INTEROPERABILITY BETWEEN THE IMPROVED METAGRAPH HAD AND GRAFCET… · IJRRAS 9 (1) October 2011 Nkenlifack & al. Improved Metagraph HAD and Grafcet 47 transportation sensor transportation

IJRRAS 9 (1) ● October 2011 www.arpapress.com/Volumes/Vol9Issue1/IJRRAS_9_1_05.pdf

45

INTEROPERABILITY BETWEEN THE IMPROVED METAGRAPH HAD

AND GRAFCET: CASE OF THE POWER NETWORK OF CAMEROON

Marcellin Nkenlifack 1,*

, Emmanuel Tanyi 2

& Landry Domche 3

1 LAIA - IUT FV – University of Dschang, Cameroon

2 ACL – National Polytechnic – University of Yaounde 1, Cameroon

3 Bachelor, Electrical Engineering, IUT Fotso Victor – University of Dschang

Email: [email protected],

[email protected]

ABSTRACT

This article presents the modeling and the control of the Power Network of the Southern Cameroon (PNSC). We are

using the effectiveness of improved version of HAD, which allows an object modeling of hybrid and complex

dynamic systems. Then we show how to derive from the models obtained, the Grafcet models of the PNSC, by

implementing bridges between HAD and the reference language of Grafcet. The application to the Control and the

Monitoring of the Cameroonian electrical Power Network enables us to better determine the complexity of the

system, thanks to an approach structuring the various objects by taking in consideration the physical structure and

the causality of their interactions. On the other hand, interoperability with Grafcet would facilitate the

implementation and the synthesis of discrete or hybrid controllers.

Keywords: Grafcet, HAD, UML, Modeling, Control, Power network, Hybrid Dynamic Systems

1. INTRODUCTION

The development of the industrial systems caused some reflections on the physical, descriptive and representative

behavior of the industrial processes. Today, within sight of the studies led by some researchers [1] [2] [3] [4] [5] [6]

[7] [8] [9] [10] [11], it is important to note the difficulty to analyze the complex industrial systems correctly.

Several approaches of description of the systems exist [9]: the Discrete Systems (DS), the Continuous Systems (CS)

and the Hybrid Dynamic Systems (HDS) regrouping the two first.

HAD (Hybrid Activity Diagram) is the synthesis of a brewing of knowledge and the concepts of the Control

Engineering (CE) and the Software engineering (SE). The setting up of his formalism and his meta-model, is the

subject of numerous research [1] [4] [7] [12] [8]. His principle is at a time to model the Discrete part and the

Continuous part of a Hybrid Dynamic System (HDS), contrary to other tools like the Grafcet [23] that concentrates a

lot more on the discrete part. HAD has been enhanced lately by [2] [3], in his formalism and his application field. It

allows him of model more big and more complex dynamic systems. In this article, we show compatibility between

the improved version of the HAD meta-graph and the Grafcet, while considering the computerization of the

command of the Power Network of the Southern Cameroon (PNSC / RISC – Réseau Interconnecté Sud Cameroun).

This project is interesting, since as regards to research, the electric power networks are the subject of several jobs in

the scientific community [13] [14] [15] [16]. But most are about the modeling of the network, the measures, the

diagnosis and the analysis of the networks in general. But the hybrid modeling problem remained an open research

field.

In the structure of this article, we present the Electric Power Networks in section 2, the analysis of the hybrid

computer systems and the modeling of the PNSC by HAD in section 3. Once the HAD model constructs, we present

the transformations between the meta-graphs of HAD and the Grafcet, with an application to the case of the PNSC

in section 4. The article ends by the conclusion and perspectives in section 5.

2. DESCRIPTION OF THE POWER NETWORK

The Electric Power Network (PN) is a set constituted of the direct current or alternative generators, of the

transformers, the electric components, the equipments, the aerial and underground cables permitting to serve the

subscribers in the optimal technical-economic conditions.

The Cameroon power system consists of two separate networks – the northern and southern grids.

- The northern grid is supplied by one hydro-electric generating station (Lagdo) and a thermal plant. The southern

grid is supplied by two hydro-electric generating stations and four thermal power stations. Hydroelectric power

accounts for 77% of the total output of 1028MVA.

- The southern grid, which produces and distributes over 91% of the total energy, is considerably more extensive

than the northern grid. The rest being provided by thermal stations. The figure 1 presents the map of distribution of

mailto:[email protected]

mailto:[email protected]

IJRRAS 9 (1) ● October 2011 Nkenlifack & al. ● Improved Metagraph HAD and Grafcet

46

the two networks on the territory of Cameroon. The PNSC is characterized by different levels of voltage, presented

in table 1.

Oyomabang

Songloulou

Nkong-NjockEdéa

Mangombe

LogbabaBekoko

N'kongsamba

Bafoussam

Bamenda

Limbe

Dei

do

Bon

aber

iLagdo

Ngaoundéré

Garoua

Guider

Maroua

225kV/90kV Station

Distribution Station

Hydro-electric Station

90kV/30kV/15kV Station

Southern grid

Northern gridHydro-electric Station

110kV/15kV Station

110kV/15kV and 15kV/90kV Station

90kV/15kV Station

Key :

Figure 1. The Cameroon electric power system [17]

Table 1. Levels of voltage of the PNSC by types of Power Networks in Cameroon

Domains of

voltage

Abbreviations General level of

volltage (kW)

Level of voltage to

Cameroon (kW)

Type of Power

Network

Very High-

Voltage VHV 100 à 800 110 et 225

Transportation and

interconnection

High-Voltage – B

HV B 50 à 90 90 Distribution

Medium

Voltage HV to MV 1 à 40 15 et 30

Distribution

Medium Voltage (MV)

Low Voltage LV 0 ,1 à 0,44 0,22 à 0,41 Distribution

Low Voltage

The energy produced in the Hydro-electric Control Stations is transported toward big regions of consumption by

VHV networks. At the time of the passage of the current or at the time of the interruptions, the elements of the PN

are governed by physical phenomena, illustrated by mathematical equations. It is the dynamic part of the PN. The

set of the technical processes permitting to assure the continuous check of the PN (tele-driving or monitoring center)

include some elements as: power station of measure, sensors etc.

The Hybrid Dynamic System meets in the Command of a PN on the shutter of the power adaptation. The production

(input) must be adapted to the consumption (output). Thus, we must control the parameters (current, voltage,

power), what requires some sensors having to fetch the output parameters to compare them to the input parameters.

The Command of the PN is considered like a closed-loop system and his transfer function is characterized by the

product of the mathematical equations of every object of the system in order to translate the physical and dynamic

model of the system. It will be constituted of static and continuous properties of the system. This function expresses

itself by the equation (1), useful for the stability, the velocity and the precision of the Network.


47

sensortiontransporta

tiontransporta

production

nconsumptio

*1 (1)

The heart of the system is the power station of measure that to every instant calculates, check and informs then with

the help of his sensors and microcontrollers, on the balance between the production and the consumption, reacts on

the circuit breaker in case of unbalance, via Programmable Logic Controller. Thus, as in [16], the power station of

measure and his setting represent the Commands Part. The Operational Part being constituted of the circuit breakers

that in High Voltage (HV), undergo the destructive effect of the electric bow, mathematically modeled by the

difference equation [2]. The reader will be able to consult [18] and [19] for the establishment of the equation and his

resolution.

Lu

iL

Ri

ligne

arc

ligne

ligne

ligneligne te

dt

d

)(. (2)

3. ANALYSIS OF THE HYBRID SYSTEMS AND THE POWER NETWORK

3.1. Classic approaches of Description The modeling of the Hybrid Dynamic Systems can be performed with the help of various formalisms, among whom:

Hybrid State-charts, Mixed Petri Networks, Hybrid Machines, Colored Petri Networks, etc. For a description

retailed of these different formalisms, the reader will be able to consult [20] [11] [9] [10] [21]. One notes as [7] that

the presented above approaches don't provide the modularity and the upgradeability required for the representation

of complex systems. On the other hand, these tools have more developed for theoretical research and are not always

adapted to the precise representation of the physical or real systems. It is why we grant an interest to the contribution

of the new approaches inspired of the Software engineering and object-oriented methods, like HAD.

3.2. Fast presentation of HAD (Hybrid Activity Diagram) HAD meta-model is a tool adapted to the modeling of the HDS that articulates mainly around the following aspects:

- Modification of Activity Diagrams to incorporate discrete and continuous signals at the inputs and outputs

- Incorporation of structures for specifying causal relationships between components of a control system

- Convertibility between a UML specification of a hybrid model and a Grafcet of the system

With HAD, a module of activity or influence is constituted of a set of activity classes. Some of the components are

presented to the figure 2.

Figure 2. Sub-classes of "ActivityModule" [4]

The notion of "ActivityClass" that we present here constitutes an organ of causality (it undergoes some influences

and also exercise some influences). This aspect is illustrated to the figure 3.

Figure 3. Fundamental principle of HAD: causality [4]

For a better description of the HAD meta-model, the reader will be able to consult [3] [2] [4] [7] [12] [8].

3.3. Modeling of the Power Network We described all models schematically corresponding HAD to the command of the PNSC (about ten), that we won't

present in this article, due to a lack of space. On the other hand the reader will find all details of all these diagrams in

[19] and [18]. The system being very big, we present an excerpt of the HAD models rightly to the figures 4a and 4b.

On the figure 4b, the objects 15, 16, 17 and 18 show the calculations (the Grafcet doesn't make it) performed by the

power station of measure serving to the starting up of one or all reserves at a time, whereas the E2 object represents

a component having to influence the system of the outside: it is a switch of activation.

ActivityClass ActivityCause

ActivityEffect ActivityNoEffect

Cau

ses

Mechanisms Influences

exerted

on the entity

Influences

exerted on

other entities

Effect


48

SP

1

S 1

Ou

: ‘’ ok ‘’ >

0 >

IN : <

BE

<

S : <

BE

<

OP

: d 0

=0

; d 1 =

0; d

2 =0

; d ij =

1; R

0 =0

;R 1 =

0; R

2 = 0

OU

: ‘’ ok ‘’ >

1 >

IN : <

0 <

E 1

<

S : <

0 <

AN

D <

E 1

<

OP

: close

the

circuit b

reake

r d 0 ()

d 0 =

1 ; R

0 =1;

OU

: ‘’ ok ‘’ >

S 1 >

[ ]

;)

(.

0

L ui

L Ri

o

o

o

oa

rct

e

dt

do

0

ua

rcO

1 0

IN : <

1 <

S : <

1 <

or <

S 6 <

OU

: ‘’ ok ‘’ >

SP

1 >

IN : <

S1 <

S : <

S1 <

OP

: up

da

te c

on

su

mp

tion

()

OU

: "‘ok" >

T1 >

T1

E 1

Ou

: "pre

ss k

0

Turn

on

the

the

hyd

ro-e

lectric statio

n" >

1 >

[k0

= 1

, m0

= 1

]

IN : <

SP

1 <

S : <

SP

1 <

OU

: ‘’ ok ‘’ >

2,3

,4,5

,6,7

,8,9

,10

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

A 1

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

A 2

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

A 0

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

B 1

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

B 2

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

B 0

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

C 1

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

C 2

()OU

: ‘’ ok ‘’ >

T 2

>

IN : <

T 1

<

S : <

T 1

<

OP

: Evalu

ation

of

the

app

aren

t p

ow

er o

f load

C 0

()OU

: ‘’ ok ‘’ >

T 2

>

;22

.1

2P

DA

PD

AK

AC

A

;00

.1

0P

DA

PD

AK

AC

A

;11

.1

1P

DB

PD

BK

BC

B

;22

.1

2P

DB

PD

BK

BC

B

;00

.1

0P

DB

PD

BK

BC

B

;11

.1

1P

DC

PD

CK

CC

C

;22

.1

2P

DC

PD

CK

CC

C

;00

.1

0P

DC

PD

CK

CC

C

;11

.1

1P

DA

PD

AK

AC

A

T 2

w

IN : <

2 <

3 <

4 <

5 <

6 <

7 <

8 <

9 <

10

<

S : <

2 <

AN

D <

3 <

AN

D…

AN

D <

9 <

10

<

OU

: ‘’ ok ‘’ >

T 2 >

23

67

10

59

84

Loo

p

Ob

ject e

xecu

ting th

e su

b-p

rogram

Figure 4-a. Excerpt of the HAD diagram of the command system of the PN


49

w

T 4

IN : < T 3 <

S : < T 3 <

OP : Evaluation of the apparent power at node A ()

OU : ‘’ ok ‘’ >T 4 >

IN : < T 3 <

S : < T 3 <

OP : Evaluation of the apparent power at node B ()

OU : ‘’ ok ‘’ >T 4 >

IN : < T 3 <

S : < T 3 <

OP : Evaluation of the apparent power at node C

()

OU : ‘’ ok ‘’ > T 4 >

T 3

IN : < T 2 <

S : < T 2 <

OU : ‘’ ok ‘’ > 11,12,13>

;2

0

j

CjCS DT C

;2

0

j

BjCS DT B

;2

0

j

AjCS DT A

12 1311

IN : <11<12<13 <

S : < 11< AND <12< AND <13<

OP : Evaluation of demand in active power ()

OU : ‘’ ok ‘’ > 14 >

14

IN : < T 4 <

S : < T 4 <

OP : Comparison of active power produced and demanded ()

OU : ‘’ ok ‘’ > S 2 > [ Δ ]

;3

1

i

DTiSD

;DS

IN : < 14 <

S : < 14 <

OP : evaluation of Δ factor and selection of an action ()

OU : "ok" > S 6 > [Δ > 0 production by hydro-electric station sufficient] "ok" > T 5> [Δ < 0 production by hydro-electric station insufficient]

IN : < S 2 <

S : < S 2 <

OU : ‘’ ok ‘’ >15,16,17,18 >

S 2

w0

T 5

IN : < T 5 <

S : < T 5 <

OP : Add reserve 1,

Compare to demand ()

{ OU : ‘’ ok ‘’ > T/ S 1>

;11 DS

;11 SSS thH

;22 DS

;22 SSS thH

;

1212 SS thth

IN : < T 5 <

S : < T 5 <

OP : Add reserve 2,


{ OU : ‘’ ok ‘’ > T/ S 1>

IN : < T 5 <

S : < T 5 <

OP : Compare reserve 2 to

reserve 1 ()

OU : ‘’ ok ‘’ > T/ S 1>

IN : < T 5 <

S : < T 5 <

OP : Add reserve 1,


{ OU : ‘’ ok ‘’ > T/ S 1>

;213 SSSS ththH

;33 DS

T / S 1

15 16 17 18

IN : <15< 16< 17 <18 <

S : < 15 < AND <16 < AND < 17 < 18 <

OP : evaluation of Δ1, Δ2, Δ3, Δ12 factors and selection of an action ()

OU : ‘’ok‘’>19>[ reserve 1 sufficient ]

‘’ok‘’>20>[ reserve 2 sufficient ]

‘’ok‘’> T 6 >[ If not load shedding reserves 1 and 2 ]

);012

.02

.01

()02

.01

.(03

);012

.02

.01

()02

.01

.(03

.);03

()02

.01

.(03

IN : < T/S 1 < AND < E 2

S : < T/S 1 < AND< E 2

OP : Synchronisation of

alternator of reserve 1 ()

OU : ‘’ ok ‘’ >2 > [ t= 15 min]

19

IN : < T/S 1 < AND <E 2

S : < T/S 1 < AND< E 2

OP : Synchronisation of alternator

of reserve 2 ()

OU : ‘’ ok ‘’ >2 > [ t= 15 min]

IN : < T/S 1 < AND< E 2

S : < T/S 1 < AND< E 2

OU : ‘’ ok ‘’ >21,22 > 20

w1 w2

W3'

T6

22 IN : < T 6 <

S : < T 6 <

OP : Synchronisation of

alternator of reserve 1 ()

OU : ‘’ ok ‘’ > 25 > [ t= 15 min]

IN : < T 6 <

S : < T 6 <

OP : Synchronisation of alternator

of reserve 2 ()

OU : ‘’ ok ‘’ >26 > [ t= 15 min]

21

w3

E 2

Ou: "Turn on the thermal station 1" >19> [m1=1]

Ou: "Turn on the thermal station 2" >20> [m2=1]

Ou : "Turn on the thermal station 1 et 2" > T6 >

[m1 =1, m 2 =1]

Figure 4-b. Excerpt of the HAD diagram of the command system of the PN (following)


50

These HAD models respect the classic Control concepts: The objects are identified by their unique number [20]. The

parallel sequences and the relations of causality are explicit as prescribed in [2] [4] [7] [9] [20], and without

forgetting the delayed actions as is illustrated in the objects 19, 20, 21, and 22 of the figure 4b. HAD performs a

modeling structured of abstract elements of complex and hybrid (continuous and discrete aspects) systems,

encourage the modularity and the exchanges (transfers) between objects, he also shows his/her/its efficiency in the

analysis and the simulation of the physical phenomena [3] [7]. The models present a more precise abstraction of the

system, the operations performed preceding the command of the circuit breakers are materialized (tasks of the

measure power station), the representative model of the electric bow under the shape of difference equation is

present, and the links of causalities are expressed.

4. RULES FOR CONVERTING HAD MODELS INTO GRAFCET

We are going to describe the tools of transformation of the HAD meta-graph toward the Grafcet, to complete and

finalize compatibility between the two metamodels, and whose principle has already been introduced in [3]. It puts

in evidence and guarantees a compatibility between HAD and the Grafcet language.

To permit to achieve the objectives easily, we cut the process of passage in two phases: the Intermediate Model (IM)

and the Final Grafcet Model (FM). It will present in more an advantage during the implementation of software of

HAD simulation and automatic transformation HAD - Grafcet.

4.1. Conversion between HAD and Intermediate Grafcet Model The table 2 present an excerpt [18] of the meta-graph of the rules of conversion between HAD and Intermediate

Grafcet Model (IM).

Table 2. Representation of the rules of conversion of HAD in the Grafcet (IM)

Parallelism and

Conditional

Behaviour - HAD

Parallelism and

Conditional Behaviour -

Grafcet IM

Comments

S

The ActivitySelectONs when they comprise an operation, one inserts

a stage before the selection in OR and the operation is considered like being the associated action.

S

The ActivitySelectOFFs when they comprise an operation, one

inserts a stage after the selection in OR and the operation is considered like being the associated action.

T

The ActivityThreadONs when they comprise an operation, the

selection in AND generate two stages thereafter and the operation is considered like being the associated action.

T

The ActivityThreadOFFs when they comprise an operation, the

selection in AND precede a stage and the operation is considered like

being the associated action.

T

The ActivityThreads / Thread when they comprise an operation, one

intersperses a stage between the two selections in AND, the operation

is considered like being the associated action.

T

The ActivitySelects / Thread when they comprise an operation, the

selection in AND generate two stages thereafter and the operation is considered like being the associated action.

T/S

The ActivityThreads / Select when they comprise an operation, one

intersperses a stage between the two selections in AND and OR, the

operation is considered like being the associated action.

The objects "Begin" and "End" disappear.

1

1

0

1

The ActivityClasses become the stages and the first ActivityClass that receive the message coming from "Begin" is considered like initial

stage.

The operation that it comprises is considered like an action.

E

Transition

This object specifies the transition;

the output message translates the condition to the receptiveness.

Links that join the different objects of the system

Link: external-object or Object – external


51

4.2. Conversion between Intermediate Model and Final Grafcet Model For this phase, it is sufficient to cover the intermediate Grafcet of the top downwards, while applying the following

rules:

- To suppress all parallel and selective sequences not having any embedded stages.

- To perform the logic product of the Transition-conditions consecutively aligned without stages separating them.

The different cases are presented in the table 3.

Table 3. Meta-graph of conversion of the IM to the Final Grafcet Model (FM)

(IM) Consecutive Transition

Conditions

(FM) Equivalence in final

model

a

b

a.b

c

ba

b.ca.c

c

ba

b.ca.c

a b

c

a.b.c

c

a b

a.b.c

It is necessary to specify here that all rules of conversion presented above have been examined with success on

different systems [19] [18].

4.3. Application in the PNSC We are going to present the intermediate and final models for the case of the PNSC. While applying the rules and

correspondences of the table 2, we get the excerpt of intermediate model of the figure 5.

For reasons of readability and simplification (of Grafcets), we are going to conduct an affectation of the

equivalences of Transition-conditions that will be used for the Grafcet of the system:

One will indicate by :

> 0 →b ( > 0. > 0.

> 0 →c ( > 0. > 0.

( < 0. > 0) → ( < 0. <0) →e

One will indicate by : >0 →

=∑S – (D - - ) ; =∑S – (D - ) ;

=∑S – (D - ) ; = ( - )

In an analogous way, we can have [18]:

;

The complements are the next one:

;


52

0

2

d0=0; d1=0; dij=1;

R0=0; R1=0; R2=0

5

7

109

4

8

14

17

13

AB

Closing of the circuit breaker d0

d0=1

sp

d1=1

d2=1

Synchronization of the

reserve R1; R1=1

d1=1

dA1=0 dA2=0

0 0

UarcA2=Er

UarcA2=ErUarcA1=Er

Uarc2 =0Uarc1=0Uarc1=0

Uarc0 =0

K0 =1

(t/3/15min)

K1=1

(t/4/15min)

K2=1 (t/7/15min)K2=1

KA1=0KA2=0KA1=0

)012

.02

.01

()02

.01

(

AAAAA

)012

.02

.01

()02

.01

(

)0

12.0

2.0

1()0

2.0

1(

)012

.02

.01

()02

.01

(

AAAAA

030

3

3Evaluation of

Selection of the process

12Evaluation of

selection of the process

Evaluation of


Updating of the

consumption

11

Uarc2 =0

d2=1

K1=1

15 16dA1=0 dA2=0

UarcA1=Er

KA2=0

6

(t/6/15min)

.)03

()02

.01

.(03

.);0()02

.01

.(0

AAAA

3

A

1 Synchronization of the alternator R0

R0=1

m0=1

Synchronization of the

reserve R2; R2=1 Synchronization

of the reserve

R1;R1=1

Synchronization

of the reserve

R2; R2=1

m2=1 m1=1 m2=1m1=1

Figure 5-a. Excerpt of the intermediate model HAD-Grafcet of the PN


53

0

2

d0=0; d1=0; dij=1;

R0=0; R1=0; R2=0

5

7

109

4

8

14

17

13

AB

Closing of the circuit breaker d0

d0=1

sp

d1=1

d2=1

Synchronisation of the

reserve R1; R1=1

d1=1

dA1=0 dA2=0

0 0

UarcA2=Er

UarcA2=ErUarcA1=Er

Uarc2 =0Uarc1=0Uarc1=0

Uarc0 =0

K0 =1

(t/3/15min)

K1=1

(t/4/15min)

K2=1 (t/7/15min)K2=1

KA1=0KA2=0KA1=0

)012

.02

.01

()02

.01

(

AAAAA

)012

.02

.01

()02

.01

(

)0

12.0

2.0

1()0

2.0

1(

)012

.02

.01

()02

.01

(

AAAAA

030

3

3Evaluation of


12Evaluation of


Evaluation of


Updating of the

consumption

11

Uarc2 =0

d2=1

K1=1

15 16dA1=0 dA2=0

UarcA1=Er

KA2=0

6

(t/6/15min)

.)03

()02

.01

.(03

.);0()02

.01

.(0

AAAA

3

A


R0=1

m0=1


reserve R2; R2=1Synchronisation

of the reserv

R1;R1=1

Synchronisation

of the reserv R2;

R2=1

m2=1 m1=1 m2=1m1=1

Figure 5-b. Excerpt of the intermediate model HAD-Grafcet of the PN (following)

For the generation of the Grafcet FM, we present to the figure 6, just an excerpt of the final Grafcet gotten after

transformation. The final Grafcet pulled of [18] is in conformity with the standards of the Control Engineering and

can be implemented.


54

0

m0=1

Uarc0=0

2 Synchronization of the alternator d0

d0=1

d0=0; d1=0; dij=1;

R0=0; R1=0; R2=0

5

7

9

(t/3/15min).K1

4

8


Reserv R1; R1=1

a

a’.[b.e+b’]a’.[c’+d’]a’.[c+d]

14

17Evaluation of


13

b

b’.KA2'.[gA’+hA’]

1918

12Evaluation of


sp

(t/3/15min).K2

(t/3/15min).K2

Uarc1=0Uarc2=0

Uarc1=Uarc2 =0

d1=1 d2=1

b’.KA1'.[gA+hA] b’.[fA.lA+fA’].KA1’.KA2’

dA2=0dA1=0

UarcA1=Er UarcB1=Er

UarcA1=UarcB1=Er

fA

fA’.KB1'.[gB+hB] fB’.KB2'.[gB'+hB']

dB2=0dB1=0

UarcB1=Er UarcB2=Er

UarcB1=UarcB2=Er

AB

3

1

Evaluation of


Updating of the

consumption

10 11d1=1 d2=1

20 21

15 16dA1=0 dA2=0

6

(t/3/15min).K1

dB1=0 dB2=0

fA‘.[fB.lB+fB’].KB1’.KB2’

3

A


R0=1

K0=1

m1=1 m2=1 m1=1 m2=1


Reserv R2; R2=1 Synchronisation

Of the Reserv R1;

R1=1

Synchronisation

Of the Reserv R2;

R2=1

Figure 6-a. Excerpt - Final Model, Grafcet of command of the PN gotten of the HAD


55

24

22

23

UarcC2=ErUarcC1=Er

AB

UarcC1=UarcC2=Er

dC1=0 dC2=0

Evaluation of


fB’.KC1'.[gC+hC] fB’.KC2'.[gC’+hC’] fB‘.[fC.lC+fC’].KC1’.KC2’

fB

25 26dC1=0 dC2=0

B

Figure 6-b. Excerpt - Final Model, Grafcet of command of the PN gotten of the HAD (following)

Comments on the Grafcet (FM - Final Model) gotten

After this "transformation", we can say that the model gotten present more of readability (more explicit), in relation

to the source mode presented in [19] [18]. For example, we observe that the stage 11 is subdivided in two stages 15

and 16. The separation (detachment) of these stages represents a simultaneous sequence. This case is illustrated to

the figure 7.

11 dA1=0 dA2=0

.Uarc1=Uarc2 =0b’.[fA.lA+fA’].KA1’.KA2’ b’.[fA.lA+fA’].KA1’.KA2’

15 16dA1=0 dA2=0

Figure 7. Comparison between Grafcet of departure - Grafcet gotten by HAD transformation

We also observe the apparition of abstract stages as the stage 3 that brings some more clarity to the diagram. The

HAD modeling can be seen under an angle of simulation. Thus, we observe more details again on this model. It is

relative to the "sub programs" that explains the variation of the electric consumption in the PN.

The gotten model (final Grafcet) respects the rules of the Grafcet [23] above all. It puts in evidence the "detachment"

of the objects (it presents the operational sequences of functioning of an object opposite the other). We observe the

presence of the abstract stages that brings more precisions to the new Grafcet.

All this shows that the HAD-Grafcet conversion succeeds to an even more explicit model, while remaining

compatible with the source Grafcet.

5. CONCLUSION AND PERSPECTIVES

In the setting of the modeling of the HDS and the PN, we applied the improved version of the HAD Metamodel,

associated to the Grafcet language, in order to enhance their efficiency. To the term of the modeling of an exercise

of adaptation of the apparent power constantly produced to a consumption increasing in an excerpt of the PN of

Cameroon, we showed compatibility between HAD and Grafcet with satisfaction. It allows the adepts of the Grafcet

to benefit from advantages of HAD in the modeling, while having the possibility to produce and to operate their

traditional Grafcet, for specific needs.

In short, HAD confirms its strength between two poles:

- Enhancement of UML that he spreads for the modeling of the HDS

- Modeling of the HDS better than the Grafcet that he completes

What makes of HAD an interesting tool and whose subsequent function is to facilitate by his/her/its syntax and his

structure, the implementation of a software application that is the subject of our present works. We put in practice

the technical specifications and the implementation of the simulator of HAD models in order to generate the

corresponding Grafcet, automatically (from a HAD diagram constructs) while implementing directly in the software

the rules of conversion described.


56

7. REFERENCES [1]. E. Tanyi, M. Nkenlifack, Modélisation Unifiée Hybride et Simulation des Systèmes de Contrôle, Revue des Sciences et

Technologies de l’Automatique, Volume 8, N°1 - Premier semestre 2011, pp 31-43, ISSN 1954-3522

[2]. M. Nkenlifack, E. Tanyi, F. Fokou, Amelioration of the HAD Metamodel for the Modeling of Complex Hybrid Systems,

International Journal of Advances Research in Computer Science, Volume 2, No. 1, Jan-Feb 2011, pp 370-380, available

online at http://www.ijarcs.info, ISSN 0976 - 5697

[3]. M. Nkenlifack, E. Tanyi, F. Fokou, Establishing bridges between UML, HAD and Grafcet Metamodels for the Modeling

of Dynamic Systems, International Journal of Scientific and Engineering Research, Volume 2, Issue 3, March 2011, pp 1

- 12, available online at http://www.ijser.org, ISSN 2229-5518

[4]. E. Tanyi, M. Nkenlifack, Une Adaptation d’UML à la Modélisation des Systèmes Hybrides, Revue des Sciences et

Technologies de l’Automatique, Volume 7, N°2 - 2ème semestre 2010, pp 46-57, ISSN 1954-3522

[5]. J. Vareille, Arnaux M., Le langage Grafcet et ses interprétations, La Revue de l’EPI NO9.

[6]. E. Tanyi, T. Noulamo, M. Nkenlifack, J. Tsochounie, A Multi-Agent Design and Implementation of An Internet Based

Platform for the Remote Monitoring and Control of the Cameroon Power Network, Special Issue on Advances in

Information Engineering, Engineering Letters, 13:2_18, ISSN: 1816-0948 (online); 1816-093X (print version), 2006

[7]. M. Nkenlifack, Modélisation objet et développement d’un atelier de simulation des automatismes industriels hybrides,

PhD Thesis, National Polytechnic Institute, University of Yaounde 1), Cameroon, 2004.

[8]. M. Nkenlifack, E. Tanyi, « Hybrid Activity Diagrams : Extendind UML for the modeling of Hybrid Systems », poster in

ECOOP, 17th European Conference on Object-Oriented programming, July 21-25, 2003, Darmstadt University of

technology Germany, www.st.informatik.tu-darmastadt.de :8080/ecoop/posters/index.phtlm - SiteWeb poster

[9]. J. Zaytoon, Systèmes Dynamiques Hybrides, Traité Systèmes automatisés, Information Commande et Communication,

Hermes, Paris, 2001.

[10]. L. Thevenon, Représentation des systèmes Hybrides Complexes par Flux de Données : Développement d’un outil de

Modélisation et de Simulation des procédés Batch, PhD Thesis, National Polytechnique Institute, Grenoble, France, 2000.

[11]. H. Gueguen, Lefebvre M., A comparison of mixed specification formalisms, 4ème Conf. Int. sur l’Automatisation des

Processus Mixtes (ADPM), 2000, Dortmund, Allemagne.

[12]. M. Nkenlifack, E. Tanyi, « An extended UML for the Modeling Hybrid Control Systems », in Burnham K. J., Haas O. C.

L. (editors), proc. of the sixteenth int. conference on Systems Engineering, Convertry, UK, 9-11 September 2003, vol.2,

pp.681.

[13]. J. Lienou, Nkenlifack M., Tanyi E., Noulamo T., A generic multi agent-based platform for reliable diagnostic by DGA,

Advances in Computer Science and Engineering Journal, Vol. 5, Issue 1, Pages 11 – 23, Aug 2010, Pushpa Publishing

House, ISSN 0973-6999

[14]. M. Avele, Conception d’un outil informatique pour l’analyse des réseaux électriques, Thèse du DEA en Sciences de

l’Ingénieur, Ecole Polytechnique, Yaoundé, Octobre 2007.

[15]. Y. Pelenc, Appareillage électrique d’interruption à haute tension, Technique de l’ingénieur Doc. (4700, 4701, 4702,

4703.), 1988, 58 pages.

[16]. Chauvin, Une Centrale de mesure et ses paramètres ,

http://www.chauvin_arnoux.com/groupe/pdf_mag/reseau_CAM_F_51.pdf, (2009).

[17]. E. Tanyi, J. Tsochounie, M. Nkenlifack, T. Noulamo, An Internet-Based Distributed Real-time Control System for the

Cameroon Power Network, IEEE Proc Int. Conf. on Signal & Image Technology and Internet-based Systems (SITI’05),

Yde-Cameroon, Dec 2005.

[18]. L. Domche, Méta modèle HAD-Grafcet : Analyse et mise en œuvre dans la Commande des Réseaux Electriques, Projet

de fin d’études en Licence Génie Electrique IUT/FV de Bandjoun 2009.

[19]. F. Fokou, Méta modèle HAD-UML : Analyse et mise en œuvre dans la Commande des Réseaux Electriques, Projet de fin

d’études en Licence Génie Electrique IUT/FV de Bandjoun 2009.

[20]. R. David., H. Alla, Du Grafcet aux réseaux de Petri, 2e édition, Hermes, France, 1997.

[21]. Traoré S., Une Approche Orientée Objet dans la Simulation des Systèmes Dynamiques Semi-Continus, Thèse de Doctorat

de l’Université de Rennes I, Supelec, Rennes, France, 1994.

[22]. M. Fowler, UML Distilled, Second edition, Addison-Wesley Longman, Inc 2000.

[23]. CEI-IEC (Commission Electrotechnique Internationale), Grafcet specification language for sequential function charts,

Norme Internationale IEC 60848, 2002.

Documents

INTEROPERABILITY BETWEEN THE IMPROVED METAGRAPH HAD AND GRAFCET… · IJRRAS 9 (1) October 2011 Nkenlifack & al. Improved Metagraph HAD and Grafcet 47 transportation sensor transportation