ANNUAL REPORT

2013

BIOMASS REFINERY & PROCESS DYNAMICS

CONTROL GROUP

ANNUAL REPORT

2013

Pictures frontpage:

Prof.dr. Harry Bitter Dr.ir. Ton van Boxtel Marja Portengen

Dr.ir. Karel Keesman Dr.ir. Gerard van Willigenburg Dr.ing. Rachel van Ooteghem, MSc Ing. Kees van Asselt Ir. Ellen Slegers Prof.dr.ir. Gerrit van Straten Elvira Bozileva, MSc Sanne Moejes, MSc Rungnapha Khiewijit, MSc Ir. Hans Cappon Farnoosh Fasaei, MSc Nurul Khairudin, MSc Emmanuel Amankwah, MSc Ir. Jan Klok Ir. Xin Jin James Atuonwu, MSc

PREFACE GENERAL INFORMATION OVERVIEW PROJECTS SYSTEMS AND CONTROL APPLICATIONS IN AGRICULTURE Optimal Digital Control PROCESS CONTROL IN FOOD AND BIO PROCESS TECHNOLOGY a) Modelling and control of food and biochemical processes b) Energy efficient adsorption drying of healthy food c) Scenario studies for biodiesel production with algae d) Systems analysis of algae biorefinery e) LCA-driven process design for milk-to-powder production f) Processing at farms MATHEMATICS, COMPUTERS AND SIMULATION IN BIO-BASED AND ENVIRONMENTAL SCIENCES a) Acoustic waves for water disinfection, filtration and purification b) Development of new bioreactor for a process for biological sulphide oxidation c) Modelling and optimization of nitrogen (N) transformation in a fertilised flooded rice paddy field for loss reduction d) Computer–aided design and monitoring of WWTPs towards energy and nutrient recovery e) Real-time accounting and simulation of dynamic energy-water-material balances for achieving liveable and sustainable cities of the future f) Model-based innovative water management for resource efficiency in integrated multitrophic aquaculture and horticulture systems (INAPRO) g) Intelligent sensor network and system technologies for fish farming (Sensorfish) CLIMATE AND NUTRIENT CONTROL IN GREENHOUSE PRODUCTION a) Optimal Management of Energy Resources in greenhouses – OMER EDUCATION PhD courses The Art of Modelling Uncertainty Analysis Subjects PUBLICATIONS PhD Theses International Journal Papers Conference Papers Professional Publications, Abstracts and Reports

Preface We are pleased to offer you the 2013 annual report of the Chair for Biomass Refinery & Process Dynamics (BRD) of Wageningen University. The transition from the Chair for Systems and Control (SCO) to the chair for Biomass Refinery and Process Dynamics (BRD) started in 2012 and continued in 2013. Attempts to fill the vacancy of full professor in BRD were not successful. Currently the chairholder of Biobased Commodity Chemistry is acting chair BRD. It is the intention to merge the two groups in the next two years. This will further focus the research of the BRD group towards applications in the field of biomass and waste stream conversions which are relevant for achieving a biobased economy. The expertise of the BRD group is in the field of modelling and control, learning from the data, analysis of the dynamic properties, optimization etc. These expertises are very relevant for the conversion of a biobased economy. Research areas such as reactor control, reaction kinetics, process design etc. can be addressed. This annual report illustrates the research activities during 2013. In addition to the achievements, some facts and figures about PhD courses and MSc thesis work are given, as well as a list of the subjects we teach. If you would like more information about our education and other activities, please consult our website www.wageningenur.nl/sco A word of appreciation has to be addressed to our secretary, Marja Portengen, who collected and compiled all the material, and to all staff members and PhD students for their cooperation and support during the transition we started.

Harry Bitter Ton van Boxtel Chair holder Biobased Chemistry Interim Chair Coordinator Biomass Acting Chair Biomass Refinery and Process Dynamics Refinery and Process Dynamics

Wageningen University

Department of Agrotechnology and Food Sciences

Chair Biomass Refinery & Process Dynamics

General information Address: Wageningen University, Biomass Refinery & Process Dynamics, Bornse Weilanden 9, 6708 WG, Wageningen, The Netherlands. Phone: 0317 - 484955 / 482124. Fax: 0317 - 484957. E-mail: Harry.bitter@wur.nl, Ton.vanboxtel@wur.nl

Staff members Ing. C.J. van Asselt, Prof.dr. J.H. Bitter, Dr.ir. A.J.B. van Boxtel, Dr.ir. K.J. Keesman, Dr.ing. R.J.C. van Ooteghem MSc., M.G. Portengen, Prof.dr.ir. G. van Straten (emeritus from 1 Sept. 2011), Dr.ir. L.G. van Willigenburg.

PhDs-BRD

J.A. Atuonwu MSc, E. Bozileva MSc, F. Fasaei MSc, Ir. X. Jin, R. Khiewijit MSc, S.N. Moejes MSc, ir. P.M. Slegers. External PhDs E.Y.A. Amankwah MSc., ir. H.J. Cappon, ir. J.B.M. Klok, N. Khairudin MSc.

DISC members VLAG members SENSE/WIMEK

members

prof.dr.ir. G. van Straten Prof.dr. J.H. Bitter prof.dr.ir. G. van Straten

dr.ir. K.J. Keesman prof.dr.ir. G. van Straten dr.ir. K.J. Keesman

dr.ir. A.J.B. van Boxtel dr.ir. A.J.B. van Boxtel ir. H.J. Cappon

dr.ir. L.G. van Willigenburg ir. P.M. Slegers ir. J.B.M. Klok

dr.ing. R.J.C. v. Ooteghem MSc ir. X. Jin R. Khiewijit MSc

J.C. Atuonwu MSc N. Khairudin MSc E.Y.A. Amankwah MSc E. Bozileva, MSc F. Fasaei, MSc S.N. Moejes, MSc

Cooperation with KU Leuven, Heverlee (B); University of Hohenheim, Tropical Farm Technology (D); School of Applied Sciences, Cranfield University (UK), Centro de Estudos Florestais, Instituto Superior de Agronomia,Universidade Técnica de Lisboa, 5 INRA, Montpellier (F), Bogor University, Bogor (Indonesia); Diponegoro University, Semaring (Indonesia); IIASA, Laxenburg (A); ICAM (Integrated Catchment Assessment and Management Centre), ANU, Canberra (Australia); Applikon, Schiedam; Plant Sciences Group, Wageningen; Food and Biobased Research, Wageningen Business Unit Glastuinbouw, Bleiswijk; UNESCO-IHE, Delft; Univ. Twente, Systems, Signals and Control group; Univ. Twente, Tissue Regeneration; Hogeschool Zeeland; KWR, Niewegein; Vitens; PRIVA Hortimation, De Lier; Hortimax, Pijnacker; Witteveen + Bos, Deventer; Wetsus, Leeuwarden; Ebbens Engineering, Lochem; Unilever; Friesland Campina; Agentschap NL; Shell Global Solutions International.

Keywords Environment, climate control, food processing, agriculture, greenhouses, modelling, identification, systems theory, control system synthesis, optimal control, neural networks, fuzzy control.

PhD Thesis

PhD Thesis

Activities of Biomass Refinery and Process Dynamics

PROJECTS

SYSTEMS AND CONTROL APPLICATIONS IN AGRICULTURE

Description The aim of this project is to study the applicability of systems and control theory in agricultural engineering and related fields. Major topics are robust control, optimal control, non-linear identification (AI), and modelling, with an emphasis on agricultural engineering applications. This project also serves to initiate possible new PhD projects. Project leader: L.G. van Willigenburg Project and participants: a) Optimal Digital Control (van Willigenburg)

Ongoing work

Optimal Digital Control (Van Willigenburg) The aim of this project is to develop general methodologies for the synthesis of optimal digital controllers for non-linear systems. Our recent research concerning optimal reduced-order LQG controller synthesis for nonlinear systems tracking optimal trajectories triggered a reinvestigation of the controllability, reachability, observability and reconstructability of time-varying linear systems in both continuous and discrete-time. This led to the development of the differential Kalman decomposition in continuous-time and the j-step, k-step Kalman decomposition in discrete-time. These decompositions reveal and establish the temporal changes of reachability/controllability and observability/reconstructability. The linearised system about optimal trajectories may be temporarily uncontrollable and/or temporarily unreconstructable. During these periods the LQG compensator is partly ineffective. This may lead to closed loop instability. To establish the latter temporal stability, temporal stabilizability and temporal detectability are new key system properties. One-step stabilizability and one-step detectability are the discrete-time counter parts of differential stabilizability and differential detectability. The continuous-time results have recently been published online in Optimal Control Applications & Methods. The discrete-time results have been published on-line in IET Control Theory and Applications. Recently the extension of these properties to time-varying linear systems with white stochastic parameters has been studied. They are of practical importance for robust output perturbation feedback design. Currently a publication concerning these results in discrete-time is being written. Using the delta operator the system property compensatability, that determines whether a system with white stochastic parameters can be stabilized means of dynamic output feedback, was carried over to continuous-time where it is new. Also the associated numerical tests and measures to verify this property were carried over. This enabled a complete unification of full and reduced-order LQG controller synthesis for systems with deterministic and white stochastic parameters in both continuous and discrete time. Currently the possible use of the delta operator to compute temporal compensatability properties of continuous and discrete-time systems with white stochastic parameters is being studied. U-D factorization of LQG controller algorithms enhances their numerical efficiency and stability. Currently we investigate the possibility of U-D factorization of similar algorithms for systems with white stochastic parameters. Full-order compensator algorithms have been U-D factored and the result has been published. An interesting by-product concerns the minimal representation of matrix valued white stochastic processes. Reduced-order compensator algorithms have been studied and a publication concerning the results is currently under review. Also alternative algorithms based on Lyapunov equations instead of the conventional generalized Riccati and strengthened optimal projection equations are being studied. Optimal control and LQG compensation require a reasonably well established dynamic systems model. If such a model is not (yet) obtained, active adaptive optimal controllers form an attractive alternative. A dual (active) adaptive controller structure has been proposed and studied by means of several benchmarks. The controller structure aims at enabling a cheap, fast and smooth transition from standard to advanced (multivariable optimal) digital feedback control within industrial environments. The results are very promising. Still they need some improvement before publication. As part of the European project “EU Cafe” adaptive optimal scheduling and control of beer membrane filtration has been studied, mainly by means of simulations of an adaptive digital optimal scheduling and control system. Also some preliminary experiments were performed. A publication concerning these results is almost finished and ready for submission.

Differential Kalmandecomposition

Equivalent almosteverywhere in time

Temporal linearsystem structure

Conventionaltime-varying linear

systems (having constant dimensions)

Deficientrealization theory

Piecewise constantrank systems

(PCR systems, havingpiecewise constant

dimensions)

Well roundedrealization theory

PCR equivalents

System type: Time-varying linear systems with deterministic parameters

Application: Optimal State & Output perturbation feedback design

Least Squares Parameter and State estimator

Receding Horizon Optimal Controller

Compound Cost Criterion

1c pJ J J

JcJ

pJ

0x̂

y u

0 1

x̂

p̂

0p̂

System type: Non-linear systems with a fixed structure but unknown parameters

Application: Adaptive optimal feedback controller design

Delta operator

OldDiscrete-time: Compensatability

Compensatability testCompensatability measure

NewContinuous-time: Compensatability

Compensatability testCompensatability measure

System type: Time-varying linear systems with white stochastic parameters

Application: Optimal/Robust State & Output perturbation feedback design

PROCESS CONTROL IN FOOD AND BIO PROCESS TECHNOLOGY

Description At present, in the food processing and biotechnology industries, mostly traditional control methods are being applied. However, these methods are limited in their abilities for realizing the increasing demands on product quality, and efficiency and flexibility of the production. As a consequence the significance of high performance process control methods grows. The objective of the project is the development and application of high performance control methods and strategies for food processing and bioprocess technology with the aim to (i) improve product quality, (ii) realize flexible operations, (iii) raise process efficiency. This project also serves to initiate possible new PhD projects. Project leader: A.J.B. van Boxtel. Subprojects and participants: a) Modelling and control of food and biochemical processes (E.Y.A. Amankwah, A.J.B. van Boxtel, G. van Straten, R.J.C.

van Ooteghem, A.J.B. van Boxtel) and EU-Café (L.G. van Willigenburg) b) Energy efficient adsorption drying of healthy food (J.C. Atuonwu, X. Jin, A.J.B. van Boxtel) c) Scenario studies for biodiesel production with algae (P.M. Slegers, A.J.B. van Boxtel) d) Systems analysis of algae biorefinery ( F. Fasaei, P.M. Slegers, A.J.B. van Boxtel) e) LCA-driven process design for milk-to-powder production (S.N. Moejes, A.J.B. van Boxtel) f) Processing at farms (R.J.C. van Ooteghem L.G. van Willigenburg, A.J.B. van Boxtel)

Ongoing work a) Modelling and control of food and biotechnological processes (E.Y.A. Amankwah, A.J.B. van Boxtel, G. van

Straten, L.G. van Willigenburg, R.J.C. van Ooteghem) The project aims at exploring novel ways for modelling and control of biotechnological processes, or at exploring the applicability of methods known from other areas. Drying In 2010 a project started on solar adsorption drying systems. This project has a sandwich PhD-funding for E. Amankwah at KNUST in Kumasi Ghana. The aim of this project is to dehumidify air with adsorbents before drying yam. With a good strategy, drying can be continued during the night, while the adsorbent is regenerated during daytime when excess solar energy is available. This systems is called “solar adsorption dryer system” (SADS). The photos give an impression of this system.

a solar adsorption dryer system

In 2012 and 2013 we worked on an experimental program for this system. Initially, the performance of the system was below expectation due to heat losses in the pipes and the size of the drying cabins. By changing the dimensions and piping and position of the solar collectors the performance increased. Next experiments were started to dry yam particles.

Parallel to the SADS experiments the same products were dried in a solar dryer (only during day time) and by open sun drying. The SADS dryer was beneficial in its total drying time. The dried products are now analysed with respect to the functional properties. Parallel to the drying experiments at KNUST in Ghana, the drying properties (sorption isotherms, drying kinetics and shrinkage) of yam were measured under controlled conditions at the laboratories in Wageningen. The data is now subject to analysis. As a first result, variation of the drying rate constant in the initial phase of drying was observed from recursive parameter estimation. It shows that the interaction between water and the food matrix is related to the water content. Modelling alternative methods for small scale sugar processing In the Netherland sugar beets are processed in two large factories. The cost for transportation of the beets towards these centralized locations and the transport of by-products back to the farm or other users is now half of the processing costs. As an alternative, sugar beets can be pre-processed at the farm which reduces the transport volume and costs. Ideas for this alternative approach are developed by the chair group BCH (Biobased Commodity Chemicals). In 2011 BRD and BCH together developed a model for the processing system on the farm. The processing system on the farm includes the alternative sugar processing system, anaerobic fermentation for biogas generation, recovery of energy and by--products and local energy generation. While setting-up the model we found a number of limitations in the process system. We worked on a system redesign with emphasis on the performance and technical feasibility of the sugar processing unit. Significant improvements have been proposed. Moreover, an analysis is made on the reduction of transportation costs for central vs decentral processing. It was shown that the transportation costs can be reduced till only 20-30% of the current level. With this result the perspective for decentral processing increases. Optimizing membrane filtration of food products The WUR research institute FBR (Food and Biobased Research) participates in the EU-project EU-cafe (computer aided food engineering). In 2011 BRD obtained participation in this project on optimal control. Our task in the project is to develop an optimal strategy for control membrane fouling that occurs during separation of food products. The beer membrane filtration process is represented by the block diagram in Fig. 1. Conventional control is characterized by the fact that the retentate flow (Fin) and permeate flow (Fout) are constant for each filtration phase (F). As a result also the cross flow (Fcross) is constant. For different filtration phases (F) these constant settings may be different after adaptation by operators.

Fig. 1: Block diagram of beer membrane filtration (BMF).

The objective of our part of the “EU Cafe” research project was to design, simulate and experimentally validate what may be gained from an adaptive optimal control approach as compared to conventional control of beer membrane filtration. Adaptive optimal control relies on a dynamic model of the beer membrane filtration process and an objective function reflecting quantitatively control objectives.

A first principles model was used. It enables computation of beer membrane filtration costs and furthermore provides valuable insight. Optimization and simulation of the full adaptive optimal scheduling and control system have been performed showing technical feasibility and over 20% reduction in costs. As to future research a major conclusion is that to benefit from adaptive optimal scheduling and control sensor development to measure fouling of beer at different locations is crucial. The simulation results obtained with adaptive optimal scheduling and control will be published in a control journal. This project finished in early 2013. b) Energy efficient adsorption drying of healthy food products (J.C. Atuonwu, X. Jin, A.J.B. van Boxtel)

The project “Energy efficient drying of healthy food products”, funded by AgentschapNL, concerns low temperature adsorption drying, i.e. in the range 10-50°C where conventional dryers have only an energy efficiency of 10-25%. The project focuses on drying of food products with components that have a positive effect on health. We look especially for the drying of cabbages like broccoli which contain glucosinolates. If these glucosinolates are well activated by the enzyme myrosinase during drying they can contribute to the mechanism to prevent colon cancer. In total 3 PhD-students worked on this project. The subprojects concern 1) kinetics of activation and deactivation of glucosinolates, 2) drying strategies to maintain bioactive glucosinolates, 3) dryer development and control. Subproject 2 and 3 are carried out by the chair group Biomass Refinery and Process Dynamics, subproject 1 by the chair group Product Design and Quality. The project will finish in 2014 with the final conclusions, the graduation of the 3 PhD-students and the reporting to AgentschapNL.

stalk

Floret

Φ

R

Broccoli photo and structure

Drying strategies to maintain bioactive glucosinolates (X. Jin) The drying model, developed in previous years, is tuned according to experimental results. In 2014 we worked on the development of a drying strategy and the experimental evaluation of that strategy. To combine retention of bioactive compounds in combination with high energy efficiency, optimal drying trajectories are calculated for the control variables air flow rate and temperature. For drying broccoli the considered bioactive compounds are vitamin C and glucosinolates. Optimal drying trajectories plotted in state diagrams with degradation and drying rates, explain the calculated drying strategies (Figure 1). The drying strategies allow for halving the energy consumption and a significant increase of vitamin C content.

Figure 1. Temperature trajectories of the inlet air temperature, product temperature, and air flow rate as a function of time (top), and trajectories presented in moisture-temperature state diagram (bottom)

In the last stage of this work, optimized drying trajectories are calculated for vitamin C, glucosinolates and myrosinase. Simulation and experimental results for optimized temperature trajectories are compared to constant inlet air temperature drying at 40 and 50°C. The results show that with the optimized temperature trajectories the retentions of GR, MYR and Vc improved significantly. Moreover, the experiments show that degradation and inactivation upon drying is slower than expected from kinetic studies. The deviation is explained from the difference in the physical state of the samples used in the drying experiments, i.e. original plant tissue, and the grounded state of the plant tissue used in the experiments for the kinetic studies. The results indicate that besides temperature and moisture content the physical state is also an important aspect in the degradation of nutritional components and enzymes. In an intervention study the dried broccoli was consumed by a number of test persons. It was observed that broccoli dried by the optimal drying strategy resulted in the excretion of a higher amount of bioactive compounds than standard dried broccoli. The PhD-thesis was successfully defended in 2013.

Dryer development and control (J.C. Atuonwu) In adsorption drying air is dehumidified using an adsorbent. Regeneration of the adsorbent can be done under favorable conditions which allow recovery of heat which is not possible under standard drying conditions. With the heat recovery, energy efficiency of low temperature drying is significantly enhanced. The concept of adsorption drying is under continuous development. Mathematical models of the adsorption drying system components, namely, the adsorber(A), regenerator (R), dryer and heat recovery devices (Fig. 1.) have been developed and linked to form an integrated system.

Figure1. Zeolite adsorption drying process with sensible and latent heat recovery

Controllability studies on the adsorption drying system (Fig. 1) show additional degrees of freedom for controlling moisture content and temperature compared to the conventional dryer. These include the regeneration air flowrate, temperature and the zeolite flow speed. These extra degrees of freedom allow for more flexibility of dryer control. A relationship is established between elements of the dryer process gain matrix (sensitivities of product moisture or drying rate and air temperature to air flowrate step changes) and energy efficiency. A mathematical modeling approach is developed to compute dryer energy efficiency using these elements of the dryer process gain matrix obtained from step responses, providing a link between energy efficiency and process controllability. The model is equivalent to the temperature drop between the dryer inlet and outlet air under adiabatic conditions and so decouples water evaporation from heat loss and product heating effects on dryer temperature drop. As such, the computation remains reasonably accurate even for dryers with significant heat losses where the traditional use of actual temperature measurements is grossly inaccurate (see Fig. 2).The approach is tested and verified on two experimental case studies involving significant heat losses: the first, a continuous fluidized-bed dryer and the second, a conventional and zeolite wheel-assisted batch dryer design. It is shown that the energy efficiency improvements achieved with the adsorbent system are linked to process gain alterations. This has controllability implications.

Fig. 2. Comparison of calculated energy efficiencies (actual, based on actual temperature drop measurements dT, and based on process gain-derived temperature drops) for a conventional batch dryer. The gain-based efficiency is far more

accurate than the dT-based

The PhD-thesis was successfully defended in 2013.

DryerHeater1

Wet product Dry product

Heater2

A

R

HX

HX

Dehumidified air

0 0.5 10

0.2

0.4

0.6

Time (h)

Ene

rgy

effi

cien

cy

Actual dT-based (in error) Gain-based

c) Advanced scenario studies for biodiesel production with micro-algae (P.M. Slegers, A.J.B. van Boxtel, G. van

Straten) Micro-algae have a high potential for biotechnology applications and are seen as an interesting source for biofuel production. In this project an advanced scenario study approach for the design of efficient, robust and flexible biodiesel production systems using algae will be developed. This approach is used to evaluate the interaction of different production systems and cultivation technology and to find bottlenecks in the applications. The project was started in 2009. In 2012 we started with a process library for the processing of algae to biodiesel. In addition, a systematic and flexible approach was applied to select the best processing route and the best process conditions for DSP of algae by maximisation of the Net Energy Return. The model-based combinatorial approach enables a bottleneck analysis for each process group. Figure 1 illustrates for several process routes the relative energy consumption of every processing step as a percentage of the total energy consumption. It clearly depends on the process route considered which processing steps are the most important to improve. Therefore, tailor-made improvements in process unit design should be made. The results illustrates the potential of this model-based approach for process design and can also be linked to other performance indicators such as economics, material inputs or greenhouse warming potential.

Figure 1. Contribution of process steps to total energy consumption (%) for several process groups. In 2013 BeWhere Algae was applied to analyse the algae cultivation supply logistics, this included: 1) assessment of how resource allocation influences the size and optimal locations of algae cultivation, and 2) analysis of the transport energy consumption and distances for water and CO2 supply. The optimisation results show that in the Benelux large algae cultivation areas should be planned at locations nearby the sea with a high CO2 availability. In southern France the plants are scattered over the region and for the Sahara cultivation is planned at locations with a minimised water transport. Given a realistic number of areas for algae cultivation, the energy consumed for transport of resources turns out to be a small percentage of the energy contained in the algae biomass. The transport energy consumption is the lowest in the Benelux due to good availability of water and CO2. In southern France and the Sahara water and CO2 have to be transported over significant larger distances of up to 178 km per supply point, but the proportion of transport energy is still low. In general, the share of transport energy is the lowest for photobioreactors and the highest for cultivation in raceway ponds, which is mainly attributed to the difference in evaporative losses.

0

20

40

60

80

100

Hexane-Acidic Hexane-Alkaline SC CO2-Acidic SC CO

2-Alkaline SC methanol Enzymatic Microwave

Co

ntr

ibu

tio

n to

to

tal e

nerg

y co

nsu

mp

tio

n (

%)

Harvesting

Dewatering

Disruption

Drying

Extraction

Conversion

Figure 2. Transport energy share (%) compared to the energy contained in biomass.

This project is performed in the cooperation framework of Wetsus, centre of excellence for sustainable water technology (www.wetsus.nl). d) Systems analysis of algae biorefinery (F. Fasaei, P.M. Slegers, A.J.B. van Boxtel)

Algae are considered as a potential sustainable alternative for fossil based fuels and chemicals. Cultivation of algae ends in a diluted solution with algae that contain lipids, carbohydrates and protein as main components. Processing these components to a series of selected end products request for a number of operations. The functional steps in algae biorefineries are harvesting, dewatering, disruption, extraction and conversion. Every functional step has several candidate unit operations, but they differ in their performance characteristics. The choice of a unit operation is not straightforward as each operation affects all next processes, and beneficial processes for one type of product can be disadvantageous for another type of product. Algae biorefineries are now in a phase of development A major challenge in algae biorefineries design is to propose process systems that can process algae with a high yield into a range of valuable products. Moreover, the biorefinery must also be sustainable. The objective of the project “systems analysis of algae biorefinery”, which is part of the TKI AlgaePARC Biofinery and the EU-funded project Miracles, is to develop a sustainable and economically efficient biorefinery. With this project we also want to make a step forward in the integration of sustainability (LCA) aspects in an early phase of process design. Another aspect of the project concerns the role of the logistics in planning algae biorefineries. Processing starts with huge volumes of algae broth and end with significant smaller volumes of end product. Transport of the huge volumes is very costly and a significant amount of the secondary stream can be reused. Main questions are: what is the contribution of logistic costs and energy consumption to the biorefinery?, where should algae biorefineries be located?, nearby the cultivation site or close to the end users?, is full processing at a site desirable or can parts of the refinery be shared with other algae producers?, what is the effect of storage on the quality of the products and processing conditions?, etc. These question are the second challenge in the project “systems analysis of algae Biorefinery”. This project started end 2013.

0.00

2.00

4.00

6.00

8.00

10.00

Tra

nsp

ort

en

erg

y sh

are

(%

)

Benelux

s. France

Sahara

P. tricornutum T. pseudonana

38.5 %

TKI-AlgaePARC Biorefinery

e) LCA-driven process design for milk-to-powder production (S.N. Moejes, A.J.B. van Boxtel)

Milk powder production by the dairy industry is an energy intensive operation. The current processing methods originate from the mid-20th century. These processing methods have been refined and optimized up to the current level. But, in the last decades innovative processing methodologies have been introduced, which can lead to breakthroughs in energy and water efficient processing. Examples are membrane technology, new atomizer systems, radio frequency heating, drying combined with adsorption systems, etc. The availability of these technologies and the potential of efficient combinations of the technologies ask for a reconsideration of the chain. The EU-funded ENTHALPY project concerns a reconsideration of the milk-to-powder chain. The work in this project concerns three different levels 1) the prediction of the performance of the innovative technologies for milk powder production, 2) to define the operational conditions of the technologies in connection to each other, 3) to define a processing chain with a low environmental impact. Sustainability can be expressed by LCA-methods. In the ENTHALPY project we search by using process systems engineering (PSE) methods for a new design of the milk-to-powder chain that is both economic attractive and which has at the same time a high sustainability performance. Moreover, with this project we work towards an early integration of LCA-criteria in process design. This project started end 2013. f) Processing at farms (R.J.C. van Ooteghem, L.G. van Willigenburg, A.J.B. van Boxtel)

Concentrated processing at large sites has the benefit of economy of scale. However, processing in rural areas has the advantage of lots of low cost space which can be used for energy generation and processing. If the energy and products are used at the farm or in the region, the costs of distribution remain low and significant benefits can be obtained. In 2013 we studied a number of cases for local processing at farms:

1. Production of hydrogen as a transport fuel using solar energy and water gained at a farm 2. Production of ammonia for fuel and fertilizer from solar and wind energy 3. Manure processing systems

In all these cases models for energy generation, the conversion and transport were developed. The models were used for year round simulations using weather data to predict performance of the systems and to search for the bottlenecks in the year production. Moreover, the systems were optimized. These studies showed, as expected, that the seasonal variations are the main limiting factors in the production. The occurrence of specific harvesting moments requests for intensive processing over a short time span. The challenge we see is to create, to manage storage systems, and to have smart small scale processing that can preprocess the resources.

Schematic overview of production of hydrogen at a farm for clean energy

MATHEMATICS, COMPUTERS AND SIMULATION IN BIO-BASED AND ENVIRONMENTAL

SCIENCES

Description Recently, the Dutch government has indicated water, energy and agrifood as three, out of nine, top sectors. Hence, water and the water-energy-material (WEM) nexus in a bio-based/circular economy are interesting subjects for research that ask for innovative solutions. In the past decade, there have been strong developments in communication (internet, e-mail, smart phones, wireless), computation (Moore’s exponential law - processors, memory, storage capacity), sensor networks (data warehouses) and cyber science (integration of knowledge). These developments allow the implementation of smart, high-tech, cost-effective solutions to water/energy/material related problems in a bio-based/circular economy. Our objective is to use systems theory and information & communication technologies for data acquisition, data analysis, modelling and decision support, to understand, manage and design bio-based and environmental systems for efficient use of energy, water and materials for many different purposes. In particular, the focus will be on biological nutrient cycles, material recovery, active storage of food/feed/bio-based materials, and energy-water-material balances in an urban environment. This project also serves to initiate possible new PhD projects. Project leader: K.J. Keesman. Subprojects and participants: a) Acoustic waves for water disinfection, filtration and purification (H.J. Cappon, K.J. Keesman) b) Development of a new bioreactor for a process for biological sulphide oxidation (J.B.M. Klok, A. Janssen, K.J. Keesman) c) Modelling and optimization of nitrogen (N) transformation in a fertilised flooded rice paddy field for loss reduction (N. Khairudin, M.P. Zakaria, P. Struik, K.J. Keesman) d) Computer–aided design and monitoring of WWTPs towards energy and nutrient recovery (R. Khiewwijit, H. Temmink,

H. Rijnaarts, K.J. Keesman) e) Real-time accounting and simulation of dynamic energy-water-material balances for achieving liveable and sustainable cities of the future (E. Bozileva, I. Leusbrock, H. Rijnaarts, K.J. Keesman) f) Model-based innovative water management for resource efficiency in integrated multitrophic aquaculture and horticulture systems (INAPRO) (PhD student, H. Cappon, K.J. Keesman) g) Intelligent sensor network and system technologies for fish farming (Sensorfish) (G. van Straten, K.J. Keesman)

Ongoing work a) Acoustic waves for water disinfection, filtration and purification (H.J. Cappon, K.J. Keesman)

The objective for acoustic purification research is to develop a laboratory scale acoustic separation unit, which is capable of recovering suspended solids, particles from 10 up to 500 mu, from polluted water by means of ultrasound. The work incorporates a numerical-experimental approach, using both computer simulations and experimental testing and analysis. It consists of the following steps:

1. Modelling of the separator using Matlab and Comsol software 2. Experimental demonstration of separation principles in air and water 3. Scaling and prototype construction 4. Determination of filtration efficiency 5. Control unit for optimal separation strategy

In 2013 steps 3 and 4 were the main areas of focus. Papers and thesis

A short paper on the (numerical) basic design principles for acoustic separators was submitted and presented in July at the IEEE UFFC International Ultrasonics Symposium in Prague. A second and third paper on the final industrial separator design and its evaluation were submitted to Elsevier Separation and Purification Technology in August and November. The PhD thesis was finalized in November and will be defended on March 5th 2014 in Wageningen.

System modeling and construction

Simulating separator performance incorporates a piezoelectric + acoustic simulation, a flow simulation and a particle path simulation using the results of the first two. Hundreds of simulations were run to determine the best dimensions for the acoustics of a separation unit. In the end the best performing system had a resonance chamber of 70x20x28.5 mm (l * w * h), three flow lanes and two opposite transducers, as shown in Figure 1. The criteria used were the absolute average acoustic pressure (> 0.5 MPa) and the lowest possible flux per channel at various flow rates.

Figure 1 – Simulated flow pattern inside the new separator (left) and prototype system (right) Flow simulations showed that this system was able to trap all particles up to a flow of 5 mL/s (18 L/h) using about 30W of power for 10 µm particles with 1100 kg/m3 density. It was found that the optimal frequency is not the main eigenfrequency of the system, but rather a frequency which generates enough acoustic pressure without applying large currents. A sensitivity study on this final design shows that the eigenfrequency and the best operating frequency of the system are very sensitive to the main dimensions, but that the minimum power consumption for particle trapping (resulting in 0.5 MPa average absolute pressure) is rather constant, between 22 and 35W. A prototype separator was constructed on the basis of the dimensions found with the following materials: - A body made of borosilicate glass with PVC flow dividers - 4 Noliac NCE41 transducers (2x2 pairs) - 4 in/outlets ABS (3D printed) With this system the flow characteristics were tested and whether the system could be used to retain particles. The separation efficiency was determined at three flow rates ranging from 1 to 3 ml/s, using a stock suspension of insoluble potato starch of 1 g/l (1000 ppm). Concentrations of stock, filtrate and concentrate were measured using a turbidity meter and significant effects of acoustic particle concentration were measured at both outlets of the process. The maximum filtration efficiency and concentration efficiency were 54% and 76%, respectively. The performance found

was lower than the 100% that was expected for 10 μm particles from the model based design study. The deviation in

performance is mainly a result of (i) the pulsation of the feed pump, (ii) differences between the model and the actual

prototype, (iii) the limited power supply of only 10 W used and (iv) (too) small particles, below 10 µm, occurring in the

starch suspension. b) Full scale dynamics of biological sulphide oxidation (J.B.M. Klok, K.J. Keesman, A. Janssen (ETE)) General General description In biological gas desulphurization, H2S-containing gas is absorbed into an alkaline solution. In the sulphur producing bioreactor, the dissolved sulphide is subsequently oxidized into elemental sulphur under oxygen-limited conditions whilst a part is oxidized to sulphate by a mixed population of sulphide oxidizing bacteria. In addition, non- biological sulphide oxidation occurs, merely leading to the formation of thiosulfate. Formation of (thio)sulphate in the sulphur producing bioreactor is unwanted, because it reduces the overall process efficiency and leads to an increased chemical consumption of e.g. ethanol or H2-gas. Progress 2013 Models have been developed which describe full-scale reactors. These models have been successfully validated with full scale reactor results. Furthermore, the developed models are able to identify operational upsets. Hence, these models can be used in a model-based control strategy.

Source: Albert J.H. Janssen, Piet N.L. Lens, Alfons J.M. Stams, Caroline M. Plugge, Dimitri Y. Sorokin, Gerard Muyzer, Henk Dijkman, Erik Van Zessen, Peter Luimes, Cees J.N. Buisman, Application of bacteria involved in the biological sulfur cycle for paper mill effluent purification, Science of the total environment 407 (2009) 1333 – 1343

Project common with Environmental Technology (WU) c) Modelling and optimization of Nitrogen (N) transformation in a fertilised flooded rice paddy field for loss

reduction (N. Khairudin, K.J. Keesman and M.P. Zakaria)

In Malaysian rice fields, the soil is fertilized with urea between three and five times for each cropping season. However, not all of the applied N fertilizer will be taken up by the rice crop. Our main objective is to model and to estimate N transformations in a flooded rice field for scientific understanding of the key processes and for scenario analyses. Progress in 2013 Experimental work From January to April 2013, soil and floodwater samples were collected in the experimental plots which were located in Block C of Sawah Sempadan, Tanjung Karang, Malaysia. These samples were analysed for ammonium (NH4

+). From this experiment, we concluded that monitoring of NH4

+ in the soil and the soil solution is not informative of the N transformation rates and the N availability for rice uptake when N is broadcasted at less than 35 kg N/ha and without subsequent incorporation into the soil. Based on this study, we infer that the N loss via ammonia (NH3) volatilisation and leaching is not significant at rice mid-growth stage under the prevailing agronomic and management practices in Sawah Sempadan. Nevertheless, N loss may also occur via surface runoff due to heavy rainfall after N application. Modelling work In flooded rice fields, NH3 volatilisation ranged between 10 and 60 % of the applied N (Fillery et al. 1984). We adapted the Jayaweera-Mikkelsen model (Jayaweera and Mikkelsen 1990) and made four major modifications to simulate the NH3 volatilisation in flooded rice fields. First, to simulate the two-hourly floodwater pH, we integrated a part of the floodwater routine from CERES-Rice (Godwin and Singh 1998) into the Jayaweera-Mikkelsen model. Second, instead of assuming an NH3 concentration of zero at steady state, we conceptualised the pH-and-temperature-dependent-ratio of NH3 and NH4

+ in the model. The equation is adopted from Jayaweera and Mikkelsen (1990). Third, we included the N uptake by rice in the model by conceptualising the uptake directly from the floodwater. This is based on the study by Safeena et al. (1999), who demonstrated that the rice is able to directly absorb N from the floodwater when the roots are well developed. The rice N demand is conceptualised to follow a sigmoidal curve. Fourth, the NH3 volatilisation from the floodwater is described by first order kinetics and is assumed to be independent of the wind speed. The simulated NH3

volatilisation, using estimates of the unknown N-gift and parameter values from literature, is compared with secondary data obtained from de Tian et al. (Tian et al. 2001)

Currently, the application of the modified Jayaweera-Mikkelsen model is limited to rice early growth stage and under continuous flooding conditions. Thus, future work in 2014 includes extending this model to simulate the mid-season NH3 volatilisation and under intermittent flooding conditions. References Fillery IRP, Simpson JR, de Datta SK (1984) Influence of field environment and fertilizer management on ammonia loss

from flooded rice. Soil Sci Soc Am J 48:914-920. doi: 10.2136/sssaj1984.03615995004800040043x Godwin DC, Singh U (1998) Nitrogen balance and crop response to nitrogen in upland and lowland cropping systems.

In: Tsuji GY, Hoogenboom G, Thornton PK (eds) Understanding Options for Agricultural Production, vol 7. Kluwer Academic Publishers, Dordrecht, pp 55-77

Jayaweera GR, Mikkelsen DS (1990) Ammonia volatilisation from flooded soil systems: A computer model. I. Theoretical aspects. Soil Sci Soc Am J 54:1447-1455

Safeena AN, Wahid PA, Balachandran PV, Sachdev MS (1999) Absorption of molecular urea by rice under flooded and non-flooded soil conditions. Plant and Soil 208 (2):161-166. doi:10.1023/A:1004439503573

Tian GM, Cai Z, Cao J, Li X (2001) Factors effecting ammonia volatilisation from a rice-wheat rotation system. Chemosphere 42:123-129

d) Computer – aided design and monitoring of WWTPs towards energy and nutrient recovery

(R. Khiewwijit, K.J. Keesman, H. Temmink) General description Under the pressure of rising energy prices, additional treatment steps as anaerobic digestion, have been added to provide valuable energy. However, to make wastewater treatment plants (WWTPs) energy positive further research is still needed. In addition to making the WWTPs energy positive, the WWTPs could also be designed and controlled such that nutrients recovery such as nitrogen (N) and phosphorus (P), reduction of CO2 emissions and reuse of treated wastewater are realized, see Figure 1.

Figure 1: Novel process design of municipal wastewater treatment plants.

WWTP of the future

Municipal wastewater

N N

N

P

P

P Energy positive N-P recovery Reduction of CO2

emissions

Re-usable water

The challenge of this study is to (re)design, monitor and control existing and novel WWTPs for positive energy, cost-effective nutrients recovery, reduction of CO2 emission, and good quality of the treated water, preferable suitable for reuse. In order to achieve this challenge, the following sub-objectives have been defined:

1. Identify bottlenecks in conventional WWTPs design 2. Design novel WWTPs layouts, using the latest insights in domestic wastewater treatment 3. Design monitoring and control strategies for the realization for the novel WWTPs layouts 4. Validate the novel designs by lab-scale and pilot-scale experiments.

Progress 2013 In 2013, we focused on design monitoring and control strategies of novel design to derive rules of thumb for implementation in practice, together with validate the novel design by lab-scale experiments. Four new scenarios were proposed for Dutch municipal WWTPs of the future. Processes integrated in our optimal scenario consists of bioflocculation as a pre-treatment step to concentrate the COD, cold partial nitritation/Anammox to remove nitrogen, novel phosphorus treatment to remove and recover phosphate, and anaerobic sludge digestion with combined heat and power (CHP) to generate the electricity and heat from biogas production, see Figure 2.

Figure 2: Flow chart of optimal scenario for Dutch municipal WWTPs.

The optimal scenario has proved to be more efficient than the conventional activated sludge (CAS) systems in term of the energy production and consumption: the energy production increases from 0.25 to 0.48 kWh/m3, and the energy consumption reduces from 0.42 to 0.28 kWh/m3. Furthermore, 81% of the phosphate is recovered, and CO2 emissions are reduced by 38% compared to the CAS systems. But, further investigation on each process is still necessary. Another point of our study is sensitivity analysis focused on temperature and wastewater characteristics, as this will offer insight in the development of future municipal WWTPs in other countries and regions. To validate our optimal scenario, lab-scale experiments were conducted to maximize the COD removal and to improve the value of municipal sludge waste other than methane production as in Figure 2. A membrane bioreactor (MBR) process is used to concentrate the COD, and separate it from the water effluent. The MBR sludge is subsequently fed to an anaerobic fermenter for the production of volatile fatty acids (VFAs). In the last step, an extraction of VFAs is needed to optimize the production yield and avoid product inhibition. With our innovative recovery technology, we expect to maximize the recovery of acids production, as well as to re-use alkalinity product. Project in collaboration with Environmental Technology (WU) e) Real-time accounting and simulation of dynamic energy-water-material balances for achieving liveable

and sustainable cities of the future (E. Bozileva, I. Leusbrock, H. Rijnaarts, K.J. Keesman)

General description Currently, human activity is the main driver of global environmental change. We have reached this stage through various simultaneous exponential increases in population, energy use, economic activity, emissions, and more. Consequently, this imposes unprecedented stresses on ecosystems, resources, political systems, and economies. Urbanization and rising living standards are at the core of these global changes. Within the Climate-KIC project we propose and develop a novel computer-aided method for serious gaming that integrates the Urban Harvest Approach (UHA - Agudelo, 2012, PhD thesis) and very large amounts of real-time data (Big Data). Big Data from multi-spectral satellites with thermal-infrared sensors and smart metering systems in the water, electricity, gas and food stores grid will be used to develop a (real-time) assessment and management tool for cities of the future. This serious gaming tool will enable us to support the transition from an inefficient linear urban resource use and waste production towards a sustainable (circular) urban metabolism with respect to energy, and recycling and reuse of water and (biodegradable) material. The proposed UHA - Big Data approach combines knowledge of water, energy and (biodegradable)material cycles and the involved production, recovery and treatment technologies with tools from systems theory, such as constrained dynamic. Climate-KIC project: Elvira Bozileva (PhD) Project in collaboration with Environmental Technology (WU): Ingo Leusbrock, Huub Rijnaarts

f) Model-based innovative water management for resource efficiency in integrated multitrophic aquaculture

and horticulture systems (INAPRO) (PhD student, H. Cappon, K.J. Keesman)

General description The collaborative project INAPRO challenges the concept of recent water, energy and nutrient management solutions in rural areas (with aquacultural and horticultural production) and urban areas (urban farming) to cope with global demands and to exploit all available opportunities of resource efficiency. The objective of this project is to mobilise industry, member states and stakeholders into promoting a new and innovative technical and technological approach for an Aquaponic system. Coupling the production of aquatic animals (e.g. fish) and plants (e.g. vegetables) in greenhouses forms the basis for all aquaponic systems where the wastewater from the aquaculture section is used for the nutrition of the plants and this way saves water compared to single greenhouse or single fish production systems (Fig 1).

Fig. 1 INAPRO system concept g) Intelligent sensor network and system technologies for fish farming (Sensorfish) (G. van Straten, K.J.

Keesman) General description Fish farming and aquaculture is the world’s fastest growing sector within the food industry, providing nearly 50 percent of fish consumed globally and creating more than 5 million jobs worldwide. Aquaculture is of growing importance to the European Community given the mere fact that the industry employs the local population in many rural areas of several countries in Europe. Especially the European fisheries and aquaculture must enhance their competitiveness while maintaining the sustainable production of aquaculture and marine resources. The only possible answer is to use highly advanced technologies for fish farm management; to create a competitive, sustainable and multifunctional fisheries and aquaculture systems and to optimize the quality of products along with fish health and welfare. The advance of climate change, leading to warmer water temperatures, will make the monitoring of the state of the water environment even more important. For cold-water species, such as Atlantic salmon, warmer water means that more energy is consumed for maintaining basic life functions; hence less energy is available for growth and coping with stress (i.e. farming operations, diseases, poor water environment etc). A continuous and reliable system for monitoring the aquatic environment so that the farmer can make informed decisions is therefore essential for the future growth of the industry. Very similar challenges arise in highly stocked ponds or raceways. Recirculation systems conditions are generally easier to control but their success is closely related to the reliable function of monitoring equipment. This underlines the need for continuous and precise monitoring of the aquatic environment, fish behaviour and the functionality of technical equipment, particularly in highly stocked fish farming facilities. The overall aim of this programme is networking and increased collaboration between groups in Europe, China, USA and Canada working on intelligent sensor network and system technologies for fish farming. Our contribution to the scientific objectives are:

1. To design smart monitoring strategies using a minimum of hardware sensors and using soft-sensors to estimate unknown rates and concentration in the recirculation system for insight into the dynamics and for advanced (model-based) control. 2. To design remote, advanced (model-based) control strategies for safe, energy efficient and economically optimal fish farming. 3. To deepen the study of real-world system integration for fish farming. For the realization of this an intelligent monitoring and control pilot system for fish farming will be realized and tested in China and Norway. The participation of the European and American institutions will support the laboratory experiments related to mixing processes. Progress 2013 The second management meeting took place in Ilmenau from 27-28 Feb 2013. Here the decision was made to submit an EU-FP7 WATER-INNO-DEMO project on aquaponic systems, called INAPRO, which was approved in June.

CLIMATE AND NUTRIENT CONTROL IN GREENHOUSE PRODUCTION

Description The control of indoor climate in agricultural buildings is usually done by heuristic control designs. The objective of the project is the development of advanced, intelligent control algorithms for greenhouses and other agricultural buildings in order to achieve better compliance with ever increasing requirements of reliability, safety, user flexibility, energy demand, product quality, environmental protection and economy. This project also serves to initiate possible new PhD projects. Project leader: G. van Straten. Projects and participants:

a) Optimal Management of Energy Resources in greenhouses – OMER (P.J.M. van Beveren, J. Bontsema (PRI), E.J. van Henten (FTE), G. van Straten) The idea of this project is to compute control trajectories that minimize total energy input to a greenhouse while maintaining the climate within bounds set by the grower. This leaves the grower in control over crop aspects of the cultivation. Models for temperature, relative humidity, and CO2, including crop evapotranspiration were adapted to the new controls in modern greenhouses, and were validated against almost year-round climate data in a rose cultivation, with very good results. Next, the models were used to compute cooling or heating trajectories throughout the year that minimize total energy input, while exploiting the possibilities of free natural ventilation. The results look promising, but ultimate conclusions will be drawn when also CO2 is included. In 2014 the project will focus on the most economical way to realize the optimal trajectories, and to study uncertainties and on-line aspects.

Ongoing work a) Optimal Management of Energy Resources in greenhouses – OMER (P.J.M. van Beveren, J. Bontsema, E.J.

van Henten (FTE), G. van Straten) General description This STW project is part of the Smart Energy Systems programme of STW. In the interest of reducing energy consumption and CO2 emission while maintaining productivity, greenhouse growers have installed a wide range of auxiliary equipment, such as assimilate lighting, heat exchangers, short and long term heat buffers, co-generation units and heat pumps. The task of the grower to manage his equipment in the most economical way has become very complicated, not only because of the complex intrinsic physics of the greenhouse dynamics, but also because of the uncertainty in the weather, and, in addition, the strongly fluctuating market prices of energy. The aim of this project is to develop optimal scheduling or management procedures. Scientifically, this can be seen as a non-linear dynamic optimization problem with uncertainty in external prices and external influences. While under assumed prices and future weather the problem can – theoretically – be solved off-line by known numerical methods, the real challenge is to obtain feasible on-line optimal or near optimal operation strategies. The project started at the end of 2011, with the appointment of Peter van Beveren MSc as post graduate student. His provisioned promoters are Prof. Eldert van Henten and Prof. em. Gerrit van Straten, who have been responsible within the university for the proposal. The work is supervised primarily by Jan Bontsema of Wageningen UR Greenhouse Horticulture (former SCO staff member), and is conducted in close cooperation with Hortimax (dr. Ad de Koning), Lek/Habo, and an advanced grower.

Progress 2013

The main challenge of this project is to formulate and solve this optimization problem and to develop optimal scheduling or management procedures. The optimization problem is split into two parts; first the total energy input to the greenhouse is minimized, and secondly the energy scheduling will be optimized. In order to minimize the total energy input to the greenhouse, a dynamic climate model for greenhouse air temperature, humidity, and CO2 concentration was developed and validated with measurement data. Optimization (minimizing the total energy input to the greenhouse) is done with constraints on the temperature, humidity and CO2 concentration.

Project in collaboration with Farm Technology (WU) and Wageningen UR Greenhouse Horticulture

EDUCATION

PhD courses

a) The Art of Modelling (4 ects)

Basic and more advanced elements of systems theory are offered to PhD students in Wageningen and from elsewhere in a PhD course named ‘The Art of Modelling’, organized under auspices of the Research School Production Ecology & Resource Conservation (PE&RC). This is an updated version of earlier similar courses given in previous years with the name ‘Modelling Technology and Systems Engineering’. The course is given in a block of two weeks. Most participants have an interest in modelling, but have little experience with modelling techniques and tools for analysis of dynamic systems. Therefore, the lectures are accompanied by hands-on computer exercises, using FSE, Matlab and Simulink, and by self-study hours. A particular successful part of the course is the final day, where participants are challenged to prepare a brief essay in PowerPoint format on how they think they can use the taught material in their own work. This course was cancelled in 2012 and will be renewed in the coming years. b) Uncertainty Analysis (2/3/4 ects)

Models are deliberate approximations of the real world and can be seen as a concise and useful representation of scientific knowledge. They serve as tools in research, in the design and control of systems, and in planning and management. Uncertainty is an important issue in all these applications. This course aims at presenting the state-of-the art in quantitative uncertainty analysis, in which much progress has been made in recent years. The course treats the problem from a systems point of view. The first part (three sessions) refreshes participants' knowledge on systems modelling and dynamics. In addition, basic notions from statistics as well as matrix calculus are dealt with and there are practical exercises to be solved within MATLAB. In the second part (three sessions) the focus is on analysis, basically using analytical and Monte Carlo-based uncertainty propagation techniques. The third part (two sessions) deals with parameter estimation with corresponding estimation uncertainty of non-linear systems. Special attention is given to non-linear least-squares techniques and Monte Carlo Markov Chain (MCMC) methods. The fourth and final part focuses on the analysis and estimation of distributed systems and is devoted to uncertainty propagation in spatial systems and the implications of an uncertainty analysis. Here, special attention is given to the effect of spatial correlation on the results of the uncertainty propagation analysis and on the interpretation of the results. The theory is supported by examples, case studies and practical exercises. This course will be given again in 2014.

Overview of number of MSc-BSc-theses over the past 10 years

The figure shows the origin of students in MSc and BSc-projects of the Biomass Refinery and Process Dynamics group (formerly Systems and Control) over the past 10 years, where AT (Agrotechnology), BT (Biotechnology) FT (Foodtechnology) are the most relevant studies. Here ‘large’ indicates a number of ECTS over 30, and small under 30.

BSc and MSc courses (Biomass Refinery and Process Dynamics - BRD) and contributions to courses

(other codes)

code course title

BRD-20306 Modelling Dynamic Systems

BRD-21306 Control Engineering

BRD-22306 Sensor Technology

BRD-22803 Physical Transport Phenomena

BRD-23306 Biorefinery

BRD-30806 Physical Modelling

BRD-31306 Systems and Control Theory

BRD-31806 Parameter Estimation and Model Structure Identification

BRD-32306 Advanced Biorefinery

FTE-12303 Introduction Biosystems Engineering part 1

FTE-12803 Introduction Biosystems Engineering part 2

FTE-13807 Engineering 2

FTE-32806 Automation for Bioproduction

FTE-33306 Advanced Biosystems Engineering

BPE-10305 Process Engineering Basics

BPE-33303 Quality by Design for Biopharmaceuticals

BPE-60312 Bioprocess Design

MAT-24803 Mathematics in Biotechnological Systems

XWT-21305 Process Dynamics (Wetsus Academy)

BRD: Biomass Refinery & Process Dynamics

FTE: Farm Technology BPE: Bioprocess Engineering MAT: Mathematics

0

2

4

6

8

10

12

14

16

18

2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

Nu

mb

er

of

stu

de

nts

Year

other largeother smallFT largeFT smallBT largeBT smallAT largeAT small

MSc theses

Bozileva, E. Novel designs for energy and nutrient recovery in municipal wastewater using a dynamic modelling approach. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Stefanov, M.S. Uncertainty and senstivity analysis of models for single panel algae photobioreactors. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Ou, Xiang, Distribution station planning on algae cultivation site. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Grubben, N.L.M., Agricultural storage concepts and modelling. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Olst, R. van, Continuous optimization trajectories for energy efficient drying of broccoli with high levels of nutritional components. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Prevoo, Y.K.L.M., Adative dual control for industrial processes. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Hassan, A., Deriving rules of thumb for the control of a novel WWTP using a dynamical model. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013.

BSc theses Beus, N. de Ultrasound bioreactor design and testing. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Jol, A. Performer MK2 robotarmen in LabVIEW. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Visser, A.C. de Local sustainable hydrogen production. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Hoogeveen, B. Efficient manure processing and distribution. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Haanstra, L.J. Greenfertilizer towards producing anhydrous ammonia on the farm. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Mol, J. The behaviour of diffuse light in vertical plate reactors for algae cultivation. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Rijn, R. van Modelling the pre-treatment of lignocellulose. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013. Saglibene, M.J. The canyon effect: decay of diffuse light between vertical plates. Afstudeervak Biomanss Refinery and Process Dynamics, WUR, 2013.

PUBLICATIONS 2013

PhD Theses Atuonwu, J.C. (2013, maart 01). Energy-efficient low-temperature drying using adsorbents: a Process Systems Engineering approach. WUR Wageningen UR (XII, 183 pag.) ([S.l.: s.n.]). Prom./coprom.: Prof. Emeritus dr. ir. G. van Straten & dr. ir. A.J.B. van Boxtel. Jin, X. (2013, October 30). Drying of healthy foods from mechanism to optimization. WUR Wageningen UR (168 pg.) ([S.l.: s.n.]). Prom./coprom.:Prof. Emeritus dr.ir. G. van Straten, Prof. dr. R.M. Boom, dr. A.J.B. van Boxtel, dr. R.G.M. van der Sman.

International Journal Papers

Agudelo Vera, C.M., Keesman, K.J., Mels, A.R. & Rijnaarts, H. (2013). Evaluating the potential of improving residential water balance at building scale. Water Research, 47(20), 7287-7299. Atuonwu, J.C., Straten, G. van, Deventer, H.C. van & Boxtel, A.J.B. van (2013). Synergistic process design: Reducing drying energy consumption by optimal adsorbent selection. Industrial & Engineering Chemistry Research, 52(18), 6201-6210.

Atuonwu, J.C., Straten, G. van, Deventer, H.C. van & Boxtel, A.J.B. van (2013). Improving dryer energy efficiency and controllability simultaneously by process modification. Computers and Chemical Engineering, 59, 138-144 Boelee, N.C., Janssen, M., Temmink, H., Taparaviciute, L., Khiewwijit, R., Janoska, A., Buisman, C.J.N. & Wijffels, R.H. (2013). The effect of harvesting on biomass production and nutrient removal in phototrophic biofilm reactors for effluent polishing : Online First. Journal of Applied Phycology, 1-14. Cappon, H. & Keesman, K.J. (2012). Numerical Modeling of Piezoelectric Transducers Using Physical Parameters. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 59(5), 1023-1032. Cappon, H.J. & Keesman, K.J. (2013). Numerical modeling, calibration, and validation of an ultrasonic separator. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 60(3), 614-621. Croese, E., Keesman, K.J., Widjaja-Greefkes, H.C.A., Geelhoed, J.S., Plugge, C.M., Sleutels, T.H.J.A., Stams, A.J.M. & Euverink, G.J.W. (2013). Relating MEC population dynamics to anode performance from DGGE and electrical data. Systematic and Applied Microbiology, 36(6), 408-416. Jin, X., Sman, R.G.M. van der, Maanen, J.F.C. van, Deventer, H.C. van, Straten, G. van, Boom, R.M. & Boxtel, A.J.B. van (2013). Moisture Sorption Isotherms of Broccoli Interpreted with the Flory-Huggins Free Volume Theory (Online first). Food Biophysics. Jin, X., Sman, R.G.M. van der, Straten, G. van, Boom, R.M. & Boxtel, A.J.B. van (2013) Energy efficient drying strategies to retain nutritional components in broccoli (Brassica oleracea var. Italica) Journal of Food Engineering 123, 172-178. Keesman, K.J. & Maksimov, V.I. (2013). Reconstruction of unknown characteristics in a third-order system. Computational Mathematics and Modeling, 24(2), 271-278. Keesman, K.J., Norton, J.P., Croke, B., Newham, L. & Jakeman, A.J. (2013). Set-membership approach for identification of parameter and prediction uncertainty in power-law relationships: The case of sediment yield. Environmental Modelling & Software, 40, 171-180. Michels, M.H.A., Slegers, P.M., Vermue, M.H. & Wijffels, R.H. (2013). Effect of biomass concentration on the productivity of Tetraselmis suecica in a pilot-scale tubular photobioreactor using natural sunlight (online first). Algal Research, 1-7. Porada, S., Borchardt, D., Oschatz, M., Bryjak, M., Atchison, J.S., Keesman, K.J., Kaskel, S., Biesheuvel, P.M. & presser, V. (2013). Direct prediction of the desalination performance of porous carbon electrodes for capacitive deionization. Energy & Environmental Science, 6, 3700-3712. Slegers, P.M., Beveren, P.J.M. van, Wijffels, R.H., Straten, G. van & Boxtel, A.J.B. van (2013). Scenario analysis of large scale algae production in tubular photobioreactors. Applied energy, 105, 395-406. Slegers, P.M., Lösing, M.B., Wijffels, R.H., Straten, G. van & Boxtel, A.J.B. van (2013). Scenario evaluation of open pond microalgae production. Algal Research, 2(4), 358-368. Willigenburg, L.G. van & Koning, W.L. de (2014). Equivalent optimal compensation problem in the delta domain for systems with white stochastic parameters. International Journal of Systems Science, 45(3), 509-522.

Conference Papers

Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Optimal management of energy resources in greenhouse crop production systems. In proceedings 1st SES Programme Meeting, 5 June, 2013, NWO, The Hague. The Hague: NWO. Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Minimal heating and cooling in a modern rose greenhouse. In A. Visala (Ed.), 4th IFAC Conference on Modelling and Control in Agriculture, Horticulture and Post Harvest Industry, 27-30 August, 2013, Espoo, Finland (pp. 282-287). IFAC. Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Minimal Heating and cooling in a modern rose greenhouse. In Proceedings of the 32nd Benelux Meeting on Systems and Control (Book of Abstracts), 26-28 March 2013 (pp. 54). Vroegindeweij, B.A., Henten, E.J. van, Willigenburg, L.G. van & Groot Koerkamp, P.W.G. (2013). Modelling of spatial variation of floor eggs in an aviary house for laying hens. In J. Vandermeulen & D. Berckmans (Eds.), Precision Livestock Farming '13, 6th European Conference on Precision Livestock Farming, Leuven, Belgium, 10 - 12 September, 2013 (pp. 916-925).

Professional Publications, Abstracts, Presentations and Reports Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Optimal management of energy resources in greenhouse crop production systems. An Innovative Truth V - Congres over duurzame ICT & energie: Utrecht (2013, juni 19). Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Optimal management of energy resources in greenhouse crop production systems. Presentation at NWO, SES programme meeting 5 June, 2013: The Hague (2013, juni 05). Beveren, P.J.M. van, Bontsema, J., Straten, G. van & Henten, E. van (2013). Optimal management of energy resources in greenhouse crop production systems. SES Programme Meeting, 5 June, 2013, NWO, The Hague: The Hague. Bruins, M.E., Boxtel, A.J.B. van, Eppink, M.H.M. & Strik, D.P.B.T.B. (2013). Biorefinery in Wageningen. Slegers, P.M., Leduc, S., Straten, G. van, Wijffels, R.H. & Boxtel, A.J.B. van (2013, oktober 09). Algae cultivation logistics. Laxenburg, 1st International BeWhere workshop. Slegers, P.M., Leduc, S., Wijffels, R.H., Straten, G. van & Boxtel, A.J.B. van (2013, april 10). Small versus large scale microalgae production. Wageningen, 1st Biorefinery for food, fuel and materials Congres. Slegers, P.M. (2013, april 10). Small versus large scale microalgae production. Wageningen, 1st Biorefinery for food, fuel and materials Congres.

http://tinyurl.com/cbettnn