Transientbehaviourofinductionmotorsinislandconditionsduetointerruptionsin

theship’selectricalpowersupply

Bachelor’sdissertation

FacultatdeNàuticadeBarcelonaUniversitatPolitècnicadeCatalunya

Student:HarmanpreetSinghRattan

Director:

Dr.PauCasalsTorrens

Bachelor’sdegreeinMarineTechnologies

Barcelona,19thofOctoberof2017

3

Summary

Thefocuspresentdissertationwasaproblemthatoccursinindustrialinstallationsandships.

Thesaidinstallationsexperienceovervoltageincaseofpowercut.Thisproblemisaresultof

installingcapacitivecompensationparalleltomotors.Thisdissertationstudiedthisproblem

withtwodifferentelectricmotorswidelyusedinbiginstallations.

The motors were tested in laboratory in different operational conditions; with different

connections (D and WYE), with different loads and without load and with and without

capacitivecompensation.Thecapacitorbank,sourceofcapacitivecompensation,wasalso

connectedinDandWYEduringthetests.

Theovervoltagewasobservedwithcapacitivecompensationonlywhenthecapacitorbank

wasconnectedinDandthemotorswereconnectedinWYE.Themechanicalloadontheshaft

hasdeterminedthepeakvalueofthetransientvoltageandthedurationofthewaveform.

The laboratory testing circuits, inwhich the overvoltagewas observed,were recreated in

Matlabtovalidatetheresultsobtainedinreal-world.Thesimulationresultswereverysimilar

totheirreal-worldcounterpartsin3ofthe4cases.

In conclusion, theovervoltagehasmanifestedwhen thevoltage receivedby thecapacitor

bankwassuperiorthanthemotor’s.Theloadhasalsoplayedanimportantroleindetermining

thepeakvalueoftransientofwaveform.

4

Tableofcontents1 Introduction................................................................................................................8

2 Mathematicalmodel.................................................................................................102.1 Physicalaspects............................................................................................................10

2.1.1 Workingprincipleofamotor..........................................................................................11

2.2 Equivalentcircuit..........................................................................................................112.2.1 Approximantequivalentcircuits......................................................................................13

2.3 Characterizationofinductionmotor.............................................................................142.3.1 Blockedrotortest............................................................................................................16

2.3.2 No-loadtest.....................................................................................................................17

3 LaboratoryTesting....................................................................................................243.1 Testinginstallation........................................................................................................25

3.1.1 Electriccircuit..................................................................................................................30

3.2 Formula-basedanalysis.................................................................................................343.2.1 Dynamicanalysis.............................................................................................................35

4 Test’sresults.............................................................................................................364.1 Results’analysis............................................................................................................46

5 Matlabsimulation....................................................................................................475.1 Simulationresults.........................................................................................................51

5.1.1 Simulationresults’analysis..............................................................................................53

6 Conclusion................................................................................................................56

7 Bibliography.............................................................................................................57

5

TableoffiguresFigure1:Motor'sparts............................................................................................................10

Figure2:Exactequivalentcircuit............................................................................................12

Figure3:Exactequivalentcircuitwithloadresistance...........................................................13

Figure4:Approximantequivalentcircuitwithoutload..........................................................14

Figure5:Approximantequivalentcircuitwithload................................................................14

Figure6:Squirrelcageinductionmotor..................................................................................15

Figure7:Woundrotorinductionmotor..................................................................................15

Figure8:(Right)Dconnectionand(Left)WYEconnection.....................................................15

Figure9:Testingcircuit(Source:Máquinaseléctricas:PauCasalsTorrens[1]).....................16

Figure10:TheVo^2-Pographofmotorno.1.........................................................................18

Figure11:TheVo^2-Pographofmotorno.2.........................................................................19

Figure12:Simulationofexactequivalentcircuitinno-loadcondition(motorno.1)............21

Figure13:Simulationofexactequivalentcircuitinblockedrotorcondition(motorno.1)...21

Figure14:Simulationofexactequivalentcircuitinno-loadcondition(motorno.2)............21

Figure15:Simulationofexactequivalentcircuitinblockedrotorcondition(motorno.2)...22

Figure16:Powertriangle........................................................................................................24

Figure17:Thecontactor.........................................................................................................26

Figure18:Thecapacitorbank.................................................................................................26

Figure19:Thetachometer(Onfarright),themechanicalbrake(Onfarleft)andtheammeter

(Behindthetachometer).................................................................................................26

Figure20:Theanalyser...........................................................................................................27

Figure21:DatawatchPro'smainwindow...............................................................................27

Figure22:Theanalysermountedintheprotectivecase........................................................28

Figure23:Connectiondiagramoftheanalyser......................................................................28

Figure24:Circuitimplementationoftheanalyser..................................................................29

Figure25:TheOscilloscope.....................................................................................................29

Figure26:ThemotorconnectedinD......................................................................................30

Figure27:ThemotorconnectedinWYE.................................................................................30

Figure28:MotorandcapacitorconnectedinWYE.................................................................30

Figure29:MotorconnectedinWYEandcapacitorbankinD.................................................30

Figure30:MotorconnectedinDandcapacitorbankinWYE.................................................31

Figure31:MotorandcapacitorconnectedinD.....................................................................31

Figure32:Phase-neutralconnectionofthecapacitorandthemotor....................................31

Figure33:AsynchronousMachine(SIunits)...........................................................................47

Figure34:Asynchronousmachine'sparameterswindow.......................................................48

Figure35:Three-phaseprogrammablevoltagesource..........................................................49

Figure36:Three-phaseload....................................................................................................49

Figure37:Three-phasebreaker..............................................................................................49

Figure38:Three-phaseV-Imeasurement...............................................................................50

Figure39:Voltagemeasurement............................................................................................50

Figure40:Scope......................................................................................................................50

Figure41:Step........................................................................................................................50

Figure42:Gain........................................................................................................................50

Figure43:Ground...................................................................................................................51

Figure44:Powergui................................................................................................................51

Figure45:Simulationcircuit....................................................................................................51

Figure46:Testno.5-Simulationresult..................................................................................52

Figure47:Testno.5-Labtestresult......................................................................................52

6

Figure48:Testno.7-Simulationresult..................................................................................52

Figure49:Testno.7-Labtestresult.......................................................................................52

Figure50:Testno.30-Simulationresult................................................................................52

Figure51:Testno.30-Labtestresult....................................................................................52

Figure52:Testno.32-Simulationresult................................................................................53

Figure53:Testno.32-Labtestresult....................................................................................53

Figure54:Testno.34-Simulationresult................................................................................53

Figure55:Testno.34-Labtestresult....................................................................................53

Figure56:Transientthree-phasevoltagewaveform..............................................................54

Figure57:Transientwaveformofthethree-phasecurrent....................................................55

Figure58:TransientwaveformoftheRMS.............................................................................55

7

TableoftablesTable1:Ratedvaluesofmotors..............................................................................................14

Table2:No-loadtestresults...................................................................................................18

Table3:Mechanicallossesofmotors.....................................................................................19

Table4:No-loadandblockedrotortests'results...................................................................20

Table5:Motors'parameters...................................................................................................20

Table6:Simulatedno-loadandblockedrotortest.................................................................23

Table7:Simulatedvalueofimpedances.................................................................................23

Table8:Differenttestconditions............................................................................................33

Table9:Tableofequivalentimpedances................................................................................34

Table10:Tableofresultsoflaboratorytests..........................................................................45

Table11:Relationbetweenpowerandinertia(Source:Máquinaseléctricas.JesúsFraileMora

[2])...................................................................................................................................48

Table12:Valuesofmutualinductionandinertiaofsimulatedmotors..................................48

Table13:Simulation’swaveforms..........................................................................................53

Table14:Comparisonofsimulationresultsandtestresults..................................................54

8

1 IntroductionThe inventionof alternative currentmotor revolutionized theworld. It is efficient and

requiresminimalmaintenance.Nowadays,thesemotorscanbespottedatanyindustry.

Themainreasonofthatisitsscalability.Theycanbeasbigasorassmallasitstaskrequires

ittobe.Anotherreasonforitspopularityisitsversatility.Bymakingsmallmodifications

initsdesign,theoptimaloperationalpointcanbechanged.Inotherwords,itisrelatively

easytotweakitsoptimalpower,speedandtorqueoutput.

Themostwell-known and popularmotor in industrial environment is the three-phase

inductionmotor. The principal behind the inductionmotor, also known as alternative

currentmotor, functioning isthesameasthegeneratorofthealternativecurrent, i.e.,

electromagneticinduction.Sincetheworkingprincipleofthesetwomachinesistheone

and same, theoretically, the inductionmotor canoperateas thegenerator and supply

currenttodevicesconnectedonsamenetwork.Anditdoesthat,foraverylittleduration,

everytimethepowersupplyisshutdown.

Innormaloperationalconditionsthisphenomenon isnotharmful.Becausethecurrent

that the motor produces can be considered non-existent compared to what is being

suppliedtootherequipmentbythenetwork.Theproblembeginswhenthereisapower

cutandthemotor’soperationisinterrupted.Nowthecurrentthatisbeingproducedby

themotor is reaching to the other electrical devices that are connected on the same

networkorgridasthemotor.

Themagnitudeandfrequencyofthecurrentandvoltageareanythingbutnominal.The

electronicequipmentthatishighlysensibletosmallvariationsofthecurrentareinagreat

danger under these circumstances. Theworst case scenario is when themotors have

capacitivecompensationbecauseitcanproduceovervoltage.

Theovervoltagederivedfromcapacitivecompensationcancorruptthedata,damagethe

internalcomponentsorevencauseburninthewholeequipment.Thisphenomenontakes

placeonshipsandinindustrialinstallations.That’swhythroughoutthisdissertationthere

willbenodistinctionbetweenshipandindustrialnetwork.

Sotheobjectiveof thisdissertation is toanalysethis transientbehaviourof themotor

whenthereisapowercutandithascapacitivecompensation.Differentstepstoachieve

thatgoalarelistedbelow:

Ø The transientbehaviourwill be studiedon twodifferent typesmotors that arewidelyusedintheindustry.

9

Ø Itwillbestudiedindifferentoperationalconditionsforeachmotor.

Ø Itwillbestudiedthathowtheloadconnectedtothemotorcanaffectthecurrentproducedbyitandhowtheabsenceofloadcontributestothisphenomenon.

Ø Itwillbestudiedwhilethecapacitorbankisconnectedparalleltothemotor.

Later on, the results gathered during the lab experiments will be compared with the

simulated results onMatlab. The outcomes from the real-world experimentation and

computersimulationshouldbeverysimilarifnotthesame.

10

2 Mathematicalmodel2.1 Physicalaspects

Before looking intoamodel that representsanelectricmotor it ispertinent todiscuss

aboutdifferentpartsofmotorandhowtheyinteractwitheachother.That’swhythispart

isdedicatedtothemainpartsofthemotorandtheirfunction.

Athreephaseinductionmotor,oranyelectricmotorforthatmatter,consistsoftwomajor

parts:astatorandarotor.Therotorisnotattachedtothestatorinanyway;thereisa

smallseparationdistancebetweenthetwoknownasairgap.Therotarymovementisthe

resultofinteractionofmagneticfieldscreatedbythewindingsoftherotorandstator.

Thestatorconsistsofframe,coreandwinding.Thestatorframeisanenclosureandthe

outermost part of the induction motor. It houses the junction box where cables are

connected that provide electric current to themotor. It acts as a protection fromany

external damage that might occur. It provides mechanical strength to the internal

components.

Thestatorcore’smainpurposeistoserveashousingforthestatorwindings.Themagnetic

fluxflowsthroughthecoreandtoreducetheeddycurrentsthecoreismadelaminated.

Ithasvariousequidistantperipheralslotsparalleltoitscentralaxis.

Therotoristhepartoftheinductionmotorthatisresponsiblefortherotarymovement.

Ithasashaftthatconnectstothemechanicalloadandinthemiddleislocatedalaminated

structureofcylindricalformthathousesslotsforthewinding.Thiswindingcanbeoftwo

type:

i. Conventionalthreephasewindingmadeofcopperwireii. Squirrelcagewinding.

Figure1:Motor'sparts

11

Thesquirrelcagewindingconsistsofaluminiumorcopperbarsthatareinsertedinthe

slotspreviouslymentioned.Attheextremesofthecylinderthesebarsareweldedtothe

ringsmadeof samematerialas thebars.Thisway the ringsand thebars formaclose

circuit.

Theinductionmotorsarenameddependingontheirrotorconfiguration.Theoneswith

squirrel cage windings are known as squirrel cage induction motor. The ones with

conventionalwindingsareknownaswoundmotors.Thelatterisalsoknownasslipring

motorsdue to the fact that ithasslip ringsmountedon its shaft thatareused toadd

resistancetotherotor.Bydoingthis,thetorque-speedcurvaturecanbechangedtoobtain

desiredstartingtorque.

2.1.1 Workingprincipleofamotor

The current enters through the junction box in the stator windings. When it starts

circulating through the stator windings it creates a rotary magnetic field. Due to

electromagnetism,thismagneticfieldinducescurrentinrotorwindings,i.e.,theelectrons

intherotorwindingsstartmovinginaparticulardirection.

Thecurrentthatnowcirculatesthroughtherotorwindingsalsocreatesarotarymagnetic

field.Thetworotarymagneticfieldsinteractwitheachother.Theresultofthisinteraction

istherotatorymotionoftherotor.

2.2 Equivalentcircuit

In order to analyse and study machine’s behaviour it is necessary to represent its

functionalityinacomprehensiveway.Thisrepresentationmustenglobeeverykeyaspect

ofthemachine;every importantcomponentmustbeincludedintherepresentation. It

hastobesufficienttounderstandthathowthemachineworks,whatareitsparameters

andhowtheseparametersinterconnectwitheachothertomakethemachinefunctional.

Ininductionmotors,thisrepresentationisachievedbySteinmetzequivalentcircuit.This

equivalent circuit features all the essential components required for the analysis of

inductionmotors.That’swhy,itisfeaturedinalmosteverybookthattalksaboutinduction

motors.

Althoughtheinductionmotorisathree-phasemachine,theequivalentcircuitisasingle-

phase representation for the sake of simplicity. For that reason, when using this

representation onemust procure to use phase voltage since the connection is phase-

neutral.

12

Thecircuitthatisshownbelowisknownasexactequivalentcircuit.Obviously,theactual

motorcomponentsarenotconnectedinthismanner.However,itdoesgiveanideathat

howthesepiecesinteractanddependoneachother.Themainpurposeofthiscircuitis

todeterminethathowmuchcurrentpassesthrougheachwindingwhichwillultimately

beusedtocalculatethepoweroutput.

Where:

Re=Statorresistance(W)Xe=Statorleakagereactance(W)Rr’/s=Rotorresistancereferredtostator(W)Xr’=Rotorleakagereactancereferredtostator(W)Rm=Corelosesresistance(W)Xm=Magnetizingreactance(W)

Thecurrent thatenters through stator (I1) isdivided into twocurrents,one that flows

through rotor (I2’) and another one that flows through magnetizing branch (I0). The

amount of current that passes through the two fractions depends directly on the

rotationalspeedandtorqueoutput.Thesetwoparametersdecidethevalueofslip(s)that

isillustratedinthecircuit.So,therotorresistancedependsonthespeedandthetorque

outputofthemotor.

Themagnetizingbranch isnotaphysical componentof the inductionmotor.Rather it

representstheimpedancethatisimposedbythecoreofthestator.Thecurrentthatflows

throughthisbranch is theoneresponsible forcreatingtherotatingmagnetic fieldthat

inducescurrentintherotor.

I1 I2

I0

Figure2:Exactequivalentcircuit

13

Figure3:Exactequivalentcircuitwithloadresistance

Inordertomakethecalculationseasiertherotorresistanceisdividedintotwoelements.

Rr’+Rr’('()))=Rr’+Rr’('

)− 1)=Rr’/s (2.1)

Rr’('()))=Resistancethatrepresentsmechanicalloadontheinductionmotor.

With this distinction it is very easy to calculate thedevelopedpowerby themotor as

illustratedintheformulabelow.

Pi=3*(I2’)2*[Rr’('()))] (2.2)

Bysubtractingthemechanical lossesofthemotorfromthedevelopedpowerthetotal

outputpowerisobtained.

PO=Pi-Pm (2.3)

Thedifferencebetweentwocurrents,I0andI2’,valueisratherbigindifferentoperational

states. Therefore, an approximate equivalent circuit is used inwhich either current is

considerednegligibledependingonoperationalstate.

Therearetwomainstatesunderwhichaninductionmotorcanoperate,i.e.,underload

ornoload.Thedistinctionbetweenthesetwooperationalstatesisimportantbecausethe

presence or absence of load determines the path that current takes after node of

magnetizingbranch.

2.2.1 Approximantequivalentcircuits

Whenthereisnoloadattherotorasmallfractionofcurrentpassesthroughrotor,enough

to produce sufficient torque to overcome friction and winding losses associated with

rotation.Soitcanbeconsiderednegligible.Underthisconditionthemajorityofcurrentis

usedtoproducereactivepowerthatgeneratesthemagneticfield,i.e.,I1»I0.

I1I2’

I0

14

Figure4:Approximantequivalentcircuitwithoutload

TheRminthemagnetizingbranchrepresentstheresistanceofairinairgap.Itisusually

veryhigh.Forthatreason,manytimesitisomittedfromthecircuitsalltogetherbecause

verylittleamountofcurrentpassesthroughthere.Itmakesthecalculationseasier.

Whenthemotorisrotatingsomethingi.e.thereisaloadconnected,itneedstorqueand

consequently power to rotate that load. Under load, the majority of current passes

throughtherotorandthemagnetizingbranch’scurrentcanbeneglected,i.e.,I1»I2’.Thiscurrent isusedtocalculatethepoweroutputofthe inductionmotor.Thismechanical

powerdependshighlyonthevalueofslip.

Figure5:Approximantequivalentcircuitwithload

2.3 Characterizationofinductionmotor

For thepurposeof thisdissertation,parametersof twodifferent inductionmotorsare

determined.One is a squirrel cage inductionmotorandanotherone is awound rotor

motor.Theveryfirststeptowardsthecharacterizationistocollecttheratedvaluesofthe

motor. It iseasilydonebyextracting information from thenameplate that is attached

motorhousing.

Motorno.1 Motorno.2Motortype Squirrelcageinductionmoto WoundrotorinductionmotorRatedvoltage 220/380V 220/380VRatedcurrent 4.8/2.8A 2.5/1.4ANominalpower 1000W 600WRatedspeed 1420rpm 1500rpmFrequency 50Hz 50HzPowerfactor - 0.8

Table1:Ratedvaluesofmotors

15

Asitcanbeseen,therearetworatedvoltagesandcurrentsforeachmotor.Thatiswhy

becausethreephasemachinesallowtwodifferenttypesofconnection,WYEandD(delta).

Thetypeofconnectiondependsonthenominalvoltageofthenetwork.Ifthevoltageis

thesameasthemotor’s,thentheconnectionmustbeinD.Ifthevoltageis3squareroot

timesthenominalvoltageofmotor,thentheconnectionmustbedoneinWYE.

Theapproachtakenindeterminingthesemotors’parametersisthemosttraditionalone.

Itconsistsofdoingtwotestsonsubjectmotors,namely,no-loadtestandlocked-rotortest,

andmeasuringthefollowingparameter:

Figure6:Squirrelcageinductionmotor

Figure7:Woundrotorinductionmotor

Figure8:(Right)Dconnectionand(Left)WYEconnection

16

- Activepower(W)- Voltage(V)- Current(A)- Reactivepower(VAr)- Apparentpower(VA)- Powerfactor

Withthisdatacollectedfromthebothtestsabovementionedresistancesandleakages

reactancearecalculated.Variousformulasandassumptionsareemployedintheprocess

ofcalculation.

Bothtestsaredoneataroomtemperatureof24ºC.Beforecommencingthetests,the

stator resistance of both motors is measured with an Ohmmeter. To measure the

parametersabovementioned,followingcircuitisimplemented:

2.3.1 Blockedrotortest.

As the name suggests, during this test the rotor is blocked, i.e., there is no rotatory

movement.Whenthereisnorotationthevalueofslip(s)is1.Oncetherotorissecured,

withthehelpofvariablevoltagesourcesmallvoltageincrementsareapplied.Thevoltage

supplymustbestoppedwhenthecurrentreachesitsratedvalue.Thisreadingisnamed

asIRB.

Apartfromthecurrentfollowingdataiscollected:

VRB:Thevalueofthisvoltageiswellbelowtheratedone.Sincethevoltageissmalland

theresistanceofthemagnetizingbranchisbig,theintensitydoesnotflowthroughthere.

PRB:Itistheelectricpowerconsumedbythemotor.Itrepresentsthecopperlosses,due

toJouleheating,inthestatorandtherotor.

Figure9:Testingcircuit(Source:Máquinaseléctricas:PauCasalsTorrens[1])

17

fpRB:Ithasrelativelyhighvalue.

Onceallthedataiscollectedfollowingformulasaretocalculatetheimpedances:

ZRB=>RB?RB (2.4)

ZRB=RRB+jXRB (2.5)

RRB=DRB

E∗?RB^H (2.6)

RRB=Re+Rr’ÞRr’=RRB–Re (2.7)

XRB=Xe+Xr’ (2.8)ThereisnoeasywaytoobtaintheXeandXr’.Forthemotorslessthan5.5kW,following

relationbetweenreactancesisacceptable[2].

Xe=Xr’=XRB/2 (2.9)

2.3.2 No-loadtest

Asthenamesuggests,thistestisperformedwithoutcouplinganyloadtothemotorshaft.

Andagain, voltage (V0), intensity (I0),power (P0)and fparemeasuredusing the same

equipment.Thesereadingsareusedtoestimatethepowerlossesinthemotor,namely,

mechanical, coreandcopper losses.Thepower (P0) that is consumedby themotor is

dividedasfollowing:

P0=Pm+PH+Pcu (2.10)

Pcu=3*Re*I02 (2.11)

PH:Corelossespertainingtothemagnetizingbranch

Pm:Mechanicallosses

Pcu:Copperlosses

Thistestisperformedwithvariousvoltages.Eachvoltagedeliversdifferentpower.With

differentvaluesofvoltageandpowerV02-P0graphcanbetraced.Withthehelpofthis

graphthevalueofpowercanbedeterminedwhenthevoltageis0.Withthevoltagebeing

18

0,corelossesarealso0sincethereisnocurrentflowingthroughthemagnetizingbranch.

Sothatvalueofpowerrepresentsthemechanicallosses.

Motorno.1 Motorno.2V0(V) P0(W) V0(V) P0(W)

101,7 42,4 100 70,2

125 52,1 125 83,05

150 66,7 150 92,45

175,4 76,7 175 100,73

200,4 96,5 200 109,1

219 143,4 230 128

225,2 157,4 - -

Table2:No-loadtestresults

Followingthevalues listed inthetableabove,thevaluesofPmofbothmotorscanbe

determined.Thevariationinvoltagesappliedtobothmotorsisduetothefactthatthe

voltageofthegridisneverconstant.

Figure10:TheVo^2-Pographofmotorno.1

Whentheaxisofabscissasisformedbysquarenumberandtheaxisofordinatesisformed

bynumbersthatarenotsquaredthegraphisrectilinear.That’swhy,inthegraphabove,

atrendlinehasbeenusedtodescribetherelationbetweenP0andV0.Thistrendline

passesclosetotherealpoints.

Thesamestrategyhasbeenemployedinthegraphbelow.Bothtrendlineshavetheir

correspondentequations.Tocalculatethemechanicallosses,thexissubstitutedwith0.

y=0,0027x+6,1611

0

50

100

150

200

0 10000 20000 30000 40000 50000 60000

Po(W)

19

Figure11:TheVo^2-Pographofmotorno.2

Motorno.1 Motorno.2Pm(W) 6 61

Table3:Mechanicallossesofmotors

Intheabovetable,theapproximatevalueofmechanicallossesislisted.Becauseofthe

small variationof speed inno-loadand full-loadconditions thePmcanbeconsidered

constantineveryoperationalcondition.Thereisonlyoneparameterlefttodetermine

and that is Xm. To calculate the Xm of magnetizing branch following formulas are

employed:

Z0=>0?0 (2.12)

Z0=R0+jX0 (2.13)

X0=Xe+XmÞXm=X0–Xe (2.14)

Thereadingsobtainedfromthetestperformedonthemotorsaredisplayinthetables

below.

No-loadtest Motorno.1 Motorno.2V0(V) 219 230I0(A) 2.65 1.926P0(W) 143.1 128Q0(VAr) 994.7 756.4S0(VA) 1005 767fp 0.14 0.17

y=0,0013x+61,25

0

20

40

60

80

100

120

140

0 10000 20000 30000 40000 50000 60000

Po(W)

20

BlockedrotortestVRB(V) 116.8 45.4IRB(A) 4.81 2.5PRB(W) 870 146.4QRB(VAr) 430 131.3SRB(VA) 960 196.6fpRB 0.9 0.74

Table4:No-loadandblockedrotortests'results

After applying the formulas mentioned earlier following values of impedances are

obtained:

Motorno.1 Motorno.2Re(measured) 9.5 6.68ZRB 12.6+j6.1 7.76+j7.05

Z0 6.7+j47.45 11.72+j67.94Rr’ 3.1/s 1.08/sXe=Xr’ j3.05 j3.525Xm j44.4 J64.42Ze 9.5+j3.05 6.68+j3.525Zr’ 3.1/s+j3.05 1.08/s+j3.525

Table5:Motors'parameters

Atfirst,thereseemstobenoproblemwiththevaluesofimpedancesofmotorno.1.But

acloseexaminationshowssomecontradiction in thevalues.Forexample, thecopper

lossesintheno-loadtestare,atminimum:

PCU=3*9.5*2.65^2=200.14W (2.15)

ThisvalueclearlysurpassestheP0obtainedduringthetests.TheP0mustbe,at least,

equaltothecorelosses.Theimpedancecalculatedinno-loadconditionrepresentsthe

sumofresistanceandreactancesofthestatorandthemagnetizingbranch.

Themeasuredresistanceofthestatorandthecalculatedonedonotmatchwitheach

otherinbothcases.Partlytheworkingtemperatureistoblame.Thetemperatureofthe

windingsissubjecttochangewiththetemperature.

Inordertovalidatetheseresultsandobtainthetruevalueofeveryparameter,aMatlab

simulation is employed. This simulation contains an exact equivalent circuit. The

simulatedcircuithasalotofblocksformeasuringthesignals.

21

Figure12:Simulationofexactequivalentcircuitinno-loadcondition(motorno.1)

Figure13:Simulationofexactequivalentcircuitinblockedrotorcondition(motorno.1)

Figure14:Simulationofexactequivalentcircuitinno-loadcondition(motorno.2)

22

Figure15:Simulationofexactequivalentcircuitinblockedrotorcondition(motorno.2)

The no-load and the blocked rotor test will be simulated in this circuit with the

correspondingvoltages. For theno-load test the slip (s)willbe close to0and for the

blockedrotortestitsvaluewillbe1.Thesourcevoltagewillbesameasmeasuredduring

tests.

Asmentionedbefore,theresistancesareunreliablewiththevaryingtemperature.But

thisisnotthecaseforreactances.That’swhy,thestartingpointofthesesimulationswill

betheRmandleakedreactancesofstatorandrotor.Thesevalueswillremainalmostthe

sameduringthesimulations.

Atstarting,thevalueoftheresistancewillbeclosetotheonemeasured.Fromthere,

smallchangeswillbemadeinthevaluesofresistancesuntilthereadingsarecoherent.

Theconsumedcurrentshouldbeveryclosetoonemeasured.

Afternumeroustriesfollowingvaluesareobtained.Thesimulatedvaluesareveryclose

totheirreal-worldcounterpart.

No-loadtest Motorno.1 Motorno.2 Measured Simulated Measured SimulatedV0(V) 219 219 230 230I0(A) 2.65 2.62 1.926 1.817P0(W) 143.1 182.1 128 127.4Q0(VAr) 994.7 981.5 756.4 712.4S0(VA) 1005 998.25 767 723.7fp 0.14 0.18 0.17 0.176

23

BlockedrotortestVRB(V) 116.8 116.8 45.4 45.4IRB(A) 4.81 4.95 2.5 2.55PRB(W) 870 887.2 146.4 148.5QRB(VAr) 430 456.4 131.3 134.8SRB(VA) 960 997.7 196.6 200.56fpRB 0.9 0.889 0.74 0.74

Table6:Simulatedno-loadandblockedrotortest

Motorno.1 Motorno.2 Calculated Simulated Calculated Simulated

Re 9.5 8.5 6.68 6.2

ZRB 12.6+j6.1 12.6+j6.1 7.76+j7.05 7.76+j7.05Z0 6.7+j47.45 8.73+j47.68 11.72+j67.94 12.86+j71.94Rr’ 3.1/s 4.1/s 1.08/s 1.56/sXe=Xr’ j3.05 3.05 j3.525 3.525Xm j44.4 J44.6 J64.42 J69.11Ze 9.5+j3.05 8.5+j3.05 6.68+j3.525 6.2+j3.525Zr’ 3.1/s+j3.05 4.1/s+j3.05 1.08/s+j3.525 1.56/s+j3.525

Table7:Simulatedvalueofimpedances

Nowthevaluesareintunewitheachotherwithsmallvariations.Theonlyexceptionis

theRethatdoesnotmatchwithresistanceofZ0.Everythingelse,fitsrightinplace.

These obtained values of impedances will be later used in theMatlab simulation to

validatetheresultsobtainedinthetests.Morespecifically,theymustbeintroducedin

theblockofmotorsothatitcansimulatethemotor’sbehaviour.

24

3 LaboratoryTesting

As outlined in the introduction, in this part, the transient behaviour of the motors

willbeobservedwhenthereisapowercut.Asmentionedbefore,thesubjectofthese

testsaretwothree-phaseelectricmotors,namely,squirrelmotorandwoundrotormotor

thatwillbetestedindifferentoperationalconditions.

Thesetestsarecarriedout in laboratoryconditions.Sotherearesomefactorsthatare

verydifferentfromwhatonecanfindinfactories.Forexample,theroomtemperatureis

notexcessivelyhot,therearenoEMI(electromagneticinterference)sourcesnearby,there

isnoabruptvoltagechangesinthenetwork(duetohighenergyconsumptionelements

connectinganddisconnecting)andtherearenoexternalvibrations.

Saidthat,therearefactorsthatcanbecontrolledinthelaboratoryfacility.Thosefactors

aregivenbelow.

1) Theoperationalconditionsofmotorscanvaryfromonedaytoanother,speciallytheload.Sotosimulatethoseconditionsthetestswillbedonewithdifferentload.Itis

achieved by connecting a mechanical brake with variable resistant torque to the

motor. The torque that applies this brake can be controlled from a dedicated

switchboard.

2) Other than the load that varies, there is also two different type of connectionspossible,WYEandD.Mostofthetimes,thenominalvoltageofanygivengridisfixed

since every security element and other elements are installed for one particular

nominalvoltage.So, theconnectionof themotor remains thesameas longas the

motoroperatesinthesamenetwork. Inthelaboratory,thenominalvoltageisaround230V.Sometimes,it’scloseto225V

andothertimesitmayexceeditsnominalvalue,eventhough,itcannotbeconsidered

overvoltage.Thesubjectmotorsoperate,nominally,at220V.Buttheavailablevoltage

canbeconsideredvalidtoperformthetestssinceavariationofupto10%isadmitted.

Nowasdiscussedearlier,itiscrystalclearthattheconnectioninthejunctionboxmust

beinD.However,forthesakeofexperimentation,therewillbesometestingdone

withWYEconnection.Thereasonforthatistoobservethathowtheoutcomevoltage

willbewhenthemotorispoweredwithavoltagelowerthanitsnominal.

3) Whenpeopletalkaboutpowerorelectricpowertheyrefertowatts(W)orkilowatts

(kW) which is the active power. But

there are two other powers that areFigure16:Powertriangle

25

consumedbyamachine.Thesearereactivepower(Q)andapparentpower(S)with

unitsVArandVArespectively.Therelationbetweenthreeisshownbelow.

P=V*I*cosj (3.1)Q=V*I*sinj (3.2)S=V*I (3.3)S= (PH + QH) (3.4)

j=Thepowerfactor.

The active power is the responsible for creating the output power. However, the

powerconsumedbythemachineistheapparentpower.Thereactivepower,inthe

caseofmotors,isinductiveanditisusedforcreatingthemagneticfield.Thepower

factor shows the relation between S and P, the smaller the angle smaller is thedifferencebetweenbothpowers.

Thebigger is reactivepowerbigger is theapparentpower,eventhough, theactive

power remains the same.Which translates into bigger electric bill. That’swhy the

capacitorbanksareusedtominimizethereactivepower. It isachievedbyinjecting

capacitivereactivepower,whichhasanoppositedirectionthantheinductiveone,into

theinstallation.

Very often, industrial installations have capacitor banks to minimize the reactive

powerconsumption.Since,motorsconsumeinductivereactivepower,thecapacitor

banks lower the overall installation reactive power by contributing the capacitive

reactivepower.ThesecanalsobeconnectedinWYEorDasperdemand.

Consideringalloftheabove,thereareseveralpossiblecombinationsthatwillresultin

different operational conditions. That’s why, tests will be done with different load

conditions, i.e., no load, with mechanical brake coupled with the motor and with

mechanicalbrakeperformingthetorqueof2N.m.intheoppositedirectionofthemotion

ofmotor. Additiontothat,theeffectsofcapacitivecompensationonthevoltagegenerationwillbe

observed.Forthat,testwillbedonewhenthecapacitorsareinWYEconnectionandD

connection.And,asmentionedearlier,therewillbeexperimentationwithbothtypesof

possibleconnections.

3.1 Testinginstallation

Thelaboratorytestsarecarriedoutusingfollowingequipment:

• Themotors

• ElectricCables–Forthesetests,electricwireswithcrosssectionareaof1.5mm2areemployed.Thissectionismorethanenoughbecausethecurrentthatpassesthrough

themissmall.

• A contactor - throughout the tests many power interruptions will be made.Disconnectingthewiresdirectlyfromswitchboardisveryrisky.Thispracticecanlead

26

toseriousinjuriesordeathduetoelectrocution.Toavoidthisdanger,acontactorwill

beemployed.Itisaswitchthatcloseswhenitisconnected24V(DC)current.With

thisamountofcurrentthereisnodangerofelectrocution.

• Acapacitorbank–Itconstitutesofthreecapacitorsof10µF.ThesecanbeconnectedinWYEorDconfiguration.

Figure18:Thecapacitorbank

• Anadjustablemechanicalbrake–Thisdeviceisusedtoproducecounter-torque.

• Atachometer–Thisdevicemeasurestherevolutionsperminute(rpm)ofthemotor.Itconnectsontheoppositeofthemechanicalbrake.

• Anammeter–Thisdevice isusedtomeasuretheamps(A). Itwillbeconnected inserieswithoneofthephaseofthemotor. It isverysimpleandreliableelementto

monitorthecurrent.

Figure19:Thetachometer(Onfarright),themechanicalbrake(Onfarleft)andtheammeter(Behindthetachometer)

Figure17:Thecontactor

27

Thefigure19representsthetypicalsetupduringthetests.Whenthereisnoneedfor

load,themechanicalbrakeissimplydetachedfromthemotor.Theammeterisalways

connected to one of the phases. The probe of oscilloscope connects in one of the

terminalsofjunctionbox.

• Ananalyser - Thisdevice isused toanalyseanarrayofparametersofdomesticorindustrialgrid.Itcanbeemployedinthree-phaseorsingle-phasenetwork.It isalso

equippedwithportsthatcancontrolrelaysandalarmsifnecessary.Thepowersource

ofthisdeviceisonethephases.

Figure20:Theanalyser

Themeasurementsaredisplayedonasmalldisplaymountedonthedevice.Themost

commononesarevoltageandcurrent.Theanalyserhasmanymorefunctionsandto

takeadvantageofitsfullcapabilitiesitisconnectedtoacomputeroralaptop.

The connection is via Ethernet. The readings andmuchmore is visualized using a

specializedsoftware.Thissoftwareismadebythesamecompanyastheanalyserand

is called Datawatch Pro. Once the software is installed, the analyser is simply

connectedtothecomputerusingRJ45cableandisconfiguredasperinstructions.

Every reading taken by the analyser can be visualized on a computer in a form of

numerical values or graphs in real time. Further to these readings, the device also

measuresthepowerconsumption.Everyreadingandmeasurementisstoredlocally

onthecomputerwhichcanbeconsultedonanylaterdate.

Figure21:DatawatchPro'smainwindow

28

Theanalyseralsomeasuresharmonicspresentintheelectriccurrent.Notonlythat,it

separatesthemaccordingtotheirtype.All inall, it isagreatdevicetomonitorthe

parametersandqualityofelectriccurrent.

Figure22:Theanalysermountedintheprotectivecase

Theanalyserismountedinaplasticboxinordertoprotectitfromanyphysicalabuse.

Itisconnectedbetweenthegridandthemachine,inthiscaseamotor.Thewiresfrom

bothsidesareconnectedintheterminalstripadjacenttotheprotectivecase.

Asitcanbeseen,therearefourpointstoconnectthewiresandoneofthemisfor

neutralwire.InsidetheboxtherearethreeRogowskicoilsresponsibleformeasuring

thecurrentofeveryphase.Thedelicatewireconnectionsofthesecoilsareanother

reasontouseaprotectivecase.

Figure23:Connectiondiagramoftheanalyser

In thisparticular case,between theanalyserand themotor there isa contactoras

showninthefigure24.Theanalyserturnsonautomaticallywhenconnectedtothe

network.Beforecommencingwithanyoperation,thebuttonofstartreadingisclicked

onthefirstpage.

29

Figure24:Circuitimplementationoftheanalyser

Aftercommencingwithtesting,therewasaveryapparentproblem.Thedevicewas

notregisteringthehighcurrentconsumption,around5timestheratedcurrent,atthe

momentofstarting.Thisphenomenonisverycharacteristicoftheinductionmotors.

Furthermore,itwasnotregisteringorreadinganytransientwaveformofthevoltage

afterpowershutdown.Thevalueofthevoltagedroppedto0afterstoppingthesupply

ofelectricity.Thisconditionwasconsistentineverytestperformedwiththeanalyser.

Afterinspectionofeverysettinginthesoftware,itwasfoundthatitwasreadingor

takingsampleseverysecond(s).Thatwasthesmallestpossiblesampletimeavailable.

That is why, the mentioned phenomena did not appear on screen because they

happenedinamatterofmilliseconds(ms).

For the reasonmentioned above, it is impossible to perceive the transient voltage

usingtheanalyser.Tovisualizeandcapturethetransientwaveformanoscilloscope

willbeused.

• Anoscilloscope–Thisapparatuswillbeusedtoobservethetransientwaveformofthevoltage.

Figure25:TheOscilloscope

30

3.1.1 ElectriccircuitAsithasbeenexplainedbefore,thecapacitorbankandthemotorwillbeconnectedwith

each other electrically in different forms. There are various possible configurations

becausesometimes thecapacitorswillnotbeconnectedandother times theywillbe

connectedintheWYEorDconnection.Allthesixpossiblecombinationsarelistedbelow.

Figure27:ThemotorconnectedinWYE

Figure28:MotorandcapacitorconnectedinWYE

Figure29:MotorconnectedinWYEandcapacitorbankinD

Figure26:ThemotorconnectedinD

31

Figure30:MotorconnectedinDandcapacitorbankinWYE

Figure31:MotorandcapacitorconnectedinD

Figure32:Phase-neutralconnectionofthecapacitorandthemotor

All these combinations offer different equivalent impedance that is perceived by the

electricgrid.Thefirststeptowardsfindingtheequivalentimpedancecircuitistoknow

the impedanceofmotorduringeveryoperation. Thereare variouspoints to keep in

mind:

• Therotorresistancevarieswiththeload(Rr’/s).Slipcanbecalculatedbyknowingthespeedineverycaseandsynchronousspeedofthemotor(ns).

s=b)(bc

b) (3.5)

• When there is no load connected, following the approximant equivalent

circuit,ZeandXmaresummeduptocalculatetheimpedanceofthemotor.

• Whenthereisaloadconnected,followingtheapproximantequivalentcircuit,ZeandZr’aresummeduptocalculatetheimpedanceofthemotor.

32

• If the capacitor bank is not connected to the grid, then the equivalentimpedancewillbelongtooneofthecasesmentionedabove.

• IfthecapacitorbankisconnectedinWYE,thenimpedancesofthemotorandthecapacitorbankaresummedasinanyotherparallelRLCcircuit.

Zt=

qr∗qsqrtqs

(3.6)

Ifnot,thentheimpedanceofcapacitorbankinDmustbeconvertedintoitsequivalentimpedanceinWYEusingthefollowingformula.

ZD-Y=qs^HE∗qs

(3.7)

Once the conversion is completed, the equivalent impedance is calculated

accordingtotheformula3.6.

Theimpedancehastwoparts,onerealandoneimaginary.Iftheimaginarypartisbigger

than0thenthecircuitisinductive.Iftheimaginarypartissmallerthan0thenthecircuit

iscapacitive.Theoreticallyspeaking,themorecapacitiveiscircuit,thelargeritsimpact

wouldbeonthetransientvoltage.

Additiontothedifferentcombinationsmentionedabove,themechanicalbrake isalso

connected to themotor in somecases. In total,34 testare carriedoutwithdifferent

combinationsofcapacitorbank,motorandmechanicalbrake.Thesecombinationsare

listedbelowwiththeircorrespondenttestnumber.

Testno. Motor MC CB Brake(N.m)1 1 WYE No No2 1 WYE No 03 1 WYE No 24 1 WYE WYE No5 1 WYE D No6 1 WYE WYE 07 1 WYE D 08 1 WYE WYE 29 1 WYE D 210 1 D WYE No11 1 D D No12 1 D WYE 013 1 D D 014 1 D WYE 215 1 D D 216 1 D D 4,217 2 D No No18 2 D No 019 2 D No 220 2 D WYE No

33

21 2 D D No22 2 D WYE 023 2 D D 024 2 D WYE 225 2 D D 226 2 WYE No No27 2 WYE No 028 2 WYE No 229 2 WYE WYE No30 2 WYE D No31 2 WYE WYE 032 2 WYE D 033 2 WYE WYE 234 2 WYE D 2

Table8:Differenttestconditions

MC:ItstandsformotorconnectionwhichcanbeinWYEorD.

CB:Itstandsforcapacitorbankwhichcanbeconnectedornottothemotor.Itcanbe

connectedinDorWYE.

Brake:IfitsaysNothenthemechanicalbrakeisnotconnectedtothemotor.Ifthevalue

is0thenitisconnectedtothemotorbutitisnotperforminganycounter-torque.Ifthe

valueisgreaterthan0thenitrepresentsthecounter-torqueitisproducingonthemotor.

Theequivalentimpedancehasbeencalculatedforeverytestdoneinthelaboratory.The

list of these impedances can be found below. For the calculation of these equivalent

impedances,thefactorsabovementionedaretakenintoconsideration.

Testno. Equivalentimpedance1 8,5+47,65i2 90,5+6,1i3 22,17+6,1i4 11,74+55,67i5 27,43+82,51i6 86,78-18,93i7 56,01-44,21i8 50,19-1,74i9 44,76-15,68i10 11,74+55,67i11 27,43+82,50i12 116,09-42,68i13 56,01-44,21i14 86,78-18,93i15 44,76-15,68i16 15,21-29,01i17 6,2+72,635i18 45,2+7,05i

34

19 21,8+7,05i20 10,40+93,85i21 60,25+219,11i22 46,29+0,49i23 42,92-12,032i24 25,89+5,14i25 26,89+0,78i26 6,2+72,635i27 22,91+7,05i28 12,88+7,05i29 10,40+93,85i30 60,25+219,11i31 15,86+6,43i32 24,96+1,78i33 13,45+6,651i34 13,96+5,81i

Table9:Tableofequivalentimpedances

Theseimpedanceswillbecrosscheckedwiththecasesinwhichovervoltageisproduced.

Thisway,itcanbeprovediftheequivalentimpedancehasanyimpactonthetransient

voltageornot.

3.2 Formula-basedanalysis

Acircuitinwhichacapacitor,aninductorandaresistanceisconnectedtoeachotheris

knownasasecondordersystem.Itisnamedlikethisbecauseithastwoenergystoring

device,i.e.,thecapacitorandtheinductor.ItisalsoknownasRLCcircuit.Essentially,the

RLCcircuit is thesameas thecapacitorandthemotorconnected inparallel inphase-

neutralconfigurationasillustratedinFigure25.

Whenthepoweriscutofffromthiscircuit,itproducestransientvoltageinsimilarfashion

as the inductionmotors.Thisphenomenon isdubbedas transient responseofsecond

ordersystem.This transient responsecanbestudiedwith formulas.Theoretically, the

samemethodcanbeemployedtostudythetransientwaveformproducedbythemotor.

vc(t)=2*úA1ú*e-at*cos(wdt+/A1),t>0 (3.8)

The formula 3.8 describes the waveform produced by the RLC circuit. The negative

exponentensuresthattheendvalueis0.Thespeedwithwhichthisequationequatesto

0dependsonaandt.Thebiggerisathefasteritequatesto0.TheparametersofthisformulaforaparallelRLCcircuitarethefollowing:

a= xHy (3.9)

w02=

'yz (3.10)

wd= aH + {0H (3.11)

35

p1,p2=-a±wdi (3.12)

TheR,LandCare, respectively, resistance, inductanceof inductorandcapacitanceof

capacitor. The value of A1, which is a complex number, is obtained by solving the

followingmatrixoperation.

1 1|1 |2 x

}1}2 =

~ 0t

~(0t) (3.13)

y(0+)=vc(0

+)=vc(0

-)=V (3.14)

y’(0+)=vc’(0

+)=0 (3.15)

A2iscomplexconjugateofA1.Itmeansthatthemagnitudeofrealpartandimaginary

partofA2issamebutthevalueofcomplexcomponentofA2isoppositeinsign.TheVis

thevoltageofthegridtowhichthecircuitisconnected.

This formula-based analysis can be used to validate the results of real-world tests.

However,inthecaseofinductionmotors,thephenomenonofovervoltageoccursatt=

0.Thisformulaisusefulwhenthetimeisgreaterthan0.AlsotheproductofA1and2is

equal to the voltage suppliedby the grid. So, themaximumvalueobtained from this

formulawillneverexceedthevalueofnetwork’svoltage.

3.2.1 Dynamicanalysis

The next step in formula-based analysis of motor is to analyse its behaviour during

acceleration and deceleration. Since the transient voltage is produced during the

deceleration.Thisanalysiscommenceswiththefollowingformula:

T–Tr=J dωdt (3.16)

Where:

T:Electromagnetictorqueofmotor(Nm)

Tr:Counter-torqueoftheload(Nm)

J:Inertia(kg.m2)

ω:Rotationalspeedofmotor(rad/s)

Aftertheprocessofintegrationofthegivenequation,followingequationisobtained:

τmec=ÇÉ

ÑÖÜá (3.17)

Thisequationgivestheelectro-mechanictimeconstant,alsocalledtau,thatdetermines

thetimethatarotorwithJinertiataketoacceleratetothesynchronousspeed(ω)under

theinfluenceofthemaximumtorque(Tmax)

36

4 Test’sresults

Followingeverythingexplainedupuntilnow34testareperformed.Theresultsofthese

tests are catalogued in the tablebelow. Theparametersobtained from the tests are

motor’sspeed,consumedcurrent,peakvalueoftransientwaveformandthedurationof

thesaidwaveform.Theimageoftransientvoltagewaveformproducedineverytestcan

befoundonthefarrightofthetable.

Thetransientwaveformisobservedandanalysedontheoscilloscope.Theparameters

suchaspeakvalueoftransientwaveformanditsdurationaremeasureddirectlyonthe

oscilloscope.

Theapparatusdoesnothavetheabilitytotellifthereisanyovervoltage.However,itis

relativelyeasytovisuallytellthepresenceofanovervoltage.

Thepeakvalueofthetransientwaveformiscomparedwiththepeakvalueofvoltagethat

isbeingsuppliedbythegrid.Ifthevalueoflatterishigherthenthereisnoovervoltage.

Forthesakeofthiscomparison,thepeakvalueofgrid’svoltageisalsolistedinthetable.

Testno. Motor Vp(V) RPM I(A) Vm(V) t(ms) Overvoltage Transientwaveform

1 1 179,65 1450 0,66 124 482 No

2 1 179,65 1425 1,53 122 334 No

3 1 179,65 1050 3,85 90 158 No

4 1 179,65 1450 0,3 172 940 No

38

5 1 179,65 1450 0,63 228 1060 Yes

6 1 179,65 1425 0,5 148 434 No

7 1 179,65 1425 0,7 200 432 Yes

8 1 179,65 1350 1,25 144 216 No

39

9 1 179,65 1350 1,35 148 170 No

10 1 311,17 1450 2,6 148 519 No

11 1 311,17 1450 1,7 140 990 No

12 1 311,17 1450 2,5 136 474 No

40

13 1 311,17 1425 1,7 140 508 No

14 1 311,17 1425 2,82 176 162 No

15 1 311,17 1350 2,2 134 252 No

16 1 311,17 1400 3,5 94 112 No

41

17 2 311,17 1450 2 150 180 No

18 2 311,17 1440 2,2 144 164 No

19 2 311,17 1350 2,5 148 144 No

20 2 311,17 1450 1,65 128 200 No

42

21 2 311,17 1450 0,85 164 400 No

22 2 311,17 1440 1,65 180 216 No

23 2 311,17 1440 0,9 156 286 No

24 2 311,17 1375 2,2 168 134 No

43

25 2 311,17 1375 1,55 138 190 No

26 2 179,65 1440 0,6 144 300 No

27 2 179,65 1360 0,75 168 150 No

28 2 179,65 1150 1,6 68 138 No

44

29 2 179,65 1440 0,3 158 552 No

30 2 179,65 1440 0,7 296 600 Yes

31 2 179,65 1240 1,2 112 175 No

32 2 179,65 1360 0,82 248 434 Yes

45

Table10:Tableofresultsoflaboratorytests

Motor:Ittellsthenumberofthemotorbeingtested.Vp:Itisthepeakvalueofvoltage.Itrepresentsthehighestvalueofthesinusoidalwaveformofthevoltage.ItissmallerinWYEconnection.RPM:Itrepresentsthespeedofthemotorineverycase.I:Itrepresentsthecurrentconsumedbythemotor.Vm:Itrepresentsthepeakvalueofthevoltageproducedbythemotoraftershuttingdownthepower.t:Itrepresentsthedurationofthevoltagewaveformproducedbythemotors.Overvoltage:IfitsaysNothenthereisnoovervoltageproduced.IfitsaysYesthentherehasbeenanovervoltage.

33 2 179,65 1150 1,4 132 156 No

34 2 179,65 1120 1,55 208 158 Yes

46

4.1 Results’analysis

Theovervoltageisobservedinfivedifferentcases,twoformotor1andthreeformotor2.Ineverycase,thecapacitorbankisconnectedparalleltothemotor.Andineverycase,thecapacitorbankisconnectedinDandthemotorisconnectedinWYE.Inthemotorno.1,theovervoltageisproducedwhenitisnotconnectedtoanyloadanditisconnectedtothemechanicalbrake.Inthelattercase,therewasnocounter-torqueperformedonthemotor.Apartfromtheovervoltage,thecapacitorbankalsocontributesinextendingthetransienttimeregardlessofitstypeofconnection.Even if there is noovervoltage, theVm is greater in the caseswhereCB is connectedcomparedtothoseinwhichthereisnocapacitivecompensation.TheeffectsofCBaremitigatedwhenthemechanicalbrakeisconnectedInthemotorno.2,theovervoltageisproducedinthreedifferentinstances,inno-loadcondition,when it is connected to theMBwithoutanycounter-torqueandwhen it isconnectedtotheMBwiththecounter-torqueof2N.m.WhenthemotorsareconnectedinDconnection,theresultingVmisusuallyhigherwhentheCBisconnectedinWYEconfiguration.AndwhenthemotorsareinWYEconnectiontheVmisalwayshigherwhentheCBisinDconnection.Theloadconnectedtothemotoralwaysmitigatesthetransientvoltage.Bothinitspeakvalueandthetransienttime.Ifthereiscounter-torquetheeffectsofloadareaggravated.Butthereseemstobesomeexceptionsinthistrend.As for equivalent impedance, there seems to be no relation between the value ofimpedanceandthevoltageproduced.Thetestno.7istheonlycapacitivecircuitinwhichtheovervoltageisobserved.Intherestofthecases,thecircuitsareinductive.In conclusion, themotors do produce an overvoltage when they are connected to acapacitorbank.Forthattohappen,theCBmustbeconnectedtothehighervoltagethanthemotor.Evenifthereisnoovervoltage,thecapacitivecompensationleadstohigherVmthanthecaseswithoutanyCBconnected.

47

5 Matlabsimulation

As outlined in the introduction, the data collected from tests is validated using thesimulationofthephenomenon.Forthesimulation,Matlabisused.Thecircuitusedinthelaboratorytestingisreproducedinthissoftware.

It has an extensive library of elements to choose from. Every device, machine andapparatususedinthelaboratoryisavailableinMatlabintheformofblocks.Theseblockshave several parameters that one must know in order to make it functional in thedesirableway.Once the blocks have been updated with the correct information, they must beinterconnectedwitheachotherintherightway.Thesimulationwouldnotworkifanyessentialinformationisnotintroduced;itwouldgiveanerror.Thesimulationcircuitsforboththemotorsaresame.Theonlythingsthatchangearetheparameters.Now,followingtheintroduction,thecasesthatpresentovervoltagewillbesimulated. Thewaveforms of voltage of the simulationwill be compared to the onesobtainedduringthetests.Theblocksthatareusedinthesimulationsarelistedbelow.

• Theasynchronousmachine(SIunits):Thisblockrepresentsthethree-phasemotor.It

can be used as squirrel cagemotor or wound rotormotor. It has three inputs toconnectthewires.Italsohasaninputtointroducethecounter-torque(Tm)whichisnotused.Because, if thecounter-torque isestablished, itkeepsrotatingthemotorevenifthereisapowercut.Ithasoneoutputportwhichcanbeconfiguredtoshowavarietyofparameters.Forthesakeofthetesting,therotorspeedisselectedbecausethedataofmotor’sspeedisavailablefromthelaboratorytest.

Asfortheparameters,basicmotorinformationisintroduced,suchasnominalpower,rated voltage and frequency. The motor’s resistances are also introduced in theparameterssection.Sincetherotorresistancereferredtothestator(Rr’/s)dependsonspeeditisadjustedwitheverychangeinthespeed.Anotherthanresistances,leakedreactancesarealsoaddedintheformofinduction(H).Theloadthatisdraggedbythemotoriscontrolledbyfrictionfactor.Actually,the

Figure33:AsynchronousMachine(SIunits)

48

frictionfactordirectlyaffectsthespeed,thehigherthefrictionfactorthelowerthespeed.Sincetheloadaffectsthespeed,bycontrollingthespeedtheeffectofloadonthemotorissimulated.

Figure34:Asynchronousmachine'sparameterswindow

Other parameters, such as, mutual induction and inertia cannot be found in thetechnicalspecificationsnorcanbecalculated.AccordingtotheJesúsFraileMorainhisbookMáquinas eléctricas [2], the value of the inertia can be between 0.0015 and0.0035J(kg.m2).Belowisatablecontainingthevaluesofinertiafordifferentpowers.

kW J(kg.m2)0.55 0.00150.75 0.00181.1 0.00251.5 0.0035

Table11:Relationbetweenpowerandinertia(Source:Máquinaseléctricas.JesúsFraileMora[2])

Thevaluespresentedinthetableserveasguidance.Asforthemutualinduction,itisamatteroftrialanderror.So,inordertogetthedesirableresults,manycombinationsofmutualinductionandinertiaaretried.Oncetheresultmatchwith,orcomescloseto,oneoftheresultsobtainedinthetests,thetwoparametersaredetermined.Theselectedvaluesofthesetwoparametersarelistedbelow.

Squirrelcagemotor WoundrotormotorMutualinduction(H) 0.46 0.5Inertia(J(kg.m2)) 0.003 0.003

Table12:Valuesofmutualinductionandinertiaofsimulatedmotors

49

Now for simulatingdifferent caseson the samemotor theonly things that requirealterationarethefrictionfactorandtherotorresistance.Therestoftheparametersremainunchanged.TheonlylimitationofthisblockofmotoristhatitispreconfiguredtorunwithWYEconnection.But,thisrestrictiondoesnotaffectthesimulations inanywaybecauseeveryovervoltagecaseintestswaswithmotorinWYEconnection.

• Three-phaseprogrammablevoltagesource:Withoutthisblockthesimulationwouldnotwork.Thetwoparametersthatmustbeaddedtothisblockarelinetolinevoltageandthefrequencythatare220Vand50Hz.

Figure35:Three-phaseprogrammablevoltagesource

• Three-phase load:Thisblockrepresentsthecapacitorbank.Thetypeofconnectionmustbeselected;inthiscase,itisdelta(D)connection.Andthepowerproducedbythecapacitorsmustbeintroduced.

QC=3*(VL^2)*2*p*50*C (4.1)

Byapplyingthisformula,thecapacitivereactivepowerofthecapacitorsiscalculated.SincetheCis10µF,theQCis456.16VAr.

Figure36:Three-phaseload

• Three-phasebreaker:Thisblockisusedtointerruptthepowersupplytothemotor.It

isequivalentof thecontactoremployed inthetesting. Itcutsoff thepowersupplyafter3seconds.

Figure37:Three-phasebreaker

50

• Three-phaseV-Imeasurement:Thisblockhasanoutputportsthatshowsthevoltageandcurrentofeverysinglephase.

Figure38:Three-phaseV-Imeasurement

• Voltagereader:Thisblockisusedtomeasurethevoltageofasinglephase.Thereading

ofthisblockisusedforthecomparisonsofthewaveforms.

Figure39:Voltagemeasurement

• Scopes:Theseblocksillustratethemeasurementsoftheblockstheyareconnectedto.

Figure40:Scope

• Step:Thisblockisusedtointroduceaconstantvalueinanyblock.Itisconnectedto

theinputportofthecounter-torque(Tm)withvalue0.IftheTmisleftunconnectedthesimulationwouldnotwork.

Figure41:Step

• Gain:Thisblockisusedtomultiplythesignalwiththeestablishednumber.Theoutput

signalofthemotorgivesthespeedinrad/s.Thestepblockisusedtoconvertrad/sintorpm.

Figure42:Gain

• Ground:Thisblockisusedtogroundanyelectriccircuit.Itisanessentialelementin

powergridsthatcansavelivesinthecaseofshortcircuits.

51

Figure43:Ground

• Powergui: Themajority of the abovementioned blocks fall under the category of

SimscapePowerSystemSpecializedTechnologymodels.Whentheblocks fromthiscategoryareused,thepowerguiblockmustbeincludedinthesimulation.

Figure44:Powergui

Thefinalcircuitafterconnectingalltheblockslooksasfollowing.

Figure45:Simulationcircuit

5.1 Simulationresults

Asmentionedearlier,fivedifferentsimulationsareexecuted,twoformotorno.1(squirrelcage motor) and three for motor no.2 (wound rotor motor). In order to do a visualcomparisonbetweenthe results fromtestingandsimulation, themeasurements fromtwoareplaceddownsidebysideinthefollowingtable.Alongsidetwomeasurements,thecorrespondingtestno.canbefound.

52

Testno.

Simulationresult Testresult

5

Figure46:Testno.5-Simulationresult

Figure47:Testno.5-Labtestresult

7

Figure48:Testno.7-Simulationresult

Figure49:Testno.7-Labtestresult

30

Figure50:Testno.30-Simulationresult

Figure51:Testno.30-Labtestresult

53

32

Figure52:Testno.32-Simulationresult

Figure53:Testno.32-Labtestresult

34

Figure54:Testno.34-Simulationresult

Figure55:Testno.34-Labtestresult

Table13:Simulation’swaveforms

5.1.1 Simulationresults’analysis

Thefirsttworeadings,i.e.,testno.5and7,belongtomotorno.1.Andtheotherthreereadings,i.e.,testsno.30,32and34belongtothemotorno.2.Asitcanbeobserved,thesimulationresultsareveryclosetotheonesoflaboratorytesting.Thecomparisonsofthemeasurementscanbefoundbelow.

Testno. Test Simulation5 Vp(V) 228 227 t(ms) 1060 800 I(A) 0.63 1.27 Vp(V) 200 218 t(ms) 432 500 I(A) 0.7 1.230 Vp(V) 296 295 t(ms) 600 550 I(A) 0.7 1.132 Vp(V) 248 160 t(ms) 500 300 I(A) 0.82 1.4

54

34 Vp(V) 208 120 t(ms) 160 130 I(A) 1.55 5.85

Table14:Comparisonofsimulationresultsandtestresults

Thesimulatedtransientvoltageshownintheaboveimagesissingle-phase.Ithasbeenimplementedthiswaytoensureaneasycomparisonbetweentwokindofwaveforms.Thethree-phasewaveformhasappearanceasinfigure56.

Figure56:Transientthree-phasevoltagewaveform

Thefirstthreesimulationsareveryclosetotheirreal-worldcounterpart.Thepeakvalueofthetransientvoltageisverysimilartotheonesobtainedinthetests.Theshapeofthesimulation waveforms has a similar resemblance to ones of the oscilloscope. Withexceptionofthesecondcase,thetimeoftransientwaveformalwaysfallsshort.

Thesecondsimulationhasagreaterpeakvaluethanthereal-worldtesting.Thetransientwaveformtimeisalsolongerinthesimulation.However,thewaveformlooksverymuchlikethewaveformofthetests.

Inthefourthsimulation,thereisnoovervoltagewhereasthereisclearlyanovervoltageinthelaboratorytests.Eveninthetransienttimethereisabigdifferenceof200ms.Eventheformofthewaveformisdifferent. It issafetosaythatthesimulatedwaveformismitigatedineveryaspect.

In the fifth case,during the test, theovervoltage seems tobe the resultof change inoperationalcondition.Manyfactorscontributetothemomentarilyovervoltage.Oneofthesefactorsisanabruptinterruptionoftheelectricityinacriticalmoment.Itissafetosaythatthisovervoltagewasnotproducebythecapacitivecompensation.

Eventhough,thesecondpeakoftransientwaveformhasapproximatelythesamevalueastheoneobservedintests.Itisaround100V.Andthetransientwaveformtimeisalsoclosetoitsreal-worldcounterpart.

55

The current consumed in the simulation is always superior than its real-worldcounterpart.Thereisnopatternthatcoulddescribethisaugmentationineverycurrent.Inthecaseno.34,thevalueofsimulatedcurrentisexceptionallyhigher.Thecurrentisalsoaffectedbytheovervoltage.Themagnitudeofcurrentisaugmentedin the same fashion as of voltage, resulting in an overcurrent. The waveform of thistransientcurrentisassameasthewaveformofthetransientvoltage.Thedurationofthewaveformisalsothesameasthevoltage’sduration.For instance, in figure 56, it can be seen the three-phase transient waveformcorrespondenttotestno.7.Theformofthistransientwaveformissameasthetransientvoltagewaveform.

Figure57:Transientwaveformofthethree-phasecurrent

Whensomeonetalksaboutvoltageof220V,heorshereferstotheratedvoltagewhichistheRMSvalueofvoltage.Inthefigure58,theRMSvalueofthetransientvoltagecanbe seen. Before turning the power off the RMS value of voltagewas constant. Afterinterrupting thepowersupply, theRMSvaluestarted tobehave likeawaveform.Thereasonsbehindthispatternareprobablythechangesinthepeakvalueandfrequency.

Figure58:TransientwaveformoftheRMS

Theintensificationandmitigationofthewaveformsinthesimulationinrespecttoreal-world testing can be a product of many unforeseen factors and reasons. The mostimportant thing toemphasizehere is the fact that even simulated circuits experienceovervoltageandunderthesameconditionastherealcircuits.

Havingsaidthat,itcanbeconfirmedthatthemeasurementsobservedinthelaboratorytestsarevalid.

56

6 Conclusion

After many hours of laboratory testing and simulation, it is clear that the capacitivecompensation in the motors causes overvoltage under certain conditions. Withoutcapacitivecompensationtherewasnoovervoltageproducedineithermotors.Apartfromtheeffectofload,thevoltagereceivedbythemotorandthecapacitorbankinfluenced greatly on the voltage production. The overvoltage manifested when themotorreceivedvoltagelowerthantheratedvoltageandthecapacitorbankreceivedthevoltagethatishasbeendesignedfor.TheMatlabsoftwarehasallowedthevalidationofthedatacollectedinthelaboratory.Thesimulationresultswereverysimilar to their real-worldcounterparts.Without thispowerfultool itwouldhavebeenverydifficulttoknowifthelabresultswereactuallytrueorjustsomecoincidences.Theobjectivesoutlinedintheintroductionhavebeenmet.Theovervoltageproducedbythemotorwasobservedandhowitcanbeaffectedbytheloadconnectedtothemotor’sshaft.Itwasalsoobservedthathowthevoltagereceivedbymotorandcapacitorbankscanchangethevoltageproduction.Thisdissertationhasservedthepurposeofoutliningaproblemthatexistsintheshipsandindustrialinstallations.Moreworkisneededinthisfieldinordertocollectmoredata.Thenextdissertationinthisfieldshouldalsoexperimentwithmorepowerfulinductionmotors.Thecollecteddatamayhelpthecompaniestocomeupwithasolutiontocounterattackthisphenomenon.

57

7 Bibliography

1. Casals Torrens, Pau; Bosch Tous, Ricard. Máquinas eléctricas: Aplicaciones deingenieríaeléctricaainstalacionesnavalesymarinas.Prácticas.1sted.Barcelona:UPC,2005.ISBN84-8301-813-6

2. Fraile Mora, Jesús. Máquinas Eléctricas. 2nd ed. Barcelona: Servicio depublicaciones.ColegiodeIngenierosdeCamino,CanalesyPuertos,1993.ISBN84-7493-143-6

3. Carlson,A.Bruce.Circuitos.1sted.Australia:ThomsonLearning,2001.ISBN970-686-033-9

4. Glendinning, Eric H. English in electrical Engineering and Electronics. 13th ed.Oxford:OxfordUniversityPress,1992.ISBN019437517X

5. Hayt, William H.; Kemmerly, Jack E.; Durbin, Steven M. Engineering CircuitAnalysis.8thed.NewYork:McGraw-Hill,2012.ISBN978-0-07-131706-1

6. Corrales Martín, Juan. Cálculo Industrial de Máquinas Eléctricas: Tomo I.

Barcelona:U.P.B.,1976.ISBN84-600-6752-1

7. Slemon, G.R.; Straughen, A. Electric Machines. 2nd ed. Reading- UK: Addison-WesleyPublishingCompany,1980.ISBN0-201-07730-2