have been ag-
gressively seeking to counteract it. Accordingly, much effort
has been directed into the research of various aspects of
accident occurrences. Many researchers have investigated
the causal factors in the occurrence of accidents (de Ona et
al., 2013; Dadashova et al., 2016; Mujalli and de Ona, 2011).
The primary goal in most of these instances has been to
understand what causes accidents, with the intention of
minimising their occurrence. Similarly, other researchers
have sought to develop methods for identifying road con-
flicts (Cafiso et al., 2017) or for assessing the likelihood of
an accident occurrence in a particular location (Li et al.,
2017). Further, some investigators have developed methods
for reconstructing accidents, based on data gathered from
the scene of an accident (Li et al., 2017; Evtiukov et al.,
2017). Yet further, some researchers have developed meth-
ods for the determination of the level of injury of a vehicle's
occupants, upon the occurrence of an accident (Kononen et
al., 2011; Delen et al., 2006).
A related field of study of particular interest, is the detec-
tion and identification of motor vehicles. Several researchers
have used vibration and/or acoustic data, coupled with signal
processing techniques, to develop effective vehicle recogni-
tion and detection methods. Wu et al. conducted significant
work in this area and were among the first to utilise a fre-
quency spectrum principal component analysis approach for
vehicle sound recognition (Wu et al., 1999). George et al.
(2013 a) also used vehicle sound signals to detect and classi-
fy vehicle types in an Indian context. They developed an
algorithm that processed the acoustic data and allowed for
vehicle detection, then used a neural network for classifica-
tion. George et al. (2013 b) have advocated for the use of
wavelet analyses in their detection and classification tech-
niques. Yet in another case, Ozgunduz and Turkmen (2010)
designed a vehicular classification system using a Mel fre-
quency Cepstral coefficient algorithm and extracted features
of the acoustic data which was then reduced by using a vec-
tor quantisation algorithm.
Despite these efforts, little work has been done on the de-
velopment of methods for identifying the actual occurrence
of accidents. Currently, accident identification primarily re-
lies on visual recognition. In many cases, this is based on
reports by person(s) involved in the accident or by bystand-
ers. In others, the analysis of real time traffic camera data
allows for accident identification. However, this is limited by
several environmental factors such as the state of the vehi-
cle's occupants, the presence of bystanders and their willing-
ness
Automated Identification of Vehicular Accidents from
Acoustic Signals Using Artificial Neural Networks
By: Aaron D’Arbasie and Renique Murray
Page 18
to assist, lighting conditions and the level of monitoring of
traffic camera data. In-vehicle collision systems provide an
effective alternative. However, this too is limited by the
make and model of the vehicles involved and the level of
support system architecture in a particular location.
In light of this, this work presents an automated approach
for the identification of vehicular accidents. It utilises a com-
bination of an artificial neural network and some selected
signal processing techniques, to identify the occurrence of
an accident based on acoustic signal data. Such an approach
can be incorporated into existing traffic management sys-
tems or form the basis for a standalone system. In so doing,
it can facilitate faster response times to critical accidents
and increase the chances of saving an injured occupant's life.
2.0 System Design
2.1 General Approach
By virtue of the phenomenon’s nature, there are a number
of attributes that can be considered and examined in seeking
to detect the occurrence of an accident. Some of these in-
clude visual imagery, vibration data, scents/odors and
sounds. However, not all of these features are as easily
quantified and recorded, and the level and type of infor-
mation provided by each feature varies significantly. Not-
withstanding this, the work done on vehicle detection meth-
ods suggests that acoustic data samples provide a wealth of
information that can be used for accident identification, if
processed correctly. In keeping with this, this work sought
to use acoustic sample data as the primary data source for a
proposed identification system.
Figure 1 shows a typical acoustic sample recorded for an
accident. It can be seen that the accident is defined by a dis-
tinct rise in the amplitude of the acoustic signal and for a
short period of time. This pattern repeats itself for most of
the acoustic samples examined. Given the repetitive nature
of the pattern, the use of an artificial neural network was
considered to be a feasible approach for identifying its oc-
currence within a recorded signal.
2.2 Identification of Signal Features for Characterisation
The efficacy of neural networks in pattern matching and
identification, has been steadily increasing over the past few
years. Two key contributing factors have been the increas-
ing computational power of computing systems and the
growing access to more detailed data sets. However, de-
spite this increase in computational power, there are still
some evident limitations, i.e., the processing of large data
sets by a neural network does present a challenge for most
standard computers. For instance, the car accident acoustic
sample of Figure 1, which is 2.5 seconds long and sampled at
44.1 kHz, contains 110,250 data points. Attempts to directly
utilise this sample in an artificial neural network, have prov-
en to be memory-intensive for a current, standard desktop
computer. Accordingly, an alternative approach to utilising
the acoustic data in an artificial neural network had to be
developed.
Figure 1.0: Sample of accident acoustic data signal Ampli-
tude versus Time
As an alternative, it was considered that a signal can be rep-
resented in both the time and frequency domains. In keeping
with this, either presentation of the signal presents unique
aspects of the data. Accordingly, the signal can be character-
ised by the features of either representation, or a combina-
tion of both. Thus, the features in both the time domain and
the frequency domain are identified, such that these features
are influenced by the occurrence of an accident. Then, this
subset can be used to identify the presence of an accident.
Key to this proposition is that the features must vary specif-
ically with the occurrence of an accident.
Page 19
In so doing, they provide both a basis set for representing
the accident signal and for assessing the presence of an acci-
dent within a wider signal. However, it is unclear which of
the many signal characteristics in the time or frequency do-
main would be critical in assessing the occurrence of an ac-
cident. In light of this, a number of well-known signal fea-
tures and characteristics were examined, to determine their
level of influence in accident identification from acoustic
sample data. Table 1 gives the list of the features assessed in
this work.
3.0 Data Sets and Data Acquisition
For the purposes of training and evaluation of the system,
acoustic samples of various accidents were required. How-
ever, due to limited funding availability, the recreation and/
or simulation of real time vehicular accidents were not feasi-
ble in this work. Alternatively, existing accident data sets
were used. These comprised of data obtained from various
crash intuitions namely Insurance Institute for Highway Safe-
ty (IIHS) and European New Car Assessment Programme
(Euro Ncap). Both institutions conduct crash testing on a
wide range of vehicles and various types of collisions (e.g.,
head-on, and small overlap). The sampling frequency for
audio capture used in these data sets, was given as 44100
Hz for both institutions. The distance from the microphone
to the point of impact was not given; however, it was
known to vary for both. The acoustic greater accuracy of
representation. The vehicle type and accident details for the
various samples examined, are presented in Table 2
As opposed to one type, various types of collisions were
used to ensure variability in the accident features examined.
The aim of this approach was to increase the system's likeli-
hood of identifying a random accident. A total of 45 vehicu-
lar accident samples was used in the development of the
system.
Additionally, simulated accident data was obtained from a
test rig that was setup for the purposes of the work. The
test rig consisted of a weighted automobile front bumper,
suspended in mid-air by a pulley system. The bumper was
lifted to a height of 12 feet and then allowed to fall and
strike a metal sheet, which was fitted with an accelerome-
ter.
Table 1.0: Acoustic signal features examined to determine
effectiveness in accident identification
Table 2.0: Types of vehicles for which data was acquired
Page 20
A microphone was positioned 10 feet away from the drop
site, to record the acoustic data. The data was recorded at a
sampling rate of 44,100 Hz. A picture of the setup is shown
in Figure 2. Some of the amples recorded here were used in
the identification of the set of key signal features. Addition-
ally, acoustic samples taken of a jackhammer in operation
and of random noises were also recorded for use in as-
sessing key signal features.
Figure 2.0: Image of simulated accident testing setup
4.0 Results and Discussion
4.1 Identification of Key Signal Parameters
The signal features presented in Table 1 were assessed for
all of the test signals previously mentioned. Various plots
were made to examine the performance of each character-
istic. The results obtained here were used to determine
which characteristics were most suitable for classification of
an accident. Figure 3 shows the plot of normalised mean
signal amplitude against signal variance.
Figure 3.0: Pot of mean amplitude vs. variance
It can be observed from the figure that an accident is easily
characterised by the variance of the amplitude time plot.
The variance of the accident signals is found to be lower and
exhibits less variability than the other signals examined. An
examination of the normalised, mean amplitude shows that
for an accident signal, the values are much lower than the
other signals considered. This is due to the fact that acci-
dent signals contain localised points of very high amplitude,
with the remaining portion of the signal having significantly
lower values. On the contrary, noise signals generally do not
have notable localised peaks and consequently their normal-
ised means are higher. Accordingly, both features are suita-
ble for characterisation.
Figure 4 illustrates the changes in the fundamental frequency
and the zero-crossing rate of the signal. From the figure, it is
evident that the values of the zerocrossing rate are much
higher for both sets of accident signals, as compared to oth-
er signals considered. Accordingly, this is a suitable signal
feature for characterisation. Conversely, the fundamental
frequency demonstrates a high degree of variation and does
not show any specific relationship for the signals considered.
In keeping with this, the fundamental frequency serves as a
poor characteristic and its use would lower the efficacy of
the system.
Page 21
Figure 4.0: Plot of zero crossing rate vs. Fundamental fre-
quency
An examination of the bandwidth values in Figure 5, shows
that it is difficult to differentiate an accident signal from
those of the other signals considered. Accident signals have
wider bandwidth ranges than the other signals, making char-
acterisation difficult.
Conversely, accident signals can clearly be distinguished by
the spectral crest values. The spectral crest values for both
sets of accident signals are visibly lower than the other sig-
nals considered. Accordingly, the spectral crest was selected
as a feature for characterisation, while bandwidth was not.
An examination of Figure 6 shows no clear relationship or
correlation between the occurrence of an accident and the
maximum energy or the energy flux. These two signal fea-
tures are dispersed through a large area, and hence at-
tempts to use them for accident characterisation may intro-
duce some error into the system.
Consequently, both parameters were not included in the
final subset used to develop the system.
Figure 5.0: Plot of bandwidth vs spectral crest
Figure 6.0: Plot of energy flux vs. Maximum energy
Figure 7 displays the frequency envelopes of the accident
and noise signals tested. The signals have been converted
into the frequency domain using a fast Fourier transform.
An analysis of the graph shows a distinct difference between
the noise signal (blue) and the accident data (black). It can
be seen that the frequencies present within the accident
signals are more stochastic as compared to the noise signals.
Additionally, the amplitudes of the frequencies that are pre-
sent in the accident signals are larger than those of the
noise. Accordingly, the frequency envelope was chosen as a
feature for accident characterisation.
4.2 Network Development and Architecture
Based on the previous analysis, 9 signal features were identi-
fied for the characterisation of accident signals.
Page 22
The features include both time domain and frequency do-
main identifiers. The time domain features selected were
the energy flux, mean amplitude, power, zero crossing rate
and variance; whereas the frequency domain features in-
clude frequency envelope, bandwidth,
Figure 7.0: Frequency envelope of various acoustic events
spectral crest and variance. In so doing, this allows for the
reduction of an accident signal having 110,250 points to 9
characteristics. Figure 8 shows the sequence of computa-
tional steps within the final system.
The development and subsequent analysis of the network’s
performance was done using MATLAB 2015. This process
entailed two primary decisions: a determination of the num-
ber of layers in the network and a determination of the
number of neurons required for accurate functionality. The
previous analysis indicated that the characteristics of an acci-
dent signal are not linearly separable. In keeping with this, a
multilayer approach was considered to be more suitable.
More specifically, a dual layer configuration was implement-
ed, with a hidden layer containing a linear function and an
output layer.
Figure 9 shows the final architecture of the neural network.
The batch training method was selected as the basis for
training the network, using a sample set of 22 signals. This
was implemented with randomly determined batches, using
a gradient descent algorithm via the MATLAB interface. This
Figure 8.0: Final system architecture
approach minimises the loss function as a means of adjust-
ing function weights and improving the network perfor-
mance. MATLAB subsequently validates the network with a
subset of samples.
Figure 9.0: Final architecture of the neural network
The determination of the most suitable number of neurons
was effected via the pruning approach. The proposed se-
quence of computational steps in Figure 8 was implemented
using a test network having 11 neurons in the hidden layer.
This test network was trained and validated as previously
discussed. Subsequently, its performance was assessed via
the examination of key network characteristics. More spe-
cifically, the root mean squared error (R2 value) relative to a
set target value was calculated for the test network, which
was indicative of its ability to observe trends.
Page 23
Nine other test networks were subsequently developed,
using a different number of neurons in the hidden layer,
ranging from 10 to 2 neurons. Each test network was
trained, validated and assessed in a manner that was identi-
cal to that of the 11-neuron network. Three of the test net-
works were found to have R2 values of 0.999, indicating the
ability to accurately differentiate between a car accident
signal and the other test samples.
Using Ockham’s razor principle, four neurons were selected
as the most suitable number of neurons to be used in the
network. Accordingly, the final system architecture consist-
ed of 2 layers with four neurons in the hidden layer. This
system is such that 109 points are inputted based on the 9
characterisation features and a probability value is output-
ted.
4.3 System Performance
The system was tested using a number of new data sets, i.e.,
signals that had not previously been used in the develop-
ment and training of the system. These data sets consisted
of 16 car accidents signals obtained from the Insurance Insti-
tute of Highway Safety (IIHS), 7 simulated accident signals,
12 noise signals, 9 sample signals of impact strikes on differ-
ent materials, and 9 other sample signals of noises likely to
be recorded on the roadway (e.g., emergency sirens and
jackhammering). Of the 53 tests on the system conducted,
Figure 10 presents the results of 36 outputs of the network.
Table 3 presents a confusion matrix for the predictions
made by the system.
Figure 10.0: Results of Classifications System Output for
Acoustic Data
Table 3.0: Confusion matrix for predictions made by sys-
tem
In keeping with Table 3, the following performance criteria
can be evaluated: Accuracy = (true positive + true negative)/
total = 86%, True positive rate = True positive/ Actual posi-
tive = 100%, False positive rate = False positive/ Actual no =
30.4%, Precision = True positive/ predicted yes (when it
predicts yes, how often is it correct) = 76%.
4.4 System Behaviour
In examining the system, some key relationships and behav-
ioural trends were identified. One of these concerns the
issue of the incorrect classification of the impact strike sig-
nals. It was noticed that impact strike signals where a high
force was used, had a higher chance of being classified as an
accident. This false positive classification occurred both with
strikes to steel and polyethylene materials.
Although the natural frequencies of both steel and polyeth-
ylene of similar masses contrast greatly, both were still clas-
sified as a car accident. This suggests that the neural net-
work gives precedence to characteristics in the time do-
main, as opposed to those in the frequency domain. This is
likely a consequence of the fact that the features in the time
domain display a greater correlation with the occurrence of
a car accident, than those in the frequency domain.
A second key behavioural trend concerns the nature of the
probability values obtained. The outputs for the tests con-
ducted showed a range of values between 0.7 -1.0, to pre-
dict the occurrence of an accident.
Page 24
Conversely, probabilities of 0 - 0.21 were found in cases
where the system suggested that an accident did not occur.
These ranges of probability values allowed for clear inter-
pretations to be made on whether or not an accident did
occur. This result was a consequence of the sigmoid func-
tion in the hidden layer. Its insertion reduces the probability
of having instances where the neural network predicts a
50% chance of the occurrence of a car accident. These re-
sults serve to validate the structure of the system used.
5.0 Conclusion
This paper presented the work done on the design of an
automated system for identifying vehicular accidents, using
acoustic signal data and utilising an artificial neural network
approach. The system was based upon the identification of
key signal features that were used to characterise an acci-
dent acoustic signal. A total of nine signal features was iden-
tified with five being time domain features and four of the
frequency domain.
These features allowed for large data signals to be repre-
sented by a much smaller data set; in so doing significantly
decreasing the computing requirements of the system. The
system was designed and tested using MATLAB. In designing
and training the system, 22 signals were used. These signals
consisted of actual accident recordings, simulated accident
data and other recorded acoustic data. The system was sub-
sequently tested using 53 additional signals that were not
used in the design phase.
An evaluation of the system's performance found that it had
an accuracy of 86% and a precision of 76%, with a 100%
identification of actual accidents. Testing also served to
identify that the system prioritises the time domain signal
features, due to a greater correlation between changes in
these values and the occurrence of an accident.
With correct incorporation into a wider traffic management
and/or emergency system, the approach presented here has
the potential to significantly increase the likelihood of identi-
fying vehicular accidents. In so doing, it can increase the
response time of emergency personnel and increase the
potential for saving lives.
Renique Murray is a
trained Mechanical Engi-neer and has been lectur-ing, managing projects and conducting postgrad-
uate research in various aspects of the field over the past fourteen years.
He holds both a BSc. and an MPhil degree in Me-chanical Engineering, with
the latter focusing on vibration analysis of pow-
er generation rotating machines. He also holds a PhD. in
Process and Utilities Engineering, in the area of renewable fuel technology and the techno-economics of power gen-eration. His core area of focus is in the field of renewable
energy and power generation. However, he also conducts research in the areas of vibration-based signal processing
and machine design for agricultural purpose.
Aaron D’ Arbasie is
a graduate student in the Master of Engi-neering programme at
Carleton University, Ottawa, Ontario, Can-ada. He graduated with
First class honours degree in mechanical engineering, from The
University of the West Indies. Aaron has a keen interest in solving problems using engi-
neering approaches.
Page 25
1.0 Introduction
Working in the plastics packaging industry has introduced
me to the reality of how rapidly we consume plastic prod-
ucts. Of course, I’m aware that plastics are essential materi-
als in all areas of society, but their extensive use, slow deg-
radation and petroleum-based origins contribute to pollu-
tion and depletion of non-renewable resources. A 2010
waste characterization study commissioned by the Ministry
of Local Government found that plastics comprised of an
average of 19.525% of waste in Trinidad and Tobago’s four
primary landfills (Government of the Republic of Trinidad
and Tobago, 2015). This figure does not consider the plastic
waste that is currently littered throughout the country, find-
ing residence at roadsides, and in drains, rivers, and our
coastline. Moreover, plastic waste takes, on average, 450
years to decompose in the environment (National Oceanic
and Atmospheric Administration Marine Debris Program,
2018).
In recent years, sustainability has become an important con-
cern for governments, businesses and everyday citizens, as
the negative impact of human activities needs to be ad-
dressed. In 2015, the United Nations (UN) launched its Sus-
tainable Development Goals (SDG) mandate which focuses
on 17 key areas of global sustainability. But what does the
term sustainability mean? In the literature, sustainability is
commonly defined as “development that meets the needs of
the present without compromising the ability of future gen-
erations to meet their own needs” (Shah, 2008).
It encompasses three main pillars, namely environmental,
social and economic development as illustrated in Figure 1.
This short article considers the concept of environmental
sustainability in terms of environmental pollution and how
material innovation can potentially make plastics less harm-
ful to the environment and thus more sustainable.
Figure 1.0: The three interconnected pillars of sustainabil-
ity
According to Goodland (1995), the concept of environmen-
tal sustainability refers to sustaining global life-support sys-
tems indeterminately. Pollution is the biggest barrier to
achieving environmental sustainability, as the improper dis-
posal of waste materials harms ecosystems around the
world. With respect to plastic pollution, it is estimated that
about 79,000 tonnes of plastic are dumped in the Pacific
Ocean alone, continuously increasing the size of the now
infamous ‘great Pacific garbage patch’, that is currently three
times the size of France (Arora et al., 2018) as shown in
Figure 2. The innovation of products and processes are
needed to combat pollution and move closer towards creat-
ing a sustainable environment.
Bioplastics and Environmental
Sustainability: Some Thoughts
By: Trishel Gokool, BSc (UWI), MSc (Manchester), AMIMechE
Page 26
Figure 2.0: Thermographic satellite image of the Great
Pacific Garbage Patch (Canelo, 2018)
Conventional plastics are petroleum-based and are typically
not biodegradable, both of which are undesirable character-
istics for environmentally sustainable materials. To this end,
new materials are being produced to reduce the environ-
mental impact of plastic products, namely, bio-based plastics
and biodegradable plastics, both of which are generally
termed ‘bioplastics’. Bio-based plastics, like the name sug-
gests, are derived from renewable resources such as corn,
soybean, bioethanol and lignin. Ideally, their production does
not rely on our ever-depleting petroleum resources, so,
producing these bio-based plastics will not only reduce cur-
rent petroleum usage, but we will still be able to produce
them when these reserves are exhausted. Biodegradable
plastics, on the other hand, are plastics that undergo physi-
cal and chemical deterioration and completely degrade into
carbon dioxide or methane, and water by microorganisms.
This action will considerably reduce the amount of time the
plastic remains in the environment. Bio-based plastics can be
biodegradable or nondegradable, and can also be molecular-
ly similar to existing plastics, such as bio-based PET
(polyethylene terephthalate), or completely new materials
such as PLA (polylactic acid).
However, although many bio-based polymers are biode-
gradable, not all biodegradable plastics are bio-based (Babu
et al., 2013). Some biodegradable plastics can also be de-
scribed as compostable and undergo degradation through
Figure 3.0: Degradation of a compostable bioplastic bottle
over the span of 80 days (Echo Instruments, 2016)
biological processes in industrial and home composts. Figure
3 shows the degradation of a compostable bottle over the
span of just 80 days. As observed, after 80 days it has de-
graded enough for the fragments to be invisible to the naked
eye. However, it must be verified that complete biodegrada-
tion has in fact occurred and not simply fragmentation, as
these fragments can remain in the environment for a long
of period of time and be just as damaging to the environ-
ment.
A major drawback of several of the new bioplastics is that
they cannot be used in current processing equipment. To
combat this issue, and maintain material performance, many
companies are offering additives that can be added in small
quantities to current feedstock, making the resulting plastic
product degradable. These companies use terms such as
“oxy-degradable”, “oxy-biodegradable” and “degradable” to
describe the products manufactured from the additive and
conventional plastic combination.
Page 27
There has been much controversy surrounding these claims,
as a few emerging studies have shown that the additives do
not reduce the degradation rate of plastics (Selke et al.,
2015; Lambert and Wagner, 2017), whilst the additive man-
ufacturers are providing data that show otherwise. Standard
testing bodies like ISO and ASTM have been trying to devel-
op standard testing methods for degradability of plastics, but
with the absence of such standards, plastics manufacturers
and consumers must decide for themselves if the claims are
justified and the risk is acceptable.
At present, the testing done on these additives often refer
to ASTM D5338 and D5511, but although these are stand-
ard test methods, they are not standard specifications
(PLASTICS, 2018). Third party or in-house testing is a good
idea before introducing such additives to conventional res-
ins, as well as toxicity testing for plastics used in food pack-
aging and mechanical testing for high strength applications.
In advertising these blends as biodegradable, companies
must also ensure that they do not inadvertently promote
littering, as consumers may get the impression that these
products will harmlessly disappear in the environment.
Proper end of useful life recovery and disposal is required to
successfully dispose of these plastics, and care should be
taken to avoid the contamination of the waste stream by the
mixing of biodegradable plastics with nondegradable ones.
In order to satisfy the UN’s SDG, which includes ensuring
sustainable consumption and production patterns and build-
ing resilient infrastructure, promoting sustainable industriali-
zation and fostering innovation; Trinidad and Tobago must
step up and reduce environmental pollution.
Local plastic manufacturers should be looking towards sus-
tainable materials and processes to reduce future plastic
waste and engage in recycling drives to reduce the plastic
waste currently littered throughout the country. Moreover,
with the reduced supply of petroleum, bio-based plastics are
becoming increasingly attractive.
However, due to the absence of established standards, com-
panies and consumers alike have to decide on whether the
bioplastics they produce and/or use are in fact beneficial to
the environment, and both need to be proactive in the end
of life disposal and recovery of these plastic products.
Trishel Gokool is a part-time graduate student
at The University of the West In-dies, pursuing her PhD Manufactur-
ing Engineering. She is also em-ployed at ANSA
Polymer as a Management Trainee under the
Champions De-velopment Pro-gramme. In 2015,
she graduated with a BSc Mechanical Engineering (First Class Honours) from The University of the West Indies.
In 2018 graduated with an MSc Advanced Manufac-turing Technology and Systems Management (Distinction) from the University of Manchester,
UK. Her research interests include but are not limited to computer-aided design and manufactur-ing (CAD/CAM), additive manufacturing, process
optimization and engineering education.
Page 28 APETT Engineering Magazine June 2019
Page 29 APETT Engineering Magazine June 2019
Assessing The Sensitivity Of The East
Coast of Trinidad To Oil Spills
By: Nyoka Sinanan
1.0 Introduction
The east coast of Trinidad is host to both oil and gas pro-
duction and transportation activities. There are over thirty
gas fields and ten oil fields owned by both local and foreign
companies located off the east coast of Trinidad. Because
of the proximity of these fields to the coastline, the chances
of an oil spill occurring is very high and due to the wind and
current direction, spilled oil is most likely to impact the
coastline. The arrival of oil on beaches which are used for
recreation, sports and other amenities can adversely impact
on tourism. The biological effects on shoreline are also of
great concern especially in environmentally sensitive areas
such as the Mathura Turtle Nesting Ground and the Nariva
Swamp and also in areas where mangroves persist and are
in direct contact with the open sea.
Castanedo et al. (2009) stated that to respond quickly and
successfully to an oil spill in a defined geographic area, a
contingency plan, including information and processes for
oil spill containment and clean-up is required. A study un-
dertaken by Wieczorek, Dias-Brito, and Milanelli (2007),
aimed at developing an ESI, which followed procedures
determined by the International Convention of Oil Spill
Prevention, Preparedness and Response (OSPPR). These
procedures included separating the coastal habitats into
different littoral sensitivity indexes to oil spills. Both re-
searchers chose to classify the areas by looking at the bio-
logical, physical and socio-economic factors that would be
impacted. However, Castanedo et al. (2009) used a quanti-
tative approach when ranking the coast by using formulae
to decipher the vulnerability index of the area. Wieczorek,
Dias-Brito, and Milanelli (2007) relied on more of a qualita-
tive approach as he ranked the coastal habitats in 1- 10
vulnerability indexes such as exposed rocky headlands,
eroded wave cut platform, etc. Another research paper
done by Adler and Inbar (2007) classified the Mediterrane-
an Coastline of Israel into 13 ESI categories of sensitivity to
oil spills in a similar manner to Wieczorek, Dias-Brito, and
Milanelli (2007) . The classification is based on the concept
that shoreline sensitivity depends on the level of geological
and geomorphological characteristics, exposure to the
ocean, sediment size, biological resources, socio economic
patterns, current and ocean tides and wind speed and di-
rection.
The eastern coastline of Trinidad is classified using a quali-
tative approach. The classification is undertaken according
to the NOAA’s Guidelines as specified by the National Oil
Spill Contingency Plan for Trinidad and Tobago (Ministry of
Energy and Energy Affairs 2013). NOAA’s method for de-
veloping an Environmental Sensitivity Index (ESI) map, cate-
gorizes coastal habitats in terms of their sensitivity to
spilled oil by analyzing the physical, biological and social
factors along the coastline under study. It involves perform-
ing a coastal classification coupled with data on relative
exposure to wave and tidal energy, shoreline slope and
substrate type to rank the shoreline in terms of its suscep-
tibility to oil spills. The shoreline ranking combines with the
delineation of both biological and human use resources to
produce the ESI map.
Wave and tidal energy determine the degree of exposure
of the coastline. Wave heights exceeding 1m indicates that
the impact of oil spills on the coastline is reduced as off-
shore-directed currents generated by waves reflation pat-
terns push the oil away from the shore. Tidal-energy flux
must also be taken into consideration as there is the poten-
tial for strong tidal currents to remove stranded oil as tidal
currents generally increase as tidal range increases. High
energy means rapid natural removal, usually within days to
weeks. Low energy means slow, natural removal, usually
within years. Medium energy means that stranded oil will
be removed when the next high-energy event occurs,
which could be days or months after the spill (NOAA
2002).
Page 30
The substrate types can be classified as bedrock, sediments
and man-made material. Table 1 shows the grain size for the
different types of sediments.
Table 1.0: Grain Size for different sediment types
(NOAA 2002), states that penetration occurs when oil
stranded on the surface sinks into permeable sediments. The
depth of penetration is controlled by the grain size of the
substrate, as well as the sorting. Penetration is more preva-
lent for coarse, well sorted sediments and as such oil can
penetrate up to 1m on gravel beaches. Trafficability is also
dependent on the substrate type. Highly trafficable shorelines
are ranked lower on the ESI scale than those on which clean-
up crews will have difficulty moving on or where their clean-
up efforts will cause additional damage. Fine-grained sand
beaches are typically compacted and hard with little chance of
workers trampling oil deep into the substrate. Coarse-
grained beaches tend to have moderate to steep slopes, are
much less compacted, and have a high permeability, making
walking difficult and more likely to drive any stranded oil
deeper into the substrate. Wetlands have very low trafficabil-
ity due to the innate softness of the substrate which makes
clean-up difficult (NOAA 2002).
The biological productivity of shoreline habitats is also im-
portant. Mangroves have the highest ranking because of the
potential for long-term impacts resulting from both exposure
to oil and
potential damages associated with clean-up activities in these
kinds of habitats. However, the presence of other sensitive
resources on a specific shoreline segment, such as turtle
nesting on a fine-grained sand beach, does not affect the ESI
ranking. These phenomena
are addressed by mapping biological and human-use re-
sources (NOAA 2002).
The map is produced using geographic information system
(GIS) techniques. The shoreline resources are ranked and
color-coded based on their sensitivity to oiling. The data on
biological and human resources also use standardized symbols
(NOAA 2002). Pincinato, Riedel, and Milanelli (2009) mod-
elled an expert GIS system based on knowledge to evaluate
oil spill environmental sensitivity. They stated that GIS stands
as a powerful resource to fulfil the limitations of traditional
environmental sensitivity maps. They visually mapped the
habitats based on a series of aerial photographs with a spatial
resolution of 0.98m per pixel. Wieczorek, Dias-Brito, and
Milanelli (2007) also used GIS mapping in their study. They
mapped the habitats under study from an orthophoto, which
enabled resolutions up to 1:2500.
Figure 1.0: Coastal Classification adapted from the IMA for
the north-east coast of Trinidad
Page 31
Figure 2.0: Coastal Classification adapted from the IMA for
the south-east coast of Trinidad
Figures 1 and 2 show coastal classification maps that
were adapted and updated from the IMA (1983). To-
gether with the wave, tidal, slope and grain size data
(table 2), the coastal classification is compared to the
guidelines given by NOAA (table 3) and the areas are
ranked in terms of their sensitivity to oil spills (figure 3
and 4). It should be noted that table 3 is a modified
version of NOAA’s guidelines as it has been made to
fit the coastal environment of Trinidad.
Table 2: Data used to apply IMA’s modified version of
NOAA’s shoreline sensitivity ranking to the East
Coast of Trinidad (The United Kingdom Hydrographic
Office Admiraility EasyTide 2017) (Institute of Marine
Affairs 2013).
From table 2, eight out of the nine beaches along the
coast are classified as fine-grained sand beaches. A
matrix was applied as seen in table 4 and areas that
consisted of fine-grained sand, low hydrodynamic lev-
els and flat to moderate slopes were ranked as ESI 4
as oil would not be readily removed from these areas
as compared to areas that have a higher hydrodynamic
level and steeper slopes.
Table 2.0: Data used apply IMA’s modified version of NOAA’s shoreline sensitivity ranking to the East Coast of
Trinidad (The United Kingdom Hydrographic Office Admiraility EasyTide 2017) (Institute of Marine Affairs 2013)
SITE WAVE HEIGHT (CM) TIDAL ENERGY (M) SHORELINE SLOPE
GRAIN SIZE
(mm)
Salybia Bay 23-30 – 1.2 6.5˚ 0.35- 0.5
Cumana Bay - 0.1 - 1.3 7.2˚ 0.5- 1.4
Sena Bay 180 0.1 - 1.3 9˚- 13.1˚ 0.5
Balandra Bay 40 0.1 - 1.3 4.3˚ 0.18- 0.25
Saline Bay 48 0.1 - 1.3 6.1˚- 10.1˚ 0.18- 0.5
Matura Bay - 0.1 - 1.3 6.5˚- 13˚ 0.25-0.5
Manzanilla Bay 60-78 0.2 – 1.4 4˚- 5˚ 0.125- 0.25
Cocos Bay 60-78 0.2 – 1.4 5˚-7˚ 0.125-0.25
Mayaro Bay 58-71 0.1 – 1.4 7˚ 0.125- 0.25
Page 32
The six shoreline habitat rankings found along the east coast
of Trinidad are ESI 1- Exposed vertical rocky shores; ex-
posed seawalls, ESI 3- Fine-grained sand beaches, ESI 4 –
Coarse-grained sand beaches, ESI 6- Gravel beaches/Riprap,
ESI 7 -Sheltered rocky shores/ Seawalls/ Vegetated banks;
Solid man-made structures and ESI 10- Mangroves. An ESI
ranking of 1 corresponds to exposed vertical rocky shores
or exposed seawalls. It consists of elements such as the
coast being exposed to high wave energy which tends to
keep oil offshore as the impermeable substrate allows oil to
remain on the surface where natural processes will allow
for removal (NOAA 2002). This means that no clean- up is
required. Approximately 26.84km of the eastern coast is
ranked as ESI 1.
ESI 3 comprises of fine-grained sand beaches with semi-
permeable substrate where oil penetration is limited and
beach slopes are very low. Grain size analysis done on
Table 3.0. Modified classification for Trini-
dad and Tobago (IMA 1996).
Table 4.0. Matrix used to determine sensitivity
ranking for areas consisting of fine-grained sand
Figure 3.0. ESI map for the north-east coast of Trinidad
Figure 4.0. ESI map for the south-east coast of Trinidad
Page 33
samples collected found the modal grain size to be between
0.06mm to 1mm which is classified as fine-grained sand by
NOAA (2002). 65.6km of the east coast of Trinidad is
ranked as ESI 3. A ranking of ESI 4 relates to permeable
substrate where oil penetration occurs up to 25 cm deep,
the slope is intermediate and the sediments are soft with
low trafficability (NOAA 2002).Grain size analysis done de-
duced the grain size to be between 1-2mm which indicates
it is of a coarse nature. Coarse-grained sand beaches are
ranked higher than fine-grained sand beaches as it is easier
for oil to penetrate which makes clean-up difficult. 9.7km of
the coast is ESI 4.
Cumana is the only area consisting of coarse- grained sand
but due to the selection matrix (Table 4) Salybia, Balandra
and Saline are also ranked as ESI 4 as the other sensitivity
ranking factors increase their susceptibility to spilled oil.
Manzanilla, Mayaro and Cocos are fine- grained and are cat-
egorised in the orange region of the matrix. They however
are ranked lower as ESI 3 because the hydrodynamic levels
in these areas are higher which would aid in the removal of
oil that makes landfall along these shorelines.
Due to erosion of the road along the Manzanilla/ Cocas
stretch, a boulder revetment was constructed just North of
the Ortoire river. This 0.7km of the coast was given a sensi-
tivity ranking of ESI 6. The riprap is in direct contact with
the water which makes it easy for oil to reach the boulders.
Flushing can be effective for removing oil but large amounts
of residue may remain depending on the type of oil (NOAA
2002).
Recent diversification of Manzanilla beach has also seen the
construction of a seawall along the foreshore area. This
0.2km of the coast receives an ESI ranking of 7. Seawalls are
man-made and are prone to oil spills as clean-up is difficult
for aesthetic reasons. Seawalls may also have a large number
of attached organisms which supports an ecosystem that
can be affected by a potential oil spill (NOAA 2002).
An ESI rank of 10 was designated to one of the most sensi-
tive parts of the east coast of Trinidad. All of the mangrove
systems along the east coast of Trinidad are not in direct
contact with the open sea but are located behind, and are
thus protected by beaches.
However, where river mouths connect to the ocean, a
threat is introduced in terms of the oil reaching the man-
groves via the river routes. In Figure 3 and 4, the areas
ranked ESI 10 are very minute because of how small the
areas at the mouth of the rivers are. Collectively 0.5km of
the east coast is ranked as ESI 10. Mangroves, marshes and
other vegetated wetlands are the most sensitive habitats
because of their high biological use and value, their difficulty
to clean-up, and the potential for long-term impacts to
many organisms (NOAA 2002).
Observations from previous spill events have shown that
mangroves undertake tremendous amounts of degradation
when oil comes in contact with it and that these ecosystems
are difficult to protect and clean up because of their intri-
cate nature (NOAA 2014b). The oil covers the pneumato-
phores of the mangrove trees which reduces its oxygen
supply.
Toxicity is another factor of concern as oil containing low
molecular weight aromatic compounds can damage the cell
membranes in the roots of the mangrove which results in
an increase in the intake of salt from the water which slowly
kills the trees. The fauna present in mangroves are also at
risk as oil may penetrate burrows in the sediments which
kills microorganisms, crustaceans etc. that are present
there.
Along with mangroves, there are biological species along the
coast that would be at risk because they are endangered
and highly valued due to their rarity. These Species include
the Leatherback sea turtle, Donax Clams and the West
Indian Manatee. Leatherback sea turtles use areas such as
Matura, Fishing pond, Manzanilla, Mayaro and Salybia as
nesting grounds. Matura (ESI 3), is a major turtle nesting
sites but based on shoreline sensitivity, it is ranked low on
the ESI map however, because of the presence of Leather-
back turtles, the areas become high priority and protecting
them are of utmost importance.
Page 34
An oil-contaminated environment can be lethal to both sea
turtles and incubating eggs. Since a sea turtle exists in differ-
ent habitats throughout its life cycle, an oil spill can affect it
at every stage in its life. The ESI maps (fig. 3 and 4) also
shows turtle breeding areas offshore which are indicated by
the large circles. Satellite telemetry show that sea turtles
spend much of their time directly off the nesting beaches
and up to 30 km offshore (Eckhart 2010). It would there-
fore be important that oil be prevented from reaching these
areas to ensure the protection of the breeding turtles off-
shore. Protecting the sea turtles that visit the island holds
environmental as well as economic importance.
Another biological resource that is indigenous to the east
coast of Trinidad is the Donax clams, commonly known
locally as “chip-chip.” These tiny shell type creatures are
prolific filter feeder that provides an important link in
coastal food chains, including sea birds and people. Because
they are filter feeders, Donax clams tend to absorb pollu-
tants through direct contact with contaminated water and
suspended particles (Snyder et al. 2014) which makes it pos-
sible for the transfer of oil components to other species in
the food chain.
An oil spill not only damages the biological factors along the
coast but also aesthetically degrades the environment as
well as causes pungent aromas that can cause health prob-
lems. Along the East Coast of Trinidad, there are two main
human use resources, fish landing sites and beaches. Beach-
es span the length of the coast and there are many popular
recreational bathing areas commonly visited by tourists and
locals. Fish landing sites are small infrastructures present in
areas where fishing takes place i.e. where fish is caught and
sold locally and on a small scale. When exposed to oil, adult
hain. Indefinitely, this means that consumers would desist
from buying produce caught in areas where oil spill have
occurred.
All in all, the amount of oil which reaches a shoreline is de-
pendent on the volume of oil spilled, the degree of weather-
ing, the nature of the coastal and marine environment and
the point of release of the oil. Depending on the type of
shoreline, the effects of the oil varies. If an oil spill does
reach the shoreline, the methods put in place to remove the
oil slicks should cause no further harm to the environments.
Dispersants such as Corexit 9500A and 9527A which were
used in the Deepwater Horizon Oil spill which occurred in
the Gulf of Mexico has been said to have negative impacts
on marine life. Dispersants also pose a significant threat to
human health as well as it contains 2-Butoxyethanol which
may affect the liver, kidney and red blood cells (Centre for
Biological Diversity 2017).
It is therefore necessary that authorities create clean-up
strategies prior to the occurrence of an actual oil spill that
would not depreciate the coast more than the effects of the
oncoming oil. The ESI maps produced from this research,
Figures 3 and 4, shows the sensitive areas along the coast
and would allow officials to conceptualize methods to effi-
ciently and effectively protect what is at risk.
Nyoka Sinanan has completed her BSc. in
Civil with Environmental Engineering from the Uni-versity of the West Indies
and is also pursuing her MSc in Coastal Engineer-ing and Management in
UWI St Augustine. Cur-rently she works as a Coastal Design Engineer with the Ministry of Works and Transport where she
hopes to gain substantial work experience in the field.
Page 35 APETT Engineering Magazine June 2019
A Sustainable Solution for Red Mud in
the Caribbean.
By: Gino Hosein and Professor Indrajit Ray
1.0 Introduction
Due to the current mining of bauxite in the Jamaica and
Guyana, there has been a large quantity of bauxite waste
material or “red mud” being produced. Red mud is one of
the major solid waste residue generated due to digestion of
bauxite ores with caustic soda during the Bayer process of
alumina production. According to the Jamaica Bauxite Insti-
tute (JBI) approximate 4 million tonnes of red mud are pro-
duced annually. The bauxite company of Guyana states on
its website that the annual production capacity of bauxite is
2.3 million tonnes. About 1 tonne of alumina (aluminum
oxide) is produced from 3 tonnes of bauxite, and depending
on the quality of ores processed, 1-2.5 tons of red mud
waste is generated for every tonne of alumina. By applying
this rule of thumb, it may be estimated that 1.15 million
tonnes of red mud are produced annually in Guyana. This
works out to be 5.15 million tonnes of red mud produced in
the Caribbean out of an overall 170 million tonnes being
produced globally in 2015 (Hua, Heal and Friesl-Hanl, 2017)
which means an average of 3.2% of the annual global produc-
tion of red mud. This signifies that red mud is one of the
most abundant waste materials produced in the Caribbean.
2.0 Reduction of the Disposal Problem by Replacing Port-
land cement
The problem faced regionally can be solved by finding a
sustainable method to dispose of the red mud. Dust repre-
sents a major pollution factor in the bauxite industry
(Traore, Traore and Diakite, 2014). The current global prac-
tices for disposal of red mud include sea disposal, ponds (see
fig. 1), mud farming, and dry mud stacking. All these are
highly unsustainable and environmentally hazardous due to
high alkaline nature of the sludge. The projected global pro-
duction of cement by 2020 is 4.4 billion metric tonnes. If we
apply the pro rata method of the Caribbean population (44
million) against the world population (7500 million) we get
the following projected consumption of cement for the Car-
ibbean: (44/7500)*4.4 = 0.03 billion metric tonnes. This is 30
million metric tonnes of cement estimated to be consumed
by the Caribbean in 2020. It is well known that manufactur-
ing of cement causes large amount of greenhouse gas (1000
kg of Portland cement produces about 927 kg of CO2)
(Portland Cement Association) and cement kiln dust (CKD)
that is disposed as landfill. The energy consumption due to
Portland cement manufacturing is 5 GJ/tonne. Due to large
quantities of Portland cement production, the total energy
consumption becomes quite high. If a large fraction of the
5.15 million tonnes of estimated red mud from the Caribbe-
an can be recycled to replace part of the Portland cement
clinker – this will readily lead to a sustainable solution.
Figure 1.0: Typical Pond Disposal for “Red Mud”
3.0 Reduction of the Disposal Problem by Replacing Port-
land cement
For manufacturing purpose, the red mud can be either used
to make composite cement by replacing part of the Portland
cement or it can be used to produce alkali activated cement.
Due to simplicity and availability of more research data the
greater potential will be to manufacture composite cement.
Research revealed that up to 20 wt% of cement can be sub-
stituted by red mud to produce durable mortars and con-
crete (Ribeiro et al. 2013). The incorporation of red mud to
produce composite cement has many benefits such as: (1)
recycling of the hazardous red mud will eliminate the dispos-
al problem; (2) the replacement of Portland cement clinker
Page 36
with red mud will reduce CO2 emission and CKD production; and
(3) reduction of the clinker will reduce energy requirements. Fur-
ther, the concrete made with red mud composite cement will re-
duce the heat of hydration and as a result will reduce the chance of
cracking of concrete structures. The red mud inclusion in the clink-
er will increase the resistivity of concrete which eventually will re-
duce the corrosion potential of reinforcements embedded in con-
crete.
Table 1.0: Chemical Composition for Portland Cement and
Red Mud
As seen in table 1, above, the silica (SiO2) content of red mud is
almost the same as the Portland cement but the alumina (AI2O3)
content is much higher -- both of these values make it a good blend-
ing material with cements. The major concern is the high content of
alkalis which tend to raise the pH value. The high pH is good against
the corrosion attacks on reinforcement and carbonation of con-
crete, but excessive sodium oxide (Na2O) and potassium oxide
(K2O) in the presence of moisture will increase the chance of alkali-
aggregate reaction if the aggregate is a reactive aggregate. However
reactive aggregates are very rare in the Caribbean.
4.0 Potential Challenges for the Implementation of Red
Mud Cement
As with all new products, general acceptance by all stakeholders
including homeowners, clients, contractors, Government, lending
agencies etc. will be slow. More research is needed for quality con-
trol testing of the properties of composite cement with varying
quantities of red mud. Key tests are needed for complying with the
ASTM standards for physical and chemical properties of blended
cement such as oxide compositions, insoluble residues, X-ray dif-
fraction, electron microscopy, pH values, and particle distributions,
surface areas, setting time, mortar strengths, soundness and other
quality control tests. If these tests are done systematically and a
database can be created for the process and the results, it can be
used repeatedly. The import costs have to be accounted for all of
the island states, other than Jamaica and Guyana, who may wish to
use the red mud. The only viable way to import the red mud is to
firstly treat it to a safe environmentally friendly state and barge it to
the various islands. The only other possibility is for Jamaica and
Guyana to produce the actual cement blends and export. The initial
quality control testing along with the import of the materials may
initially raise the cost of production, but for large productions later
on the cost will eventually be much less. The benefits of lower
greenhouse gas, more durable concrete, and recycling of red mud
will be much higher -- thus will significantly lower the cost/benefit of
the product.
5.0 Conclusions
As discussed above, the main conclusions are:
1). Red mud is the most abundant waste material produced in the
Caribbean.
2). There is a need for recycling this waste material in order to
contribute to a sustainable development.
3). It is possible to achieve this by using it as a partial cement re-
placement, up to 20wt%, for mortar and concrete applications.
4). Regionally some of the problems faced are the current disposal
methods, and the potential testing required for implementation of
the “red mud” as a recycled product.
Mr. Gino Hosein. BASc. in Civil Engineering.
MSc. in Construction Management. Diplomas in
Construction and Building and Draughting. Asso-
ciate Member APETT, Chartered Member of the
CIOB (MCIOB). Chartered Construction Manag-
er, 16 years construction experience in civil infra-
structure projects including roads, bridges, retain-
ing walls, infrastructure works to accommodate
MEP services i.e. plumbing, electrical etc., devel-
opments and building works including residential,
commercial, educational, health and recreation buildings. Worked in
pre and post contracts not exceeding US$ 20M. Mr. Hosein’s re-
search, to date, is focused on local challenges of the construction
sector in relation to construction management processes.
Professor Indrajit Ray. BS (Hons) in Civil Engi-
neering, MS in Structural Engineering, Ph.D.in Civil
Engineering. Dr. Indrajit Ray is currently the Pro-
fessor and Programme Coordinator of Construc-
tion Engineering and Management in the Depart-
ment of Civil and Environmental Engineering UWI.
Professor Ray’s research is focused on advanced
and sustainable construction materials, fiber rein-
forced polymer/concrete composite for repair and
strengthening, and local challenges of construction
sectors. Professor Ray has published over 125
peer-reviewed papers in Journals and proceedings, reviewed several
international Journal papers, and funding proposal. He led over US
$5 million externally funded projects. He has supervised over 45 MS
and PhD students and made several international presentations as
invited speaker. Professor Ray is the voting member of ASTM Inter-
national committees on cement and concretes & aggregates, and
member of institution of engineers and geotechnical society.
Page 37 APETT Engineering Magazine June 2019
KMA IN-PLANT COLD RECYCLING
TECHNOLOGY IN ROAD CONSTRUCTION
AND REHABILITATION IN TRINIDAD AND TOBAGO
By: Laurence Bridgemohan, BSc., MSc (UWI), REng. MAPETT, PMP
1.0 Introduction
The cold recycling technology offers a sustainable option for
road construction and rehabilitation, with increased global
applications, of significant success. Its accompanying eco-
nomic, environmental and energy efficient benefits has pro-
moted its use in the construction and rehabilitation of major
heavily trafficked carriageways in Trinidad. The introduction
of the Cold In Place Recycling rehabilitation technique in
Trinidad and Tobago by Danny’s Enterprises Company Lim-
ited has positively impacted the local road building industry,
with its entry over a decade ago, as an innovative pavement
engineering solution
In light of ongoing local infrastructural development, in-
creased demands for suitable road building aggregates have
been placed on our depleting local natural deposits of lim-
ited supplies of quality virgin granular aggregates. Fortunate-
ly, global advancements in equipment and road construction
techniques currently provides opportunity for the improve-
ment of a wide range of locally available aggregates and sub-
sequent satisfactory inclusion into road bases and sub bases.
The versatility offered by the Wirtgen KMA 220 Mobile
Cold Recycling Plant now provides for the utilization of the
Cold In-Plant stabilization technology towards the enhance-
ment of our local available materials. The KMA In-Plant sta-
bilisation technology allows for value to be added to local
aggregate materials not initially considered satisfactory for
pavement construction and rehabilitation application in its
parent form, providing a sustainable material source option
to the road building industry.
Figure 1.0 KMA Cold in Plant Stabilization
The KMA Cold in Plant Stabilization methodology utilizes
the enhancement provided by foamed bitumen, cement and
hydrated lime stabilizing agents, towards the improvement
in strength and durability properties of the aggregate. The
process allows for the controlled blending of dual aggregate
sources and meter controlled addition of stabilizing agents,
resulting in the production of a stabilized aggregate of im-
proved properties. The methodology provides for the
placement of these stabilized aggregates utilizing typical as-
phalt pavers with significant advantages inclusive of im-
proved efficiency, levels and grade controls.
Foamed Bitumen Stabilization
Foamed Bitumen or expanded bitumen is produced when
hot bitumen comes into contact with cold water in the
presence of compressed air. The resultant is vapour trapped
in thousands of tiny bitumen bubbles, which burst upon
mixing, producing splinters that adhere to the finer particles
of the aggregate material. Upon compaction, the mastic cre-
ated between both expanded bitumen and fine components
introduces spot welds to the coarser fraction of the aggre-
gate skeleton, resulting in the
production of a non-
continuously bound layer
(Wirtgen GmbH 2012, 107).
Figure 2.0 Section of BSM1-ACG
compacted layer core
(Bridgemohan, 2018)
Bitumen Stabilised Materials (BSMs)
BSMs produced from foamed bitumen stabilization behave
in a manner similar to unbound granular material, but with a
notable improvement in cohesive strength and a reduced
moisture sensitivity. Its main features include:
• A material with increased cohesion in excess of 3 times
its original value.
• Increased flexural strength.
• Increased moisture sensitivity
Cement Stabilization
Cement stabilization serves as one of the oldest forms of
aggregate treatment utilized in pavement engineering. The
process involves the mixing of selected aggregate materials
with cement and water, which hardens after compaction
and resulting in the production of a stiff, durable pavement
layer. The hydration of the calcium silicate compounds pre-
sent in the cement produces the calcium silicate hydrate
gel, which on hardening and curing holds the aggregate par-
ticles together. The resultant is a stabilized material of in-
creased stiff-
ness, durabil-
ity, compres-
sive and ten-
sile strength.
Figure 3.0 Stabilized Aggregate Pavement Systems
vs Conventional Granular Aggregate Pavement Sys-
tems
Figure 4.0 CBR 5 in-situ subgrade/
KMA BSM1 Stabilized Base
Estimated Capacity = 3 MESA to 4 MESA
Figure 5.0 CBR 5 in-situ subgrade
CBR 80 Unbound Granular Agg. Base
Estimated Capacity < 1 MESA
Owing to its improved shear and flexural strength proper-
ties, stabilized aggregate layers allow for the design and
construction of stabilized pavement systems with overall
increased carrying capacity. In light of these characteristics,
when proposed as alternatives to granular aggregate pave-
ment structures, stabilized pavement systems consist of
structural layers of reduced thicknesses. In addition to the
opportunity for inclusion of a wide range of available mate-
rials, the option for thinner stabilized structural high capaci-
ty layers provides for significant economic savings.
2.0 Typical Benefits of the KMA Cold In Plant Stabiliza-
tion Methodology
• Stabilized material of improved engineering strength
and durability properties consistent with increased
pavement layer carrying capacities and resultant eco-
nomical, durable pavement designs and construction.
• KMA in plant aggregates are laid in single passes, using
the conventional asphalt pavers to the desired thick-
nesses, grade and levels, allowing for higher site pro-
duction rates.
• Provides for blending and enhancement of a wide range
of locally sourced marginal aggregates, with monitored
consistency in quality and subsequent fulfilment of re-
quirements for its inclusion in pavement structures.
• Stabilized aggregate produced in-plant allows the mate-
rial to be pre-mixed, sampled, inspected and tested,
with allowed adjustments to input parameters and mix-
ing times as required.
Page 38
• In-plant stabilization provides for a durable layers as
base or sub base layers in designed pavement struc-
tures and may also be considered for surfaces for tem-
porary unpaved applications.
3.0 Construction Applications In Trinidad
Sir Solomon Hochoy Highway – Southbound vicinity
of Gasparillo Bypass Road: August 2018
Project Details: KMA BSM 1 – RAP Rehabilitation Alternative:
Cold Milling of existing thick asphaltic concrete layers, cold in-
place sub base strengthening, KMA BSM1-RAP layer and reduced
thickness HMA surfacing.
Soogrim Trace Connector, Endeavour Trinidad: 2017
- Ongoing
Project Details: Alternative Stabilized Design: KMA BSM 1 – AGG
base / KMA CSM-RAP sub base – Installation of KMA in-plant
cement stabilized sub base, reduced thickness BSM1-AGG base
and reduced thickness HMA surfacing.
3.0 Conclusion
KMA In-Plant BSM 1 – RAP layer open to vehicular traffic 7
days before asphalt surfacing
The KMA In-plant cold recycling technology serves to ad-
dress concerns with respect to increasing demands for con-
sistent supply of high quality road building aggregates. The
opportunity for
improvement and inclusion of a wider range of available
materials is indeed a sustainable option with immense po-
tential benefits to the local industry. Successful applications
in Trinidad to date are indicative of the value of the tech-
nology as an innovative engineering solution, in suitably
designed and constructed applications.
Page 39
Laurence Bridgemo-han is a national of Trini-
dad and Tobago and Civil Engineer presently in-volved in pavement engi-
neering research, design and construction. Mr. Bridgemohan is a holder
of a BSc in Civil Engineer-ing and MSc in Construc-
tion Management from the University of the West Indies (UWI), St. Augustine Campus and is a current
Civil Engineering PhD Candidate at the UWI, with ongoing research in the field of bitumen stabilized materials. He is a Registered Specialist Engineer with
the Board of Engineering of Trinidad and Tobago, in the field of Roads/Asphalt and is actively engaged in research, design and construction applications utilizing
stabilised materials, as sustainable pavement rehabili-tation and construction solutions. His recent projects include stabilization technology transfer activities in
the Caribbean and Latin/Central American regions, working alongside global industry pioneers, Wirtgen Group (Germany), Loudon International (South Afri-
ca) and Resansil Inc. (Miami).
Page 40 APETT Engineering Magazine June 2019
Page 41 APETT Engineering Magazine June 2019
The “Human” case for Energy Efficiency
in Trinidad and Tobago
By: Sheena Gosine
1.0 Introduction
Social and economic development and population growth
have resulted in increased global energy demand over the
last decade. Worldwide electricity production, one com-
ponent of energy supply, has increased by 76%. Total
worldwide CO2 emissions per Gt per year has however
also increased by 44% from 1993-2011, reaching 31.6 Gt in
2012 (World Energy Council 2013). Despite the rapid in-
crease in energy use, almost 20% of the global population
has no access to electricity (GEA 2012) and over three
billion people still rely on traditional fuels for household
cooking and heating. Consequently, the resulting air pollu-
tion leads to the occurrence of over two million premature
deaths annually, largely women and children. Energy there-
fore is central to addressing major the challenges sustaina-
ble economic and social development and global security
(GEA 2012).
According to International Energy Agency (IEA), the global
demand for electricity has increased, and electricity genera-
tion from fossil fuels has increased from 4,606 TWh in
1973 to 15,396 TWh in 2012 so that 67.9% of the world’s
electricity has been produced by fossil fuels (International
Energy Agency 2014). Governments are increasingly con-
cerned about the security of electricity supply and question
the ability of existing market design and regulatory frame-
works to continue to deliver reliable and efficient electricity
supply in a timely manner (IEA;OECD 2013).
Energy Efficiency refers to technical improvements that
result in using less energy without a reduction in consumer
enjoyment (Hofmeister, 2010). Energy efficiency has been
an active and inexpensive tool to offer environmental pro-
tection, stimulate economic growth and improve energy
security. Historically, the focus of international and domes-
tic energy law has been on maintaining adequate supply of
energy, rather than maximizing generation and utilization of
energy efficiency measures (Bradbook & Richard, 2003).
Reducing energy use through existing processes offers
many possibilities. Some of these options have little or no
cost, energy conservation actions, cost nothing to imple-
ment. Energy conservation is any action that results in the
use of less energy and energy efficiency requires us to use
technology in a way that requires less energy to perform
the same function.
The world’s demand for energy has resulted in low energy
prices, some maintained by unrealistic subsidies avoiding
the “true” social and environmental costs. Energy subsidies
depress economic growth in many different ways. Subsidies
can discourage any type of “investment in the energy sec-
tor; crowd out other public spending that would enhance
growth and over the long term diminish the competitive-
ness of the private sector” (Bauer, et al., 2013). Subsidies to
consumption or production result in the lowering of end-
user prices, this can lead to increased rates of energy use
and act as a deterrent to conserve or use energy more
efficiently (United Nations Environment Programme, 2008).
Two-thirds of global greenhouse-gas emissions are derived
from the energy sector (International Energy Agency 2013)
it will be pivotal in determination of achievement of climate
change goals. On the 12th, December 2015 at COP 21 in
Paris, Parties to the UNFCCC reached a landmark agree-
ment to combat climate change called the “Paris Agree-
ment”. The Paris agreement generally aims to increase the
ability of countries to mitigate the effects of climate change,
and at making finance flows consistent with a low GHG
emissions and climate-resilient pathway (UNFCC, 2015).
The choices people make about how they use energy af-
fects the environment and everyone's lives. The earth’s
average surface temperature has reportedly increased over
the last century. This increase in temperature change has
been linked to anthropogenic sources such as greenhouse
gases (GHG). In 2013, globally energy use accounted for
72% of the GHG emissions, 31% of this was due to electric-
ity and heating utilization (Global Emissions, 2018).
Page 42
2.0 Understanding Electricity in Trinidad and Tobago
T&T meets all of its domestic electricity needs locally and
therefore does not import or export electricity (Manickchand
2011). The power generation sector in Trinidad and Tobago
can be broken down into: one transmission and distribution
company (the Trinidad and Tobago Electricity Commission, or
T&TEC) which purchases its bulk power from three (3) inde-
pendent power producers, namely; The Power Generation
Company of Trinidad and Tobago (PowerGen), Trinity Power
and Trinidad Generation Unlimited (TGU) (Parliament of the
Government of Trinidad and Tobago 2013).
T&TEC purchases electric power for resale to its domestic,
commercial and industrial customers, the company is also
responsible for purchasing natural gas for the generation
companies from the National Gas Company of Trinidad
&Tobago Limited (NGC) (McGuire, Competition in Energy
Markets: Trinidad and Tobago 2007). The total supply availa-
ble with full TGU capacity on the grid is 2,155 megawatts and
the peak demand is 1,322 megawatts (Parliament of the Gov-
ernment of Trinidad and Tobago 2013).
Cheaper electricity is obtained through the economy of scale:
that is, the bigger the generator, the less expensive the power
produced on a per unit basis. However, large generators are
no more than 40% efficient (Martin May 2009) which results in
wasting of natural resources. The efficiency of Trinidad and
Tobago’s the centralized system has come under scrutiny over
the past decade. The evidence suggests that our generating
plants have been deemed inefficient and plans have been made
for replacing them. Our current power generation plants have
low thermal efficiencies. The TGU, La Brea Power Station had
the highest efficiency when compared to the other power
stations as a result of a combined cycle arrangement of
450MW of gas and 270MW free combined cycle that is, it
uses the steam from the 450MW of electricity to generate a
further 270MW of electricity (Parliament of the Government
of Trinidad and Tobago 2013). In T&T the IPP’s receive pay-
ment based on their ability to make power available when
called upon and cannot contractually earn revenue from the
byproducts of energy. T
his type of contractual arrangement does not encourage ener-
gy efficiency upgrades (Driver, 2017). The electricity sector in
Trinidad and Tobago is the second highest contributor to
GHG emissions. The power sector experienced an immense
growth in emissions of 174% growth from 1990-2012
(GoRTT, 2015). Residential customers currently consume
approximately 33% of all electrical energy produced, whereas
industrial and commercial customers utilized approximately
55% and 10.0% respectively (Ministry of Energy and Energy
Industries, 2019)
T&TEC purchases natural gas from NGC in US dollar denomi-
nation and this increases annually at a rate of 4% (RIC 2003).
Sixty percent of T&TEC’s operational costs are from money
spent on fuel purchases and conversion to energy, the financial
viability of T&TEC would be adversely affected if the govern-
ment changes these price structures. (T&TEC 2010). T&TEC
has identified a series of issues that must be addressed for
their sustainability and success: the reduced natural gas re-
serves, the increasing of natural gas prices, rising generation
costs and intensifying global competition and prices for electri-
cal inputs (T&TEC 2010). Over the years fuel has been sold to
T&TEC at reduced costs, however, as electricity demand has
increased so has fuel consumption for electricity generation
and as a result there has been increased expenditure by
T&TEC.
3.0 Electricity Pricing and Demand in Trinidad and Tobago
There are five classes of customers in Trinidad and Tobago:
Residential, Commercial, Industrial, Heavy Industrial and
Street Lighting. The rate structure for residential electricity
customers uses a three-tiered system where the tiers are de-
fined on the basis of electricity usage, which is measured in
kilowatt-hours (kWh) over a two-month billing cycle, the oth-
er customers are billed monthly at a specific energy rate. The
three (3) residential usage categories are as follows: 1-
400kwh, 401-1000kwh and >1000kwh.
The industries are required to also pay maximum demand
charges, they are required to submit the planned Maximum
Demand of their plant in kVA (the Reserve Capacity) when
requesting a supply (RIC 2009).
Page 43
Electricity tariffs in T&T are reviewed annually and within the
review period the annual tariff adjustments have not always
changed. NGC provides T&TEC with the required gas at a
reduced rate determined by government. This is the reason
why Trinidad and Tobago’s customers have the benefits of
the lowest price of electricity in the region (McGuire, Com-
petition in Energy Markets: Trinidad and Tobago 2007).
There has been growth in customer categories over the
years, resulting in increased demand for electricity in Trini-
dad and Tobago. The Energy Chamber of Trinidad and Toba-
go conducted research in 2017 on the analysis of the elec-
tricity subsidy through the Energy Efficiency and Alternative
Energy Committee. Despite having three (3) residential con-
sumption categories at various prices, roughly 43% of all
households in the country fell in the highest usage category
of >1000kwh. In addition the average bi-monthly consump-
tion of these households in 2015 was roughly 2100kwh (The
Energy Chamber of Trinidad and Tobago, 2017). The high
level of electricity consumption in Trinidad and Tobago is
most likely due to the low cost of energy and electricity,
which can influence customers to inefficiently utilize their
appliances and light fixtures (OLADE, 2012).
4.0 The effects of poor electricity utilization on Human
Development
The long-term progress of a country can be assessed
through the HDI and it is measured in three basic dimen-
sions of human development: a long and healthy life, access
to knowledge and a decent standard of living (United Nations
Development Programme, 2019). Trinidad and Tobago’s HDI
value for 2018 is 0.784 which is in the high human develop-
ment category, the country ranks at 69 out of 189 countries
and territories (United Nations Development Programme,
2019).
An empirical analysis concluded, that there is a clear correla-
tion between electricity consumption per capita and social
and economic development indices such as GDP (Leung &
Meisen, 2007) . T&T’s level of socioeconomic development
should be higher due to its high electrical energy consump-
tion. However, this translation is not observed when socio
economic development is measured in terms of the Human
Development Index (HDI) (UNDP, 2011).
Economic growth can be stimulated by increasing electricity
consumption per capita and hence indirectly achieve en-
hanced social development (Leung & Meisen, 2007). The HDI
of T&T can be found to be considerably lower than that of
countries that have similar per capita energy and electricity.
T&T has shown poor utilization of its electrical energy, which
has contributed to T&T’s lower HDI value, this value indi-
cates that the country should be benefitting more from its
high energy consumption, but it does not (Ugursal, 2011).
The Energy Chamber stated “there are approximately
400,000 households in T&T putting the average number of
persons per household at ~3 persons per home. While the
average North American household is slightly smaller at ~2.6
persons per household there is still a large gap between both
the standard of living and income levels between T&T and
North America”.
One contributing factor to this lower HDI value is the level
of poverty and hunger present in T&T, despite its developed
status. Poverty plagues the society, over 20 per cent of peo-
ple are assessed as living below the poverty line.
(Commonwealth Foundation, 2013). Another contributing
factor to the lower HDI value of T&T is child mortality that
is, the “probability of dying between birth and five years of
age per 1,000 live births”. Countries with high GDP values
normally have lower child mortality rates. The research con-
ducted by the Energy Chamber also identified a relationship
between the level of income and electricity utilization in
T&T. The Energy Chamber stated, “43% of homes in T&T
have a consumption level that is on par with the average
North American home, twice that of the average European
home and 3 times the global average. Moreover, 70% of all
residential power in Trinidad and Tobago is consumed by
this 43% which to some extent illustrates the level of income
inequality in the country”.
Page 44
5.0 The future of Energy Efficiency in Trinidad and Tobago
One of the policy objectives of the Government of Republic
of Trinidad and Tobago is to promote energy efficiency and
energy conservation across all sectors, in order to reduce
our “Carbon Footprint” as well as to better utilize our finite
petroleum resources. The government’s target is that by the
year 2021, at least 10% of the electricity generated in T&T
should be from renewable sources.
The United Nations Sustainable Development Goals (SDGs)
which were adopted in 2015, with goals to combat climate
change and its impact as well as to ensure sustainable con-
sumption and production patterns. In December 2015, the
Trinidad and Tobago adopted the Paris agreement and rati-
fied it on February, 22nd, 2018. This agreement serves as a
legal framework for reducing emissions in the post 2020 pe-
riod.
As part of this Agreement, Trinidad and Tobago’s Nationally
Determined Contributions was derived from the Carbon
Reduction Strategy. T&T’s Carbon reduction strategy (CRS)
identified and assessed greenhouse gas mitigation options for
the major emitting sectors: Power Generation, Industry and
Transportation. Business as Usual (BAU) scenarios were
developed to 2040.
As per the Nationally Determined Contribution, T&T aims
to achieve a reduction objective in overall emissions by 15%
in these major sectors by 2030. The mitigation options also
included direct technology interventions such as Renewable
Energy and Energy Efficiency.
Energy efficiency and energy conservation are the first steps
to proper utilization of electrical energy. Energy conservation
is not the sole responsibility of any single entity, we are all
stakeholders. All citizens, in our homes and our workplaces,
can undertake energy efficiency practices. If we can success-
fully implement energy efficiency and conservation measures,
we can potentially achieve significant changes and positive
environmental impacts.
Sheena R. Gosine is the first
woman to graduate with a dis-tinction in Renewable Energy Technology from the Depart-
ment of Physics at the University of the West Indies, St Augustine. She has a unique background
both in the fields or Education and Sustainable Energy in Trini-dad and Tobago. She has over a
decade of experience having served as an educator: in Physics, Level 1 City and Guilds MEEET and a Masters Level Energy Efficiency Course. Sheena has also served as a Sustainable Energy Policy Analyst,
an Energy Auditor and an Energy Efficiency Con-sultant . Sheena has worked throughout the Carib-bean region; participating in numerous Energy Au-
dits at some of the largest hotels in the Caribbean Region.
She has been instrumental in providing technical
assistance to the newly institutionalized Renewable Energy Division of the Ministry of Energy and Energy Industries (MEEI) in Trinidad and Tobago. Sheena
was one of three individuals to serve on the secre-tariat to the Inter-Agency Committee for the Evalu-ation of Expressions of Interests (EOIs) for a Waste
to Energy (WtE) Facility at the Beetham Landfill. She continued to serve on the secretariat to the Inter-Agency for the Evaluation of Expressions of Inter-
ests (EOIs) for Renewable Energy (RE) Projects.
Page 45 APETT Engineering Magazine June 2019
Hand-Held NPK Sensor
By: Dillon Boodoo
1.0 Objectives
The specific objectives of the project are as follows:
1. Select or build the necessary sensors and interface into the Human Machine Interface (HMI) for the Hydroponic
Experimental Automated Platform (HEAP) 2. Research models of plant growth for selected categories
of crops. 2. Research the correlation between vegetative indices and usable spectrums as they relate to nutrient concentrations
visible in the canopy.
2.0 Abstract
For the past fifteen (15) years, NASA has been working on using plants for human beings as a source of sustenance
with regards to long-term habitation in space. (Stuster, 1986) The Controlled Ecological Life Support System (CELSS) is one of the programs that addresses the prob-
lems faced with long-term human habitation. Currently, the systems being developed are hydroponic-based. However, these systems suffer from problems that results from nutri-
ent imbalances and deficiencies. (W.A. Hill) If a system can be developed that readily and accurately identifies incipient nutrient stress in plants, growth and development could be
vastly improved with regards to product quality. Since nu-trient deficiencies give rise to physical symptoms of degra-dation, the spectral responses and detection of these nutri-ents can provide useful insight into treating with deficien-
cies before they occur. There are currently remote-sensing methods used to detect the Nitrogen content in plants but very little information is present about the detection of
Phosphorus and Potassium, the two most important mac-ronutrients essential to plant growth after Nitrogen. (Chong Yen Mee) This project is geared towards the devel-
opment and implementation of control strategies to readily determine the Nitrogen (N), Phosphorus (P) and Potassium (K) levels in lettuce plants. Absorbance measurement cir-
cuits that make use of Light Emitting Diodes (LEDs) were used to detect the nutrient presence and the information was processed using the Arduino Bluno, an Ardunio Uno
with a built-in Bluetooth module. This information was then presented to the user via an Android application where
further diagnosis can be carried out. Although there were
many challenges faced, the sensors were successfully built, all major components were successfully integrated and rele-vant research was carried out on the models of plant
growth applicable to the scope of this project. 3.0 Models of Plant Growth—Green lab
The GreenLab Model is a type of Functional Structural
Plant Model (FSPM) (Zhao). It consists of two major com-ponents namely: 1. The Growth Engine Model 2. The 3-D Visualization Model
The Growth Engine Model is broken up into the Structural and Functional model. The Structural Model deals with the
plant’s organs, in particular, their production and develop-ment. The Functional Model deals with the eco-physiological aspect that is, how environmental factors in-
fluence plant growth and structure. From the Structural Model, intermediate data is set to be processed using 3-D Visualization where graphical data is produced as output.
From the Functional Model, simulation data is produced as output. (Zhao) This is seen below.
Figure 1.0: The overall workings of the GreenLab model, taken from https:/www.H_Hu_03.pdf
Upon first impression, plant growth may seem like a stag-nant process. However, by modelling plant organs and taking into consideration environmental factors, we can
optimize the growth and production rate of plants. The full model equations for the GreenLab Model can be seen in the next page.
Page 46
For the following equations, the terms are as follows:
Equation for structure, that is, the number of organs in the current cycle that appears at a current time:
Where: up,q – Number of units that originate from the shoot at a time, t for a specified physiological age
bp,q– Number of auxiliary structures appearing at a cycle, t, for a specified physiological age
It should be noted that the sequences described can be either deterministic or stochastic.
Equation for plant demand at a specified cycle, n:
Equation for the biomass denoted to the organs:
Where: i - The organ age - expressed in cycles
Equation for total functioning leaf area S (n) at a specified cycle, n:
Where: – Leaf blade function at an age, j
– Blade age with
– Number of cycles before leaf deterioration
– Thickness of the leaf blade
Page 47
Equation for the production at crop level:
By replacing S (n) in equation 5, with the S (n) expression as
seen in equation 4, we get the GreenLab production equa-
tion (Plant Growth Architecture and Production, n.d.):
The GreenLab Model has the advantage of being flexible in
that it has the ability to host many processes, allowing for a
range of applications. The detailed description of the plant’s
structure allows for visualization applications as well as pest
interaction. However, the model is difficult to calibrate and
validate. In addition, partitioning and biomass transport incur
high costs. Finally, predictive behaviours require multiple
experiments which is time-consuming.
4.0 Correlation between Vegetative Indices and Usable
Spectrums
This section deals with the spectral properties of Nitrogen
(N), Phosphorus (P) and Potassium (K) and the relevant
wavelengths at which these macronutrients can be detected.
A spectral reflectance index (SVI) gives a measure of the
green vegetation in a crop field (Spectral Vegetation Indices
(SVIs)). When data is collected via remote-sensing technolo-
gies, algebraic equations are then applied to further improve
on the results. Green vegetation has a unique spectral pat-
tern. This is seen at the Visible (VIS) and Near-Infrared (NIR)
wavelength range. A low reflectance value is seen on other
regions of the reflectance spectrum, which is why leaves ap-
pear green to human eyes (Spectral Vegetation Indices
(SVIs)). (SPECTRAL VEGETATION INDICES (SVIs), n.d.).
Most SVIs are determined by looking at the values in the
visible (VIS) and near-infrared (NIR) range. In the red region,
chlorophyll content can be detected (I. Filella, 2007) and the
NIR reflectance gives an idea of the mesophyll structure of
the plant. Two types of SVI used are the Ratio Vegetation
Index (RVI) and the Normalized Difference Vegetation Index
(NDVI) (SPECTRAL VEGETATION INDICES (SVIs), n.d.).
Ratio Vegetation Index (RVI)
This is given by the formula:
The Simple Ratio (SR) is close to 1 if the red and NIR bands
have similar reflectance values. High SR values are usually on
orders of 30.
Normalized Difference Vegetation Index (NDVI)
This is given by the formula:
For the NDVI, normalization is applied in an attempt to minimize
illumination levels. The NDVI range is from 1 to -1. A value of 0
indicates there is no vegetation while a value closest to 1 indicates a
high value of green leaf presence. This high value also indicates a
high biomass value.
4.10 Nitrogen Detection
The importance of Nitrogen (N) in plants cannot be over-
stated. Of all the nutrients essential to plant growth, Nitro-
gen is the most important as it facilitates proper plant
growth as well as the production of healthy fruits and vegeta-
bles (Phoslab Testing Laboratories, 2013). Nitrogen is an
essential component of amino acids, normally referred to as
the building blocks of plant proteins which is essential in the
development of plant tissues such as the cell membrane
(Geenway, 2016).
Page 48
In addition, Nitrogen is a major component of vitamins and
aids in the production as well as usage of carbohydrates. The
most important function of Nitrogen in plants is its role in
the process of photosynthesis. N is an important component
in chlorophyll, the molecule which allows the absorption of
light energy to facilitate plant growth (Geenway, 2016). Since
N correlates directly with the chlorophyll content in the
leaves, we can determine a plant’s N level by monitoring the
chlorophyll presence. The wavelength at which the maximum
absorption of chlorophyll occurs is at 690 nm. This is re-
ferred to as the red-edge position and the strong absorption
comes as a result of the scattering of light in the leaf due to
the mesophyll structure. Although it was demonstrated that
total N content could be detected at both the VIS and NIR
wavelengths, the VIS bands were found to be the best pre-
diction of chlorophyll content.
Figure 2.0: Wavelength at which Nitrogen is absorbed in
plants.
4.20 Phosphorus Detection
Phosphorus (P) is another macronutrient essential for plant
growth as it is required for optimum growth and reproduc-
tion of plants (Functions of Phosphorus in Plants, 1999). The
range of total P concentration ranges from (0.1%-0.5%) in
agricultural crops (Soil Nutrient Management, n.d.). In addi-
tion, healthy P content improves the quality of fruits and
vegetables by improving crop maturity and allows for early
root formation and growth (Roles of the 16 Essential Nutri-
ents in Crop Development). P is also a major component
adenosine di-phosphate (ADP) and adenosine tri-phosphate
(ATP) which affects processes such as respiration, membrane
transport and biosynthesis. P stress levels leads to an in-
crease in a molecule known as anthocyanin which causes a
purple coloration in leaves (Pimstein, 2010). Therefore, by
monitoring the presence of anthocyanin, we can determine
the level of P stress experienced by the crop. Anthocyanin is
detected in the wavelength range of 400-550 nm (Mee). At
the early stage of plant growth, P-deficiency symptoms can
be detected at the NIR wavelengths of 730 nm and 930 nm
(Bansal, Field Crops Research, 2010). It is at the V6 stage
(wavelengths around 440-445 nm), anthocyanin can be de-
tected (Pimstein, 2010).
Figure 3.0: The wavelength at which Phosphorus is ab-
sorbed in plants,.
4.30 Phosphorus Detection
Potassium (K) is considered by many to be the most essen-
tial nutrient needed by a plant after Nitrogen. Since K is not
used in structural makeup of pigments and molecules essen-
tial to plant growth, it is difficult to detect by looking at the
absorption wavelengths of these pigments needed by the
plant. K prevents the plant from wilting through maintenance
of the plant’s turgor pressure. A healthy K level results in the
proper functioning of the stomata, which allows water va-
pour, Carbon Dioxide and Oxygen to leave the plant
(Functions of Potassium in Plants, 1998). These gases are
produced as a result of photosynthesis in the plant. In addi-
tion, healthy levels of K allow for disease resistance and im-
proves the quality of seeds and fruits. In order to determine
the K level in the plant, we must be able to effectively meas-
ure the K+ ions in the plant. Potassium concentration in
plants is done through absorption of the K+ ion. This is de-
tected at a wavelength of 517 nm which lies in the green
portion of the visible spectrum (Sodium and Potassium Indi-
cators and ionophores, n.d.). Although excitation can occur
at 488 nm, 517 nm represents the wavelength where the
maximum excitation occurs. Although the monitoring of sto-
matal opening can be done with red and blue light (Assmann,
1999), using green light gives the best indication of stomatal
activity, which can be used to detect the K level in plants. In
addition, green light is also preferred to red and blue light for
stomatal activity as the stomatal pores are located deeper
within the leaf’s structure and green light penetrates deeper
than red or blue light.
5.0 System Overview
•Develop absorbance measurement circuits
•Design switching configuration for the above circuits
•Automate the acquisition of results using a microcon-troller
•Present the results to the user via an Android applica-tion
Page 49
5.10 Absorbance Measurement Circuits
Absorbance (A) is a substance’s ability to absorb light or
electromagnetic radiation at a specified wavelength
(Absorbance Definition, n.d.). Research shows at different
wavelengths, we are able to determine the Nitrogen (N),
Phosphorus (P) and Potassium (K) level in the plant. For the
Absorbance measurement circuits, light sources at the speci-
fied wavelengths were used to determine the nutrient con-
tent in the plant. The light sources used were Light Emitting
Diodes (LEDs).
On a global ranking scale, LEDs are third (3rd) when it comes
to energy conversion efficiency, that is, the conversion of
electricity to light. In addition, they cover a broad range of
wavelengths and are continuously increasing in emission in-
tensity. Also, their small size, minimal cost as well as stability
with regards to light fluctuation made them ideal light
sources to be used in this project. Typically, for light sensing
devices, LEDs are used together with photodiodes. The LED
acts as the light source emitter while the photodiode acts as
the photo-sensor. However, it was suggested that two LEDs
of similar wavelength can be used for sensing by using one
LED in forward-bias and the other in reverse-bias
(Dasgupta). In this configuration, the LED in forward bias
acts as the light source while the LED in reverse bias acts as
the photo-sensor. Since LEDs are known to be more sensi-
tive to the same wavelength that it emits (Shin, 2013) when
compared to photodiodes, this set-up was chosen for this
project. The figure below shows the Multisim circuit diagram
for the Absorbance measuring circuit.
Figure 4.0: The Multisim design Absorbance measuring cir-
cuit
As seen above, the forward-biased LED acts as the emitter
while the reverse-biased LED acts as the receiver. Light from
the emitter LED illuminates the emitting chip of the receiver
LED. When this occurs, a small amount of current is pro-
duced. The current produced is then passed through an op-
erational amplifier to amplify the photocurrent on the receiv-
er LED. This circuit amplifies the current produced by 106, as
seen by the 1 MΩ resistor. This is due to the photocurrent
produced, which is usually in the range of Nano Amperes
(nA). Since this current is minute in nature, a low noise JFET-
Input operational amplifier must be used to detect this cur-
rent. Therefore, the TLO71CP is most appropriate for use in
this circuit. This circuit was initially designed on a bread-
board and can be seen below (Figure 5). Following this is a
Multisim design (Figure 18) showing the operation of the
amplification using the TLO71 and the 1 MΩ resistor in the
current-to-voltage converter.
Absorbance Calculation
Since the voltage at the output is directly proportional to the
light received by the emitter LED, we can modify Beer’s Law
5.20 Methods of Switching
Relays and transistors were the two methods explored in
determining the switching configuration. In terms of the
switching speed, transistors are much quicker than relays,
taking a few picoseconds when compared to the 50 millisec-
onds taken by the relay (Relay vs Transistor?, n.d.). In addi-
tion, electromagnetic problems can occur when using relays
while transistors emit little to no electromagnetic interfer-
ence. When in the “on” state, relays consume a lot of cur-
rent while transistors do not. Relays also need a greater
switching voltage (around 6 Volts) when compared to the
transistor (0.5 Volts or less). Relays are often used for heavi-
er loads for example 1.5 Amperes or anything above 18
Volts. For these reasons, the transistor configuration was
chosen as the means of switching.
Page 50
Figure 5.0: Absorbance circuit with switching configuration
Printed Circuit Board (PCB) Design
The design for the sensors were done with the goal of mak-
ing the system portable so that plant health could be deter-
mined over a wide range of environments. The initial design
on the breadboard was successful but the breadboard itself
was bulky as only a portion of the board was being used for
the circuits. In an attempt to save on space, a PCB design
was done using the Autodesk Eagle software. The schematic
for the circuit was first built followed by the connection of
copper lines in the circuits to create the finished product.
This can be seen below.:
Figure 6.0: The layout of the finished PCB design
5.40 The Microcontroller
From the Absorbance measurement circuits, the manual
method of collecting results using the multimeter and calcu-
lator was time-consuming at times and the threat of human
errors when recording these values was imminent. These
materials may not be available in a hydroponic system, when
accurate determination of plant health is of paramount im-
portance. To automate the acquisition of results, a micro-
controller was selected for this project. The microcontrol-
lers that were considered included the Arduino Nano and
Raspberry Pi. The project scope was examined to determine
which would be more suitable. In this project, the microcon-
troller was used for switching between the different sensing
circuits and for collecting the Absorbance values. Due to the
simplicity and repetitive nature of these tasks, the Arduino
Nano was selected. One instance where the Raspberry Pi
would have been preferred is if there were more complex
tasks with intense calculations. Thus, the Arduino Nano was
selected. However, to further improve on the delivery of
results, it was concluded that wireless communication with
the user would be the best method to obtain results in a
variety of environments. As a result, Bluetooth technology
was examined to allow this communication. The HC-O5 is a
Bluetooth module that can be used with the Arduino Nano
to communicate with the user wirelessly. However, there
would need to be connection with the Bluetooth module
which would mean more circuitry into the system. To avoid
this, the Arduino Bluno was selected. The Arduino Bluno
Nano is essentially an Arduino Nano with a built-in Blue-
tooth module to allow wireless transfer of information. As a
result, the Bluno Nano was selected as the microcontroller
for this project.
Figure 7.0: Image of the Arduino Bluno Nano, taken from
https://thepihut.com/products/bluno-nano-an-arduino-nano-
with-bluetooth-4-0
Page 51
5.40 The Android Application
Android Studio was chosen due to its simplistic nature when
accessing native components such as the Bluetooth technolo-
gy from the Bluno Nano. Android Studio has a rich layout
editor that allows you to preview layouts on screen configu-
rations. In addition, there is adequate support and the soft-
ware allows for faster design and testing. Alternatives to An-
droid Studio included means of hybrid development which
would make writing code to the Bluno difficult because of
the levels of abstractions from the native components. An
example of hybrid development would be the creation of a
mobile responsive application, with the ability to work on
both Android and iOS devices. In order to create the inter-
face, a Bluetooth application to read the values from the Blu-
no Nano was modified to include buttons to store the values
in a database. The basic opening screen as well as the
“Nutrient History” page are shown below. Following this is a
figure which shows interconnection of all components for
nutrient detection.
Figure 8.0: The opening page of the application
Figure 9.0: The Nutrient History page of the application
Figure 10.0: Interconnection of all sub-components
6.0 Discussion
The sensing circuits designed were able to give Absorbance
values to denote the nutrient content in the lettuce leaf. It
was noted that of the three (3) circuits, Nitrogen gave the
most appropriate results when a healthy and unhealthy leaf
was placed in the circuit. For the healthy leaf, a relatively
high value was recorded which meant chlorophyll, a pigment,
Page 52
absorbed the light at this wavelength (approximately 690
nm). In addition, an unhealthy leaf gave a low Absorbance
value which meant that the chlorophyll “a” pigment was not
present in a healthy amount to give an appropriate Absorb-
ance value. In the research conducted, experiments acknowl-
edged the fact that a healthy chlorophyll a concentration
correlates directly with a healthy N level. Similar responses
were seen with P and K detection. The circuit values ac-
quired further cemented this fact.
The Hoagland’s solution is a hydroponic nutrient solution
that contains the appropriate concentrations of the major
nutrients essential to plant growth. This can be extremely
useful in the design of sensors to determine plant health. By
monitoring the nutrient content of the plant via the sensors,
we can determine exactly how much of a nutrient a plant
requires or does not require. By doing this, we are now able
to effectively allocate nutrient resource to the plant whilst
improving the natural growth of the plant. If a known con-
centration of a particular nutrient is known and the sensors
are applied it, the absorbance value can be denoted (solely
for Nitrogen at this point). Using interpolation, we can now
begin to generate a calibration curve that follows the equa-
tion:
Once this calibration curve is developed, we can now begin
to calculate the absorbance level of a nutrient at a particular
concentration. Through this method, we are also able to
better diagnose the plant with regards to nutrient allocation.
The chlorophyll “a” pigment was the only major pigment
targeted in this project. A better representation of the health
level of the plant may be able to be determined by monitor-
ing the major pigments the plant requires rather than the
actual nutrient present in the plant. On many occasions, it is
the presence of a particular pigment that leads to the deter-
mination of the nutrient content. For example, the chloro-
phyll “a” to Nitrogen concentration. If we design sensors
that target the pigment, we will be better able to diagnose
the plant using the Hoagland’s solution as well as be able to
generate calibration curves which improves the overall accu-
racy of plant health determination.
7.0 Conclusion
When compared to the existing methods used for Nitrogen
detection, proper application of the NPK sensor will yield
more accurate results in a shorter period of time. As there
are currently no existing methods of Phosphorus and Potas-
sium detection, the sensor is also preferred. Nitrogen is used
in preference to all other macronutrients as it gives an over-
all indication of plant health. However, one must not omit
the other major macronutrients and its effect on plant
health.
Dillon Boodoo holds a B.Sc. (Hons)
in Electrical and Computer Engi-neering from the
UWI, St. Augustine with a Major in Control Systems.
His professional career began as an Operations Ac-count Manager at
Ramps Logistics, then to an Associ-ate Professional at
the UWI followed by an Engineering Trainee at Qual-
itech Machining Services Limited. He is currently a Graduate Trainee at Massy Wood in the Instrumenta-tion and Controls Department and is involved with
Quality Control/Quality Assurance (QA/QC), com-missioning and engineering design. Winner of a nation-al scholarship (2014), Dillon is a private tutor at the
CSEC level and an avid member of the Alumni Associ-ation at his alma mater. In addition, he enjoys going to the gym, playing a myriad of sports including cricket
and football and is extremely passionate about the rocky journey of the West Indies cricket team and
Chelsea FC.
Page 53 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
PROCESS—Design and Selection of Choke Valves Part 1I
• Ahmed A.M Elgibal, Ibrahim S.Nashawi. ‘’Prediction of Two-Phase Flow Through Chokes of Middle East Oil Wells.’’ 1996.
• Alireza Bahadori. ‘’Natural Gas Processing.” 2014.
• Ana Paula Meneguelo, Daniel Da Cunha Ribeiro, Margotto, Romulo, Jeferson Cunha, Reginaldo Miranda, , Ully Misse Moreno Benedito, and. "Chokes in Series Provide Safe Testing Operations in Wells with Abrasive Solids." OTC Brasil, 2017. doi:10.4043/28198-ms.
• Arashi Ajayi and Konopczynski, Michael, "Design of Intelligent Well Downhole Valves for Adjustable Flow Control." SPE Annual Technical Conference and Exhibition, 2004. doi:10.2118/90664-ms.
• Ayoola, Olakunle T., Wendell Wc Delandro, and Ali M. Muslim. "Detecting Production Choke Wear in High Rate Dry Gas Wells using Graphical Trending of Production Parameters: An Offshore Saudi Arabia Gas Field Case Study." SPE Annual Technical Con-ference and Exhibition, 2016. doi:10.2118/182144-ms
• Barton, Neil, Mike Lewis, and Paul Emmerson. "CFD Erosion Prediction in Gas-Liquid-Sand Flow." SPE International Oilfield Corrosion Conference and Exhibition, 2016. doi:10.2118/179926-ms.
• Cain, L.l. "Unique Well Control Choke: Design Objectives vs. Commercial Field Performance." SPE/IADC Drilling Conference, 1987. doi:10.2118/16132-ms.
• Changzhi Lin, Li, Guomei, Yueshe Wang, Renyang He, Xuewen Cao, and Tao Meng. "Numerical simulation of predicting and re-ducing solid particle erosion of solid-liquid two-phase flow in a choke." Petroleum Science 6, no. 1 (2009): 91-97. doi:10.1007/s12182-009-0017-9.
• Ehsan Khamehchi, Ebrahim Reisi, “Sand Production Prediction Using Ratio of Shear Modulus to Bulk Compressibility (case
study).’’ 2015.
• Emad Gharaibah, John D. Friedemann, Paggiaro, Ricardo, and Yongli Zhang. "Prediction of Sand Erosion in Choke Valves - CFD Model Development and Validation against Experiment." OTC Brasil, 2013. doi:10.4043/24271-ms.
• Grant, Lee. "Montney Unconventional Gas Play: Managing Choke Wear Using Flow Coefficient Diagnostics." SPE/CSUR Uncon-ventional Resources Conference, 2015. doi:10.2118/175992-ms.
• Ioana Cristina Grigorescu1. “Erosion-Corrosion Failures in Wellhead Chokes.” Universidad Simon Bolivar, Sartenejas, Baruta, Ca-
racas 1080. Venezuela. Nace Corrosion Conference & Expo 011.
• Master Flo Valve Inc. 2014a. P2E Choke Valve Detail. http://www.masterflo.com/products/chokevalves/p2-e-hun-10000
• Master Flo Valve Inc. 2014b. P3E Choke Valve Detail. http://www.masterflo.com/products/chokevalves/p3-e-hun-10000
• Naeim Nouri Samie. "Practical Engineering Management of Offshore Oil and Gas Platforms." (2016) Google Books. Accessed Feb-ruary 07, 2018. https://books.google.tt/books?id=0XU1CwAAQBAJ&pg=PA212&lpg=PA212&dq=systems%2Band%2Bequipment%2Bfor%2Boffshore%2Bplatform%2Bdesign%2Bchapter%2B3&source=bl&ots=rE815KQ0AW&sig=N0BhfXo8At0ZMj2T-61rR7xYZag&hl=en&sa=X&ved=0ahUKEwjXh8qXoJXZAhURy1MKHZycBnoQ6AEILjAC#v=onepage&q=systems%20and%20equipment%20for%20offshore%20platform%20design%20chapter%203&f=false.
• Saeed, Abdullah J. Al, Salah A. Al Mousa, Mohammed A. Al Ajmi, and Martin Odonnell. "Successful Installation of Multistage Choke Valve Technology in Water Flowlines to Reduce High Pressure Drop across Choke Valves." SPE Saudi Arabia Section Annu-al Technical Symposium and Exhibition, 2015. doi:10.2118/177988-ms.
• S. Gorini, A. Maliardi, eni spa, S. Malavasi, G. V. Messa. Politecnico di Milano University. “Enhanced Erosion Preduction for Xtree Valves Lifetime Estimation.’’ OMC Italy, 2017.
• Sherik, Abdelmounam Mustafa. ‘’Trends in oil and gas corrosion research and technologies: production and transmission’’. Duxford, UK: Woodhead Publishing, 2017.
• Tawancy, H.m., and Luai M. Alhems. "Damage analysis of choke bean used in an oil–gas well." Case Studies in Engineering Failure Analysis7 (2016): 56-64. doi:10.1016/j.csefa.2016.08.001.
Page 54 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
MECHANICAL— Automated Identification of Vehicular Accidents from Acoustic Signals Using Artificial Neural Net-
works
• Cafiso, S., Di Graziano, A., and Pappalardo, G. In-vehicle stereo vision system for identification of traffic conflicts between bus and
pedestrian Journal of Traffic and Transportation Engineering, Vol.4, No.1, pp. 3-13.
• Dadashova, B., Arenas-Ramirez, B., Mira-McWilliams, J., and Aparicio-Izquierdo, F. (2016), “Methodological development for selec-
tion of significant predictors explaining fata road accidents”, Accident Analysis and Prevention, Vol.90, pp.82-94.
• de Ona, J., Lopez, G., Mujalli, R., and Calvo, F.J. (2013), “Analysis of traffic accidents on rural highways using Latent Class Clustering
and Bayesian Networks”, Accident Analysis and Prevention, Vol51, pp.1-10.
• Delen, D., Sharda, R., and Bessonov, M. (2006), “Identifying significant predictors of injury severity in traffic accidents using a series
of artificial neural networks”, Accident Analysis and Prevention, Vol.38, pp.434-444.
• Evtiukov, S., Kurakina, E., Lukinskiy, V., and Ushakov, A. (2017), “Methods of accident reconstruction and investigation given the
parameters of vehicle condition and road environment”, Transportation research Procedia, Vol. 20, pp. 185-192.
• George, J., Cyril, A., Koshy, B. I., and Mary, L. (2013a), “Exploring sound signature for vehicle detection and classification using
ANN”, International Journal on Soft Computing, Vo4, No.2, pp. 29-36.
• George, J., Mary, L., and Riyas, K. (2013b), “Vehicle detection and classification from acoustic signal using ANN and KNN”, Pro-
ceedings of the International Conference on Control Communication and Computing, Thiruvananthapuram, India, Decem-
ber,pp.436-439. DOI: 10.1109/ICCC.2013.6731694.
• Kononen, D.W., Flannagan, C.A.A., and Wang, S.C. (2011), “Identification and validation of a logistic regression model for predict-
ing serious injuries associated with motor vehicle crashes”, Accident Analysis and Prevention, Vol. 43, pp. 112-122.
• Li, M., Wang, X.H., and Shi, K. (2017), “Traffic conflict identification technology of vehicle intersection based on vehicle video tra-
jectory extraction”, Procedia Computer Science, Vol.109, pp.963-968.
• Mujali, R.O., and de Ona, J. (2011), “A method for simplifying the analysis of traffic accidents injury severity on two-lane highways
using Bayesian networks”, Journal of Safety Research, Vol. 42, pp.317-326.
• Ozgunduz, E., Turkmen, I., Senturk, T, Elif Karsligil, M., and Gokhan Yavuz, A. (2010), “Vehicle identification using acoustic and seis-
mic signals”, Proceedings of the IEEE Signal processing and Communications Applications Conference (SIU), Diyarbakir, Turkey,
April, pp.941 944 DOI: 10.1109/SIU.2010.5652112.
• WHO (2017), Global Health Observatory Data, World Health Organisation, Available from the website: http://www.who.int/gho/
road_safety/mortality/en/ Accessed on 10 November 2017.
• Wu, H., Siegel, M., and Khosla, P. (1999), Vehicle sound signature recognition by frequency vector principal component analysis
IEEE Instrumentation and Measurement Technology Conference, Vol.48, No.5, October, pp. 429-434.
MECHANICAL— Bioplastics and Environmental Sustainability: Some Thoughts
• Arora, N.K., Fatima, T., Mishra, I., Verma, M., Mishra, J and Mishra, V. (2018) Environmental sustainability: challenges and viable
solutions, Environmental Sustainability, Vol. 1, pp. 309-340, DOI: https://doi.org/10.1007/s42398-018-00038-w
• Babu, R.P., O'Connor, K. and Seeram, R. (2013) Current progress on bio-based polymers and their future trends, Progress in Bio-
materials, Vol. 2, No. 8. DOI: https://doi.org/10.1186/2194-0517-2-8
Page 55 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
MECHANICAL — Bioplastics and Environmental Sustainability: Some Thoughts
• Canelo, A. (2018) New Technology to Dismantle the Largest Plastic Spillway in the Pacific Ocean, The Costa Rican News [Online]
Available at: https://thecostaricanews.com/new-technology-to-dismantle-the-largest-plastic-spillway-in-the-pacific-ocean/ Accessed
on March 20, 2019.
• Echo Instruments (2016) BIOPLASTIC DEGRADATION [Online] Available at: http://www.echoinstruments.eu/applications/
bioplastic-degradation/ Accessed on March 20, 2019.
• Goodland, R. (1995) The Concept of Environmental Sustainability, Annual Review of Ecology and Systematics, Vol. 26, pp. 1-24,
DOI: https://www.jstor.org/stable/2097196.
• Government of the Republic of Trinidad and Tobago (2015) National Waste Recycling Policy 2015 [Online] Available at: http://
www.planning.gov.tt/sites/default/files/WASTE%20RECYCLING%20POLICY%202015%20Final.pdf Accessed on March 11, 2019.
• Lambert, S. and Wagner, M. (2017) Environmental performance of bio-based and biodegradable plastics: the road ahead, Chemical
Society Reviews, Vol. 46, pp. 6855-6871 DOI: https://doi.org/10.1039/C7CS00149E.
• National Oceanic and Atmospheric Administration Marine Debris Program (2018) Approximate Time it takes for Garbage to De-compose in the Environment [Online] Available at: https://www.des.nh.gov/organization/divisions/water/wmb/coastal/trash/
documents/marine_debris.pdf Accessed on March 11, 2019.
• PLASTICS (2018) Position Paper on Degradable Additives [online] Available at: https://www.plasticsindustry.org/sites/default/
files/2018%20PLASTICS%20-%20Position%20Paper%20on%20Degradable%20Additives.pdf Accessed on March 18, 2019.
• Selke, S., Auras, R., Nguyen, T.A., Castro Aguirre, E., Cheruvathur, R. and Liu, Y. (2015) Evaluation of Biodegradation-Promoting
Additives for Plastics, Environmental Science & Technology, Vol. 49, No. 6, pp. 3769-3777 DOI: https://doi.org/10.1021/es504258u.
• Shah, M.M. (2008) Sustainable Development, Encyclopedia of Ecology, Reference Module in Earth Systems and Environmental Sci-
ences, pp. 3443-3446 DOI: https://doi.org/10.1016/B978-008045405-4.00633-9.
CIVIL / ENVIRONMENTAL—Assessing The Sensitivity Of The East Coast of Trinidad To Oil Spills
• Castanedo, S., J. A. Juanes, R. Medina, A. Puente, F. Fernandez, M. Olabarrieta, and C. Pombo. 2009. "Oil spill vulnerability assess-
ment integrating physical, biological and socio-economical aspects: application to the Cantabrian coast (Bay of Biscay, Spain)." J Envi-
ron Manage 91 (1):149-59. doi: 10.1016/j.jenvman.2009.07.013.
• Center for Biological Diversity. 2017. "Dispersants." Center for Biological Diversity Accessed 07/05/17. http://
www.biologicaldiversity.org/programs/public_lands/energy/dirty_energy_development/oil_and_gas/gulf_oil_spill/dispersants.html.
• Eckhart, Karen L. 2010. Sea Turtle Recovery Action Plan for the Republic of Trinidad and Tobago.
• Environmental Management Authority. 2005. The Administrative Record for the Environmentally Sensitive Area: Nariva Swamp
Managed Resource Protected Area. Institue of Marine Affairs.
• Environmental Management Authority. 2011. Guidebook for Hydrodynamic Considerations in EIA applications-Appendix B- Coastal
Environments. Environmental Management Authority.
• Georges, Cicely. 1983. A Coastal Classification for Trinidad. edited by Institute of Marine Affairs.
• Institute of Marine Affairs. 2013. A Guide to the Beaches and Bays of Trinidad and Tobago. Institute of Marine Affairs.
• Juman, R.A , and D Ramsewak. 2010. "Status of Mangrove Forests in Trinidad and Tobago." IMA Research report.
Page 56 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
MECHANICAL — Bioplastics and Environmental Sustainability: Some Thoughts
• Canelo, A. (2018) New Technology to Dismantle the Largest Plastic Spillway in the Pacific Ocean, The Costa Rican News [Online]
Available at: https://thecostaricanews.com/new-technology-to-dismantle-the-largest-plastic-spillway-in-the-pacific-ocean/ Accessed
on March 20, 2019.
• Echo Instruments (2016) BIOPLASTIC DEGRADATION [Online] Available at: http://www.echoinstruments.eu/applications/
bioplastic-degradation/ Accessed on March 20, 2019.
• Goodland, R. (1995) The Concept of Environmental Sustainability, Annual Review of Ecology and Systematics, Vol. 26, pp. 1-24,
DOI: https://www.jstor.org/stable/2097196.
• Government of the Republic of Trinidad and Tobago (2015) National Waste Recycling Policy 2015 [Online] Available at: http://
www.planning.gov.tt/sites/default/files/WASTE%20RECYCLING%20POLICY%202015%20Final.pdf Accessed on March 11, 2019.
• Lambert, S. and Wagner, M. (2017) Environmental performance of bio-based and biodegradable plastics: the road ahead, Chemical
Society Reviews, Vol. 46, pp. 6855-6871 DOI: https://doi.org/10.1039/C7CS00149E.
• National Oceanic and Atmospheric Administration Marine Debris Program (2018) Approximate Time it takes for Garbage to De-compose in the Environment [Online] Available at: https://www.des.nh.gov/organization/divisions/water/wmb/coastal/trash/
documents/marine_debris.pdf Accessed on March 11, 2019.
• PLASTICS (2018) Position Paper on Degradable Additives [online] Available at: https://www.plasticsindustry.org/sites/default/
files/2018%20PLASTICS%20-%20Position%20Paper%20on%20Degradable%20Additives.pdf Accessed on March 18, 2019.
• Selke, S., Auras, R., Nguyen, T.A., Castro Aguirre, E., Cheruvathur, R. and Liu, Y. (2015) Evaluation of Biodegradation-Promoting
Additives for Plastics, Environmental Science & Technology, Vol. 49, No. 6, pp. 3769-3777 DOI: https://doi.org/10.1021/es504258u.
• Shah, M.M. (2008) Sustainable Development, Encyclopedia of Ecology, Reference Module in Earth Systems and Environmental Sci-
ences, pp. 3443-3446 DOI: https://doi.org/10.1016/B978-008045405-4.00633-9.
CIVIL / ENVIRONMENTAL—Assessing The Sensitivity Of The East Coast of Trinidad To Oil Spills
• Castanedo, S., J. A. Juanes, R. Medina, A. Puente, F. Fernandez, M. Olabarrieta, and C. Pombo. 2009. "Oil spill vulnerability assess-
ment integrating physical, biological and socio-economical aspects: application to the Cantabrian coast (Bay of Biscay, Spain)." J Envi-
ron Manage 91 (1):149-59. doi: 10.1016/j.jenvman.2009.07.013.
• Center for Biological Diversity. 2017. "Dispersants." Center for Biological Diversity Accessed 07/05/17. http://
www.biologicaldiversity.org/programs/public_lands/energy/dirty_energy_development/oil_and_gas/gulf_oil_spill/dispersants.html.
• Eckhart, Karen L. 2010. Sea Turtle Recovery Action Plan for the Republic of Trinidad and Tobago.
• Environmental Management Authority. 2005. The Administrative Record for the Environmentally Sensitive Area: Nariva Swamp
Managed Resource Protected Area. Institute of Marine Affairs.
• Environmental Management Authority. 2011. Guidebook for Hydrodynamic Considerations in EIA applications-Appendix B- Coastal
Environments. Environmental Management Authority.
• Georges, Cicely. 1983. A Coastal Classification for Trinidad. edited by Institute of Marine Affairs.
• Institute of Marine Affairs. 2013. A Guide to the Beaches and Bays of Trinidad and Tobago. Institute of Marine Affairs.
• Juman, R.A , and D Ramsewak. 2010. "Status of Mangrove Forests in Trinidad and Tobago." IMA Research report.
Page 57 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
CIVIL / ENVIRONMENTAL—Assessing The Sensitivity Of The East Coast of Trinidad To Oil Spills
• Ministry of Energy and Energy Affairs. 2013. National Oil Spill Contingency Plan of Trinidad and Tobago.
• NOAA. 2002. Environmental Sensitivity Index Guidelines Version 3.0. In NOAA Technical Memorandum NOS OR&R 11 National
Oceanic and Atmospheric Administration.
• NOAA. 2014a. "How does oil impact marine life?". National Ocean Service Accessed 01/05/17. http://oceanservice.noaa.gov/facts/
oilimpacts.html.
• NOAA. 2014b. Oil Spills in Mangroves: Planning and Response Considerations. National Oceanic and Atmospheric Administration.
• NOAA. 2014c. "What is an environmental sensitivity index map?". National Oceanic and Atmospheric Administration Accessed
26/04/17. http://oceanservice.noaa.gov/facts/esimap.html.
• Pincinato, F. L, P. S Riedel, and J. C. C Milanelli. 2009. "Modelling an expert GIS system based on knowledge to evaluate oil spill
environmental sensitivity." Ocean & Coastal Management 52 (9):479-486. doi: 10.1016/j.ocecoaman.2009.08.003.
• Snyder, Richard A., Alexandra Vestal, Christina Welch, Gracie Barnes, Robert Pelot, Melissa Ederington-Hagy, and Fredrick Hi-leman. 2014. "Study Finds Clams are Oil Indicator Species for Gulf of Mexico Surf Zones." Gulf of Mexico Research Initiative Ac-
cessed 07/05/17. http://gulfresearchinitiative.org/study-finds-clams-oil-indicator-species-gulf-mexico-surf-zones/.
• The United Kingdom Hydrographic Office Admiraility EasyTide. 2017. "Predict - Select port." Accessed 7/11/17. http://
www.ukho.gov.uk/Easytide/easytide/SelectPort.aspx.
• Wieczorek, Arthur, Dimas Dias-Brito, and João Carlos Carvalho Milanelli. 2007. "Mapping oil spill environmental sensitivity in Car-
doso Island State Park and surroundings areas, São Paulo, Brazil." Ocean & Coastal Management 50 (11-12):872-886. doi: 10.1016/
j.ocecoaman.2007.04.007.
CIVIL / ENVIRONMENTAL—A Sustainable Solution for Red Mud in the Caribbean
• Google search. https://rusal.ru/en/about/11/
• Google Search. http://nepa.gov.jm/LBS-Workshop/Waste%20from%20Bauxite%20and%20Alumina%20Industry%20-%20JBI.pdf. Last
Viewed 6-2-18.
• Hua, Yumei., Kate V. Heal and Wolfgang Friesl-Hanl. 2017. “The use of red mud as an immobiliser for metal/metalloid-
contaminated soil: A review.” Journal of Hazardous Materials. 325. 17-30 DOI 10.1016/j.jhazmat.2016.11.073.
• Ribeiro, D.V., A.S. Silva, J.A. Labrincha, and M.R. Morelli. 2013. “Rheological properties and hydration behaviour of Portland cement
mortars containing calcined red mud.” Canadian Journal of Civil Engineering 40: 557-566 DOI: 10.1139/cjce-2012-0230.
• Traoré D. L., Traoré S. and, Diakité S. 2014. “Bauxite Industry in Guinea and Value Opportunities of the Resulting Red Mud as
Residue for Chemical and Civil Engineering Purposes.” Journal of Civil Engineering Research, 4(1): 14-24 DOI: 10.5923/
j.jce.20140401.03.
Page 58 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
CIVIL / ENVIRONMENTAL— KMA in-plant Cold Recycling Technology in Road Construction and Rehabilitation in
Trinidad and Tobago
• Wirtgen GmbH 2012. Wirtgen Cold Recycling Technology, 1st edition Germany: Wirtgen Group.
ELECTRICAL - The “Human” case for Energy Efficiency in Trinidad and Tobago
• Hofmeister, B. (2010). Bridging the Gap: Using Social Psychology to Design Market Interventions to Overcome the Energy Efficien-
cy Gap in Residential Energy Markets. Southeastern Environmental Journal(19), 7-8.
• IEA. (2014). Energy Security. Retrieved October 10, 2014, from http://www.iea.org/topics/energy security/subtopics/
whatisenergysecurity/
• IEA;OECD. (2013). Secure and Efficient Electricity Supply During the Transition to Low Carbon Power Systems. Retrieved October
7, 2014, from Publications: http://www.iea.org/publications/freepublications/publication/secureandefficientelectricitysupply.pdf
• International Energy Agency. (2014). Key World Energy Statistics 2014. Retrieved October 7, 2014, from Publications: http://
www.iea.org/publications/freepublications/publication/KeyWorld2014.pdf
• Leung, C. S., & Meisen, P. (2007, April 24). How electricity consumption affects social and economic development by comparing low, medium and high human development countries. Retrieved October 17, 2014, from Global Energy Institute Network: http://
www.geni.org/globalenergy/issues/global/qualityoflife/HDI-vs-Electricity-Consumption-2005-07-18.pdf
• Manickchand, N. M. (2011). Renewable energy development in T&T Exploring scenarios for the deployment of solar photovoltaic
systems. Sweden: IIIEE, Lund University.
• McGuire, G. (2007). Competition in Energy Markets: Trinidad and Tobago. Calgary: Latin American Energy Organization.
• MEEA. (2014, September 16). Energy Ministry Always in Open Discussion with Point Lisas Energy Companies. Retrieved October
16, 2014, from http://www.energy.gov.tt/energy-ministry-always-in-open-discussion-with-point-lisas-energy-companies/
• Methanex. (2014, August 26). METHANEX ANTICIPATES LOWER TRINIDAD GAS SUPPLY IN THE SECOND HALF OF 2014.
Retrieved October 16, 2014, from https://www.methanex.com/news/methanex-anticipates-lower-trinidad-gas-supply-second-half-
2014
• Ministry of Energy and Energy Industries. (2019). LNG Exports.
• Newspower. (2014, November 27). T&TEC Update. Retrieved December 2, 2014, from http://news.power102fm.com/ttec-update-
26316
• OLADE, L. A. (2012, November). Energy Economic Information System Energy Statistics, 22.
• Parliament of the Government of Trinidad and Tobago. (2013). Fifteenth Report of the Joint Select Committee on Ministries, Statu-
tory Authorities and State Enterprises on the Administrations and Operations of T&TEC. Port of Spain: GOTT.
• Regulated Industries Commission. (2008). Regulated Industries Commission. Retrieved November 2, 2013, from http://
www.ric.org.tt/cms/content/view/56/49/
• RIC. (2003). Shedding Light on the Regulated Industries Commission's rate review process. Port of Spain: RIC.
• RIC. (2009). Summary of Electricity Rates. Retrieved December 12, 2014, from Trinidad and Tobago Electricity Commission:
https://ttec.co.tt/services/tariffs/images/TTECRatesSummaryEffective1Sept2009.gif
Page 59 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
ELECTRICAL—The “Human” case for Energy Efficiency in Trinidad and Tobago
• Stabroek News. (2014, September 15). Proman set to acquire CLICO Methanol Holdings. Retrieved November 21, 2014, from
http://www.stabroeknews.com/2014/news/regional/09/15/proman-set-acquire-clicos-methanol-holdings/
• T&TEC. (2010). T&TEC Business Plan 2011-2016. Port of Spain: T&TEC.
• The Energy Chamber of Trinidad and Tobago. (2017, May 23). How much electricity do we use in our homes in T&T. Retrieved
from Energy Chamber of Trinidad and Tobago: https://energynow.tt/blog/how-much-electricity-do-we-use-in-our-homes-in-tt
• Ugursal, V. I. (2011). Energy use and changing energy policies of Trinidad and Tobago. Energy Policy 39, 5791–5794.
• UNDP. (2011). United Nations Development Programme. Retrieved November 1, 2013, from United Nations Development Pro-
gramme Reports: http://hdr.undp.org/en/statistics/hdi/
• UNFCC. (2015). What is the Paris Agreement? Retrieved 2019, from https://unfccc.int/process-and-meetings/the-paris-agreement/
what-is-the-paris-agreement
• United Nations. (2005). 2002 Energy Statistics Yearbook. New York: United Nations Publications.
• United Nations Development Programme. (2019). Human development Reports. Retrieved October 17, 2014, from http://
hdr.undp.org/sites/all/themes/hdr_theme/country-notes/TTO.pdf
• United Nations Environment Programme. (2008). Reforming Energy Subsidies Opportunities to Contribute to the Climate Change
Agenda. UNEP Division of Technology, Industry and Economics.
• Williams, C. (2014, September 3). Oil and Gas Journal. Retrieved November 21, 2014, from Ryder Scott: Trinidad and Tobago gas
reserves fell in 2013: http://www.ogj.com/articles/2014/09/ryder-scott-trinidad-and-tobago-s-gas-reserves-fell-in-2013.html
• Williams, C. (2014, October 12). T&T running out of natural gas. Retrieved October 16, 2014, from http://www.guardian.co.tt/
news/2014-10-12/tt-running-out-natural-gas
• Wilson, A. (2014, September 4). Guardian Newspapers. Retrieved October 16, 2014, from http://www.guardian.co.tt/business/2014
-09-03/tt-losing-energy
• World Energy Council. (2010). World Energy Resources Survey 2010. London: World Energy Council.
• World Energy Council. (2013). World Energy Resources 2013 Survey. London: World Energy Council.
• World Energy Council. (2019). Sustainability Index. Retrieved July 14, 2014, from http://www.worldenergy.org/data/sustainability-
index/
Page 60 APETT Engineering Magazine June 2019
ARTICLE REFERENCES
ELECTRICAL—Hand-Held NPK Sensor
• "Changes in Hyperspectral Reflectance Signatures of Lettuce Leaves in Response to Macronutrient Deficiencies." Advances in Space
Research. March 04, 2011. https://www.sciencedirect.com/science/article/pii/S0273117711001499.
• Detecting and Monitoring Plant Nutrient Stress Using Remote Sensing Approaches: A Review. https://scialert.net/fulltext/?
doi=ajps.2017.1.8.
• "ESTIMATING NITRATE, PHOSPHATE, - Plant Physiology." http://www.bing.com/cr?
IG=7E5444361F6B4E6D8C8D6E52B5CA3D3F&CID=176E3E933CFC6808142635543D5369C4&rd=1&h=tYhSoftj8YXzBsxScaY9K
1RRGb3wV00kDaoJ0JpJSMc&v=1&r=http://www.plantphysiol.org/content/plantphysiol/7/2/315.full.pdf&p=DevEx.LB.1,5068.1.
• "Plant Growth Architecture and Production Dynamics." UVED - Plant Growth Modelling - GreenLab Model - Production/Expansion
- Function Equations - Full Model Equation Set. http://greenlab.cirad.fr/GLUVED/html/P2_GLab/Prod/GLprod_equa_013.html.
• "Relay vs. Transistor?" Electrical Engineering Stack Exchange. https://electronics.stackexchange.com/questions/10092/relay-vs-
transistor.
• Stabroek News. (2014, September 15). Proman set to acquire CLICO Methanol Holdings. Retrieved November 21, 2014, from
http://www.stabroeknews.com/2014/news/regional/09/15/proman-set-acquire-clicos-methanol-holdings/.
• Terashima, Ichiro, Fujita, Takashi, Inoue, Takeshi, Chow, Wah Soon, Oguchi, and Riichi. "Green Light Drives Leaf Photosynthesis
More Efficiently than Red Light in Strong White Light: Revisiting the Enigmatic Question of Why Leaves Are Green | Plant and Cell
Physiology | Oxford Academic." OUP Academic. February 25, 2009. https://academic.oup.com/pcp/article/50/4/684/1908367.
• "The Role of Chlorophyll Fluorescence in The Detection of ..." http://www.bing.com/cr?
IG=8F2300217992474B88216997CDC7C9A8&CID=389BDB53A65A6FB21FB6D094A7F56E4B&rd=1&h=ZAF_FDe1hPIeM_nncCTHFLydnJIh4odtU6kA2TyzrH0&v=1&r=http://www.tandfonline.com/doi/abs/10.1080/15476510.1988.10401466?
journalCode=batc19&p=DevEx.LB.1,5508.1.
• "Use of Spectral Reflectance Values for Determining ..." http://www.bing.com/cr?IG=A24FFDE6B259493E9913631B946BB013&CID=18DDD895A4A06B613D3AD352A50F6A60&rd=1&h=gD0jACNXvW4I-
8aKaRkQQQKsk-82tVjyhuFQCHoIb7I&v=1&r=http://jast.modares.ac.ir/index.php/ejb/article/viewfile/56560/journal/
article_10232.html&p=DevEx.LB.1,5488.1.