AUTOMATION & CONTROL - Theory and Practice Part 8 pptx

The neuctric power plant y implemented inresponses 0.05 pllers for one unit power station wperforms better n the one unit op in both cases wh h neural PDF and ency control modtribution t

nally, the output

nction that was u

on (3) where S is

p j S

efficient Sjp is pr

in this case the si

h

be accomplished upropriate function

y to define an ehis is obtained wh

p k

r

t function of the used to connect th

n n

i

p i ij

p j

S

p j

e S



 1

1 ) (

using alternative

n depends on theexciting functionhen the weights V

p

jk h

V

1 output neuron ‘

he input and hidd

p

r

p k

e



 1 1

Neural Networkntation of controowledge of the stused to improve t

ed here, PDF The

for each layer is

a weight Wij to tion coefficient

p j

1

 output function equation (4)

p range-limited fu

e final application

n, in order to acVjk have been calc



k l

V 1 ,

‘k’ is obtained usden layers, equati

ks (ANNs) have e

ol strategies is ttructure of the mthe tuning of typere are many cont

s as follows Thethe intermediate

n which represen unctions, such as

n As shown in figccess the output culated

sing the same si

on (6)

earned their posittheir flexibility Imodel of a systemical controllers, stributions in the a

e input

e layer, (3) nts the

(4)

tanh-1, gure 5, layer, (5) gmoid (6)

tion in

If it is

m to be such as area of

artadjGaThGahyfuzpreturhyuse

In dis(20Ne

Eqintfunthe

Fig

tificial neural nejust the parametarcez & Garcez, 19here have been searcez and Garcezydroelectric powezzy inference to esented a self-learbine governor Rybridized controll

ed as governors o this work a bascrete PDF regula000) with great seural-PDF scheme

quations 7 and 8terconnection Vj nction for the err

e change of sign i

g 6 Neural PDF

tworks aimed atters of discrete P995) In this workeveral works wh

z (1995) applied

er plant Djukanov control a low harning control sysRecently, Shu-Qiler based on gen

of a hydroelectricck-propagation sator This strategsucess in practica

in the evolution o

t defining fast anPID control system

k a similar strateghere ANN have

d PI neural contrvic, et al (1997) vhead hydropowestem using a PID

ng et al (2005) hnetic algorithms a

c power plant mostrategy has bee

gy was used to ad

al implementatioregulation can be

1  

j

e h v

t E

er system Yin-So fuzzy NN and ahave compared aand fuzzy NN wodel

en used to adjusdjust a PID controns Figure 6 sho

e calculated by:

j u

e

adjust the weighevelop the minimgradient of the f

tegies to calculat Mukhopadhyay

e a discrete PDF hydroelectric sysimulator of a 20ptive-network baong, et al (2000)applied it to a hyd

a PID controller wwhen the controlle

st the parameterroller by Aguadoows the scheme

hts for each neumization of the trfunction and to e

te and , 1996; ystems

0 MW sed on ) have draulic with a ers are

rs of a Behar

of the (7) (8)

(9) uronal ransfer express

nally, the output

nction that was u

y to define an ehis is obtained wh

p k

r

t function of the used to connect th

n n

i

p i

p j

S

p j

e S



 1

1 )

(

using alternative

n depends on theexciting functionhen the weights V

p

jk h

V

1 output neuron ‘

he input and hidd

p

r

p k

e



 1

1

Neural Networkntation of contro

owledge of the stused to improve t

ed here, PDF The

for each layer is

a weight Wij to tion coefficient

p j

1

 output function

equation (4)

p range-limited fu

e final application

n, in order to acVjk have been calc



k l

V 1 ,

‘k’ is obtained usden layers, equati

ks (ANNs) have e

ol strategies is ttructure of the mthe tuning of typ

ere are many cont

s as follows Thethe intermediate

n which represen unctions, such as

n As shown in figccess the output

culated

sing the same si

on (6)

earned their posittheir flexibility Imodel of a systemical controllers, s

tributions in the a

e input

e layer, (3) nts the

(4)

tanh-1, gure 5, layer, (5) gmoid (6)

tion in

If it is

m to be such as area of

artadjGaThGahyfuzpreturhyuse

In dis(20Ne

Eqintfunthe

Fig

tificial neural nejust the parametarcez & Garcez, 19here have been searcez and Garcezydroelectric powezzy inference to esented a self-learbine governor Rybridized controll

ed as governors o this work a bascrete PDF regula000) with great seural-PDF scheme

quations 7 and 8terconnection Vj nction for the err

e change of sign i

g 6 Neural PDF

tworks aimed atters of discrete P995) In this workeveral works wh

z (1995) applied

er plant Djukanov control a low harning control sysRecently, Shu-Qiler based on gen

of a hydroelectricck-propagation sator This strategsucess in practica

in the evolution o

t defining fast anPID control system

k a similar strateghere ANN have

d PI neural contrvic, et al (1997) vhead hydropowestem using a PID

ng et al (2005) hnetic algorithms a

c power plant mostrategy has bee

gy was used to ad

al implementatioregulation can be

1  

j

e h v

t E

er system Yin-So fuzzy NN and ahave compared aand fuzzy NN wodel

en used to adjusdjust a PID controns Figure 6 sho

e calculated by:

j u

e

adjust the weighevelop the minimgradient of the f

tegies to calculat Mukhopadhyay

e a discrete PDF hydroelectric sysimulator of a 20ptive-network baong, et al (2000)applied it to a hyd

a PID controller wwhen the controlle

st the parameterroller by Aguadoows the scheme

hts for each neumization of the trfunction and to e

te and , 1996; ystems

0 MW sed on ) have draulic with a ers are

rs of a Behar

of the (7) (8)

(9) uronal ransfer express

norwig has a dig

e two control loo

action when big c

gnal (control sign

e vane is adjusteddroop The Droonfiguration is usedower control looped-forward loop

he Dinorwig Govust maintain th

ty When the re

on On the otherrnor valve will clates with two d

or, 2000) The powdefines the operapart load respons

al, directly sets tchanges in the ponal) is produced bsignal The powebetween guide va

ID

g, PI control can raints and their b

es, the performan

vernor

e speed within eference is raised

r hand, when thelose, decreasing groop settings; 1%

wer reference sigating point for thse) Changing thethe guide vane ower reference ap

by adding the out

er feedback loopane opening and

offer good and rbehaviour changence of a linear

ons

nfiguration is shoansoor, 2000) In the power devia

to change the s The frequency cortional to the fresystem to respond

an operational

d the governor

e output signal isgeneration (Wrig

% for high regulgnal sets the refer

he unit when it is

e power referencposition, in ordeppear The guide vtput signals from

p compensates thpower

d to a rapidly-cha

band defined bvalve will open

s lowered the refght, 1989) At Dinlation and 4% foence position for

s working in freq

ce, which also ac

er to produce a vane position ref

m the P, I and D p

he system for the

nce However, alltraints are activa

as PI, can deter

There

ol loop

by the

of the vides a here is anging

by the

n, thus ference norwig

or low

r guide quency

ts as a rapid ference parts to

e

non-l Pnon-lant ated In riorate

sigbecancau200becthiousyssigAtFigAt

a nsat

be gailim

is ctra

Fig

5

Thgowe

gnificantly (Peng,comes excessivel

d it then “winds used by the satur01) In other worcause increasing

is behaviour perutput of the planstem back to its c

gn for a long timtherton, 1995)

gure 8 shows a getherton, 1995) Thnegative value anturation is used t integrated is mo

in (K i) are adjustmit and the dead zcommonly used

acking anti-windu

g 8 General sche

he Simulink© softwvernors This tooere constructed

, et al., 1996) Wh

ly large compare up” In addition,ration effect (Penrds, windup is p the control signsists the integrat

nt As a consequcorrect steady-sta

me The result iseneral PI controllhis controller has

nd forces the out

to reduce the inteodified by the prted in order to mzone depend on t

In this work, the

up structure will

eme of PI anti-win

del and Progra

ware tool was use

ol has libraries ofusing these stan

hen the plant has

ed to a linear resp, a higher integrat

ng, et al., 1996; Broduced when th

al can no longer tor value can beuence, when recate value require

s a large overshler that includes a

an internal feedbput of the systemegrator input As roportional gain maintain equivalethe constraints fix

actuator saturatiponse (an actuattor output and a Bohn & Atherton,

he control signal accelerate the recome very largecovering from sa

s the control errohoot and a long

a tracking anti-wback path, which

m to be in the lin can be seen from

or to be of the opsettling time (Bowindup scheme (Bdrives the integrnear range The in

m Figure 8 the sig

e values of the inassic PI The satutor; a value of 0.9

s governed by a P

er plant under dif

he power plant ming a dialog bo

r value ration),

me are , et al., tuator, lant If

ng the

ng the pposite ohn & Bohn & ator to nternal gnal to ntegral uration

95 p u

PI with

fferent models

ox, the

norwig has a dig

e two control loo

action when big c

gnal (control sign

e vane is adjusteddroop The Droo

nfiguration is usedower control loop

ed-forward loop

he Dinorwig Govust maintain th

ty When the re

on On the otherrnor valve will clates with two d

or, 2000) The powdefines the opera

part load respons

al, directly sets tchanges in the ponal) is produced bsignal The powebetween guide va

ID

g, PI control can raints and their b

es, the performan

vernor

e speed within eference is raised

r hand, when thelose, decreasing groop settings; 1%

wer reference sigating point for thse) Changing thethe guide vane

ower reference ap

by adding the out

er feedback loopane opening and

offer good and rbehaviour changence of a linear

ons

nfiguration is shoansoor, 2000) In

the power devia

to change the s The frequency cortional to the fresystem to respond

an operational

d the governor

e output signal isgeneration (Wrig

% for high regulgnal sets the refer

he unit when it is

e power referencposition, in orde

ppear The guide vtput signals from

p compensates thpower

equency error Th

d to a rapidly-cha

band defined bvalve will open

s lowered the refght, 1989) At Din

lation and 4% foence position for

s working in freq

ce, which also ac

er to produce a vane position ref

m the P, I and D p

he system for the

nce However, alltraints are activa

as PI, can deter

There

ol loop

by the

of the vides a here is anging

by the

n, thus ference norwig

or low

r guide quency

ts as a rapid ference parts to

e

non-l Pnon-lant ated In riorate

sigbecancau200becthiousyssigAtFigAt

a nsat

be gailim

is ctra

Fig

5

Thgowe

gnificantly (Peng,comes excessivel

d it then “winds used by the satur01) In other worcause increasing

is behaviour perutput of the planstem back to its c

gn for a long timtherton, 1995)

gure 8 shows a getherton, 1995) Thnegative value anturation is used t integrated is mo

in (K i) are adjustmit and the dead zcommonly used

acking anti-windu

g 8 General sche

he Simulink© softwvernors This tooere constructed

, et al., 1996) Wh

ly large compare up” In addition,ration effect (Penrds, windup is p the control signsists the integrat

nt As a consequcorrect steady-sta

me The result iseneral PI controllhis controller has

nd forces the out

to reduce the inteodified by the prted in order to mzone depend on t

In this work, the

up structure will

eme of PI anti-win

del and Progra

ware tool was use

ol has libraries ofusing these stan

hen the plant has

ed to a linear resp, a higher integrat

ng, et al., 1996; Broduced when th

al can no longer tor value can beuence, when recate value require

s a large overshler that includes a

an internal feedbput of the systemegrator input As roportional gain maintain equivalethe constraints fix

actuator saturatiponse (an actuattor output and a Bohn & Atherton,

he control signal accelerate the recome very largecovering from sa

s the control errohoot and a long

a tracking anti-wback path, which

m to be in the lin can be seen from

or to be of the opsettling time (Bowindup scheme (Bdrives the integrnear range The in

m Figure 8 the sig

e values of the inassic PI The satutor; a value of 0.9

s governed by a P

er plant under dif

he power plant ming a dialog bo

r value ration),

me are , et al., tuator, lant If

ng the

ng the pposite ohn & Bohn & ator to nternal gnal to ntegral uration

95 p u

PI with

fferent models

ox, the

odels may be cha

stem and linear o

mulink© power

mbining the f

rbine/generator a

r this study; they

stance, there are

onlinear non-elast

ithout rate limita

odel can be adjus

ock has the option

gorithm takes aro

ange) to find the

y can be selecte three models atic and nonlinear ation and saturatsted to represent

n of classical and ink© S-functions ncorporated withrs) and sample tiutput is the cont

he full hydroelems: Guide van

s Each block is p

ed to represent aavailable to simu elastic The guidetion The sensor

t different condit advanced controwere developedhin Simulink© mo

me Its inputs aretrol signal The vcontrol algorithmgorithm calculate

of optimality is q

; however this cri

ns (the exact valu

ne dynamics, part of the Simul

a diversity of moulate the hydrau

e vane dynamics filters block is aions of the natioollers

d for the neural Podels The neura

e the reference anversatility of Simu

m to be modified

es the optimal vaquadratic error, iterion can be cha

ue depending onues (training timuntil the set-poin

perating point of ant as a SISO or Mhows a schematicodel is constructhydraulic subsylink© library deveodes of operatiolic subsystem - can be selected w

a fixed block Thnal grid The govPDF algorithms

n the magnitude me) When these r

nt or the plant

linear MIMO

c of the ted by ystem, eloped

on For linear, with or

he grid vernor These cepts η nals of

e plant

s to be rol law

is the

ry The

of the ranges model

6

AsprosystesFoanconpabaplashoconconopdri

Figcas

Simulation res

s discussed previovide timely andstem The actual sting, it can be spe

r all simulations,

d 50 Hz, and assunnected to the nrameters fixed asis of comparisonant under anti-wows the small snnected In bothntroller, being reperational cases Tiving the process

t K=0.1 and Ti=0

n Figure 10 showwindup PI and nestep responses (0

h cases, the hydespectively 10% aThe undershoot

response of the operation

f a hydroelectric ply of its deman

er demand is rela

of step, ramp andxpressed in the ptem with infinite

of the hydroelec0.12 (as currently

ws the small step reural PDF control0.05 p.u.) of the droelectric plant and 30% faster in

is also reduced i

hydro plant with

station in frequended power contated to Grid frequ random input siper-unit system, nbusbars The neuctric power plant

y implemented inresponses (0.05 pllers for one unit power station wperforms better

n the one unit op

in both cases wh

h neural PDF and

ency control modtribution to the uency variation bgnals

normalized to 30ural PDF controll

t A PI controlle

n practice) is use.u.) of the hydroe

t operational Figwhen all six un with the neuraperational and sixhen a PDF contro

d PI controllers f

de is to power but, for

00 MW

er was

er with

ed as a electric gure 11 its are

al PDF

x units oller is

for the

odels may be cha

stem and linear o

mulink© power

mbining the f

rbine/generator a

r this study; they

stance, there are

onlinear non-elast

ithout rate limita

odel can be adjus

ock has the option

gorithm takes aro

ange) to find the

four sub-systemand sensor filters

y can be selecte three models atic and nonlinear ation and saturatsted to represent

n of classical and ink© S-functions

ncorporated withrs) and sample tiutput is the cont

aviour may be sel

he full hydroelems: Guide van

s Each block is p

ed to represent aavailable to simu elastic The guide

tion The sensor

t different condit advanced contro

were developedhin Simulink© mo

me Its inputs aretrol signal The v

control algorithmgorithm calculate

of optimality is q

; however this cri

ns (the exact valu

a diversity of moulate the hydrau

e vane dynamics filters block is aions of the natio

ollers

d for the neural Podels The neura

e the reference anversatility of Simu

m to be modified

es the optimal vaquadratic error,

iterion can be cha

ue depending onues (training timuntil the set-poin

perating point of ant as a SISO or M

hows a schematicodel is constructhydraulic subsy

link© library deveodes of operatio

lic subsystem - can be selected w

a fixed block Thnal grid The gov

nt or the plant

linear MIMO

c of the ted by ystem, eloped

on For linear, with or

he grid vernor These cepts η nals of

e plant

s to be rol law

is the

ry The

of the ranges model

6

AsprosystesFoanconpabaplashoconconopdri

Figcas

Simulation res

s discussed previovide timely andstem The actual sting, it can be spe

r all simulations,

d 50 Hz, and assunnected to the nrameters fixed asis of comparisonant under anti-wows the small snnected In bothntroller, being reperational cases Tiving the process

t K=0.1 and Ti=0

n Figure 10 showwindup PI and nestep responses (0

h cases, the hydespectively 10% aThe undershoot

response of the operation

f a hydroelectric ply of its deman

er demand is rela

of step, ramp andxpressed in the ptem with infinite

of the hydroelec0.12 (as currently

ws the small step reural PDF control0.05 p.u.) of the droelectric plant and 30% faster in

is also reduced i

hydro plant with

station in frequended power contated to Grid frequ random input siper-unit system, nbusbars The neuctric power plant

y implemented inresponses (0.05 pllers for one unit power station wperforms better

n the one unit op

in both cases wh

h neural PDF and

ency control modtribution to the uency variation bgnals

normalized to 30ural PDF controll

t A PI controlle

n practice) is use.u.) of the hydroe

t operational Figwhen all six un with the neuraperational and sixhen a PDF contro

d PI controllers f

de is to power but, for

00 MW

er was

er with

ed as a electric gure 11 its are

al PDF

x units oller is

for the

he large ramp resural PDF controlleu.) of the power stter using the neely, the one unit uces the undershooss coupling inteand the perturbaponse has a high

rshoot

hydro plant with

sponses (0.35 p.uers for one unit ostation when six eural PDF contro operational andoot

eraction a 0.8 p

tion of unit 1 obher overshoot, the

h neural PDF and

u.) of the hydroeoperational Figur units are generaoller, the respons

d six units operat

u step was appbserved Figure 14

e PI response has

d PI controllers f

electric plant with

re 13 shows largeating In both cas

se being 15% antional cases Agaplied simultaneou

4 shows that, alt

ng time

Figon

Figsix

the hydro plant w

the hydro plant w

he large ramp resural PDF controlleu.) of the power stter using the neely, the one unit uces the undersho

oss coupling inteand the perturba

ponse has a high

rshoot

hydro plant with

sponses (0.35 p.uers for one unit ostation when six eural PDF contro operational and

oot

eraction a 0.8 p

tion of unit 1 obher overshoot, the

h neural PDF and

u.) of the hydroeoperational Figur units are genera

oller, the respons

d six units operat

u step was appbserved Figure 14

e PI response has

d PI controllers f

electric plant with

re 13 shows largeating In both cas

se being 15% antional cases Agaplied simultaneou

4 shows that, alt

ng time

Figon

Figsix

the hydro plant w

the hydro plant w

f pumped storage

er plant The modystems and contaihown how the nimprove its dyna

he system with n

to represent clos

he nonlinear modand encourage u

ments

o thank First Hyd

e of the hydro pla

models a hydroollers has been dually increasing

he rapid inclusioncontrol approach

e stations has beedel includes reprins the principal fneural PDF can bamic response Ineural PDF is impsely the real plandel These are prom

en discussed Thisresentation of thefeatures of the pl

be applied to a h

n particular, it haproved Multivar

nt The coupling bmising results for

e issue of robustn

their assistance

eural PDF control

d allows comparimodular nature o

of the simulation

l methods and alalready included

s model was app

e guide vane, hydlant’s dynamics

hydroelectric pu

as been shown thriable effects havebetween penstoc

r the use of neuraness of the respo

llers

ison of

of this

ns The lso the

d The plied to draulic mped-hat the

e been cks has

al PDF onse in

9 References

Aguado-Behar, A., “Topics on identification and adaptive control” (in Spanish), Book edited

by ICIMAF, La Habana, Cuba 2000

Bohn, C and Atherton, D P., "An analysis package comparing PID anti-windup strategies",

IEEE Control Systems Magazine, vol 15, p.p 34-40 1995

Djukanovic, M B., Calovic, M S., Vesovic, B V., and Sobajic, D J., “Neuro-fuzzy controller

of low head hydropower plants using adaptive-network based fuzzy inference

system”, IEEE Trans on Energy Conversion , 12, pp 375-381 1997

Garcez, J N., and Garcez, A R., “A connectionist approach to hydroturbine speed control

parameters tuning using artificial neural network”, Paper presented at 38th IEEE

Midwest Symposium on Circuits and Systems, pp 641-644 1995

Goodwin, G C., Graebe, S F and Salgado, M E., "Control system design", Prentice Hall,

USA 2001

Gracios, C., Vargas, E & Diaz-Sanchez A., “Describing an IMS by a FNRTPN definition: a

VHDL Approach”, Elsevier Robotics and Computer-Integrated Manufacturing, 21, pp

241–247 2005

Kang, J K., Lee, J T., Kim, Y M., Kwon, B H., and Choi, K S., “Speed controller design for

induction motor drivers using a PDF control and load disturbance observer”, Paper

presented at IEEE IECON, Kobe, Japan, pp 799-803 1991

Kundur, P., Power System Stability and Control, New York, NY: Mc Graw Hill 1994 Mansoor, S P., “Behaviour and Operation of Pumped Storage Hydro Plants”, Bangor, U.K.:

PhD Thesis University of Wales 2000

Mansoor, S P., Jones, D I., Bradley, D A., and Aris, F C., “Stability of a pumped storage

hydropower station connected to a power system”, Paper presented at IEEE Power

Eng Soc Winter Meeting, New York, USA, pp 646-650 1999

Mansoor, S P., Jones, D I., Bradley, D A., Aris, F C., and Jones, G R., “Reproducing

oscillatory behaviour of a hydroelectric power station by computer simulation”,

Control Engineering Practice, 8, pp 1261-1272 2000

Miller T., Sutton S.R and Werbos P., Neural Networks for Control, Cambridge Massachusetts:

The MIT Press 1991

Minsky, M L., and Papert, S A., Perceptrons: Introduction to Computational Geometry

Cambridge, USA: MIT Press 1988

Munakata, T., Fundamentals of the New Artificial Intelligence: Neural, Evolutionary, Fuzzy

and More London, UK: Springer-Verlag 2008

Narendra, K S., and Mukhopadhyay, S “Adaptive control using neural networks and

approximate models”, Paper presented at American Control Conference, Seattle, USA,

pp 355-359 1996

Peng, Y., Vrancic, D and Hanus, R., "Anti-windup, bumpless, and conditioned transfer

techniques for PID controllers", IEEE Control Systems Magazine, vol 16, p.p 48-57

1996

Rumelhart, D E., McClelland, J L., and Group, T P., Parallel distributed processing:

Explorations in the microstructure of cognition (Vol 1) Cambridge, USA: MIT Press.1986

Shu-Qing, W., Zhao-Hui, L., Zhi-Huai, X., and Zi-Peng, Z “Application of GA-FNN hybrid

control system for hydroelectric generating units”, Paper presented at International

Conference on Machine Learning and Cybernetics 2, pp 840-845 2005

makes possible thvement of the c

ments

o thank First Hyd

e of the hydro pla

models a hydroollers has been

dually increasing

he rapid inclusioncontrol approach

e stations has beedel includes reprins the principal f

neural PDF can bamic response In

eural PDF is impsely the real plandel These are prom

en discussed Thisresentation of thefeatures of the pl

be applied to a h

n particular, it haproved Multivar

nt The coupling bmising results for

e issue of robustn

their assistance

eural PDF control

d allows comparimodular nature o

of the simulation

l methods and alalready included

s model was app

e guide vane, hydlant’s dynamics

hydroelectric pu

as been shown thriable effects havebetween penstoc

r the use of neuraness of the respo

llers

ison of

of this

ns The lso the

d The plied to draulic mped-

hat the

e been cks has

al PDF onse in

9 References

Aguado-Behar, A., “Topics on identification and adaptive control” (in Spanish), Book edited

by ICIMAF, La Habana, Cuba 2000

Bohn, C and Atherton, D P., "An analysis package comparing PID anti-windup strategies",

IEEE Control Systems Magazine, vol 15, p.p 34-40 1995

Djukanovic, M B., Calovic, M S., Vesovic, B V., and Sobajic, D J., “Neuro-fuzzy controller

of low head hydropower plants using adaptive-network based fuzzy inference

system”, IEEE Trans on Energy Conversion , 12, pp 375-381 1997

Garcez, J N., and Garcez, A R., “A connectionist approach to hydroturbine speed control

parameters tuning using artificial neural network”, Paper presented at 38th IEEE

Midwest Symposium on Circuits and Systems, pp 641-644 1995

Goodwin, G C., Graebe, S F and Salgado, M E., "Control system design", Prentice Hall,

USA 2001

Gracios, C., Vargas, E & Diaz-Sanchez A., “Describing an IMS by a FNRTPN definition: a

VHDL Approach”, Elsevier Robotics and Computer-Integrated Manufacturing, 21, pp

241–247 2005

Kang, J K., Lee, J T., Kim, Y M., Kwon, B H., and Choi, K S., “Speed controller design for

induction motor drivers using a PDF control and load disturbance observer”, Paper

presented at IEEE IECON, Kobe, Japan, pp 799-803 1991

Kundur, P., Power System Stability and Control, New York, NY: Mc Graw Hill 1994 Mansoor, S P., “Behaviour and Operation of Pumped Storage Hydro Plants”, Bangor, U.K.:

PhD Thesis University of Wales 2000

Mansoor, S P., Jones, D I., Bradley, D A., and Aris, F C., “Stability of a pumped storage

hydropower station connected to a power system”, Paper presented at IEEE Power

Eng Soc Winter Meeting, New York, USA, pp 646-650 1999

Mansoor, S P., Jones, D I., Bradley, D A., Aris, F C., and Jones, G R., “Reproducing

oscillatory behaviour of a hydroelectric power station by computer simulation”,

Control Engineering Practice, 8, pp 1261-1272 2000

Miller T., Sutton S.R and Werbos P., Neural Networks for Control, Cambridge Massachusetts:

The MIT Press 1991

Minsky, M L., and Papert, S A., Perceptrons: Introduction to Computational Geometry

Cambridge, USA: MIT Press 1988

Munakata, T., Fundamentals of the New Artificial Intelligence: Neural, Evolutionary, Fuzzy

and More London, UK: Springer-Verlag 2008

Narendra, K S., and Mukhopadhyay, S “Adaptive control using neural networks and

approximate models”, Paper presented at American Control Conference, Seattle, USA,

pp 355-359 1996

Peng, Y., Vrancic, D and Hanus, R., "Anti-windup, bumpless, and conditioned transfer

techniques for PID controllers", IEEE Control Systems Magazine, vol 16, p.p 48-57

1996

Rumelhart, D E., McClelland, J L., and Group, T P., Parallel distributed processing:

Explorations in the microstructure of cognition (Vol 1) Cambridge, USA: MIT Press.1986

Shu-Qing, W., Zhao-Hui, L., Zhi-Huai, X., and Zi-Peng, Z “Application of GA-FNN hybrid

control system for hydroelectric generating units”, Paper presented at International

Conference on Machine Learning and Cybernetics 2, pp 840-845 2005

Werbos, P J., Beyond regression: New Tools for Prediction and Analysis in the Behavioral

Sciences, Cambridge, USA: PhD Thesis Harvard University 1974

Widrow, B., and Hoff, M E., “Adaptive switching circuits”, Paper presented at IRE

WESCON Convention Record 4, pp 96-104 1960

Working group on prime mover energy supply, I, “Hydraulic turbine and turbine control

model for system dynamic studies”, IEEE Trans.s on Power Systems , 7, 167-179

1992

Wright, R M., "Understanding modern generator control", IEEE Transactions on Energy

Conversion, vol 4, p.p 453-458 1989

Yin-Song, W., Guo-Cai, S., & Ong-Xiang, “The PID-type fuzzy neural network control and

it's application in the hydraulic turbine generators”, Paper presented at Power

Engineering Society meeting 1, pp 338-343 2000

Intelligent Network System for Process Control: Applications, Challenges, Approaches

Qurban A Memon

X

Intelligent Network System for Process Control:

Applications, Challenges, Approaches

Qurban A Memon,

UAE University, United Arab Emirates

1 Introduction

The ever increasing size, sophistication and complexity of sorting and handling systems, as

industries strive to increase efficiency and offer greater consumer choice, is a constant

problem for system controllers and integrators who build such systems In particular, the

growing trend towards larger capacity systems, which increase throughput rates and

provide greater flexibility to deal with product variation and changing supply demand,

place increased pressure on system controllers to get things right on time and within

budget, often in the face of loose specifications resulting in large and complex programs In

view of these demands it comes as no surprise that, when designing and integrating a

control scheme, control engineers prefer to use past practice to minimize risk Current

practice includes distributed and modular systems like programmable logic controllers

(PLC’s) to break down complex control

The need for modularity, real timeliness, integrated diagnostics, decentralized control,

expanding physical setups, and functionality has resulted into various types of distributed

control system (DCS) Presently, the DCS concept is not only applied to the process control

and manufacturing automation but it has targeted many areas like communications

engineering (Yang, 2006), physics and nuclear science (Cavineto et al., 2006; Kleines et al.,

2004), power system engineering (Stewart, et al., 2003; Ioannides, 2004), a lighting system

design (Alonso, et al., 2000) etc The innovative ideas are being applied in field, for example

in (Alonso, et al., 2000), the authors have made use of the power line as the communication

medium for developing the DCS of the lighting system, based on a custom built large scale

integrated Neuron chip incorporating both the control and communication features in a

three microprocessors based system Another domain of application is industrial plant

automation to reduce operational cost, reduce product development time to market and

improve flexibility of the plant in manufacturing a range of products The decentralization

in the control systems has brought communication and coordination amongst various

sub-control systems to surface as an important aspect in building an optimal sub-control system

design The research and development in the area of DCS network design is quite active

because of developments in the communications and information technology field As an

example, the authors in (O’Hearn et al., 2002) discuss an approach for combination of

11