OPTIMALSIZINGOFDISTRIBUTEDGENERATIONSINMICROGRIDSYSTEM �

SchoolofElectricalengineering,WuhanUniversity

Prof.DENGChanghong

Team:(HEJun,XUQiushi,LIUCuilin,PANHua)

IntroducGon � 0�

TheOpGmalsizingmodelofDistributedGeneraGoninMicro‐gridsystem �

1�

conclusion � 4�

SimulaGonexample � 3�

Method:ProbabilisGcMethod � 2�

0IntroducBon�

worldenergy‐relatedcarbondioxideemissionsrisefrom30.2billionmetrictonsin2008to35.2billionmetrictonsin2020and43.2billionmetrictonsin2035‐‐‐‐43%1.Thelow‐carbonpowertechnologieshavearousedwideconcerned.

Crisis:carbondioxideemissions�

1.U.S.EnergyInformaGonAdministraGon.InternaGonalEnergyOutlook2011[R],2011.�

WhatcanMicrogriddo?‐‐‐‐�weconcern:1.HowcanMicrogridefficientlyuGlizewind,solarenergy?2.Howcanweprovideelectricalpowerwithhighreliability?3.Howcanuseraffordexpensiverenewableenergy?

1TheOpBmalsizingmodelofDistributedGeneraBoninMicro‐gridsystem�

(1)Objective Function ThispaperestablishedobjecGvefuncGonincludinglow‐carbonformsofenergyandhigh‐carbonenergy,setobjectofminimizingthefulllifecyclecosts,describedbelow.：�

fulllifecyclecosts�

Maintenancecosts�

IniGalinvestmentcosts�

Discountrate�

CarbonPunishment �

(J–Lifecycleform)�

(2)Constraints�

Outputofpower �RandomnessofthelocalweatherresourcesaffectrenewableenergyuGlizaGonefficiency,anditsoutputisodenlessthanthecapacityoftheunits �

Reliability�

Outputofunits �

Outputofasingleunit �

Numberofonetypeofunits �

Reliabilityofthesystem

8760h�

(3)Problem�

duetowindandsolarrandomlychangewithweather �

Itishardtocalculate:1.  electricquanGtythatproducedbyeachgeneraGon2.  reliability

TosimulatetheprobabilisGcproducGon:

Method:ProbabilisGcMethod

2.Method‐‐‐‐ProbabilisBcMethod�

2.Method‐‐‐‐ProbabilisBcMethod

AccordingtotheloadandgeneraGonoutputcurve，PowerShortageiscalculatedasfollows:�

Loadcurve�

OutputofgeneraGonsystemintegratedwithwindfarmandphotovoltaicsystem�

OutputoftradiGonunits �

tocalculate:1.  electricpowerthatproducedbyeachgenerator2.  Fuelcosts3.  reliability

2.Method‐‐‐‐ProbabilisBcMethod

IfthestateofchargecannotaffordPowerShortage，theloadwillbelost. �

accumulatethelostloadhours�

Reliability：

Chargeofstorageenergy�

PowerShortage�

2.Method‐‐‐‐ProbabilisBcMethod

BasedonMonteCarlomethod，calculatetheabovevalue：�

MonteCarlo： �

1．AccordingtoprobabilitydensityfuncGonψ(x)togeneraterandomvariablex,andcalculatetheoutputf(x)

2．accumulatef(x) at every turn ，and averageit.�

3.SimulaBonexample�

month 1 2 3 4 5 6

Windspeed（m/s） 9.4 7.2 7.0 6.0 7.3 7.3

month 7 8 9 10 11 12

Windspeed（m/s） 7.0 5.4 5.8 9.2 6.4 7.9

Table 1 Data of wind speed

month 1 2 3 4 5 6

SolarradiaBon(kWh/m2/d） 2.993 3.323 3.798 5.279 5.825 5.113

month 7 8 9 10 11 12

SolarradiaBon(kWh/m2/d） 5.495 5.083 5.706 5.111 3.626 3.449

Table 2 Data of solar radiation

Takeanindependentislandmicro‐gridatChinaasanexample.themaxmumofthesystemloadis7.5MW�

Natureresourcesofthisplacesisasfollowings:�

Case1: Limit the capacity of tradition units �

LOLE（h） Wind photovoltaic Storage energy TradiGonunits Virtualunits

9.15 0 0 0 7Mw 1.4Mw

9.22 0 0 0 8.9Mw 0

9.28 9.1Mw 0 0.95Mwh 7Mw 0

9.24 4Mw 2.1Mw 0.95Mwh 7Mw 0

9.18 0 11.35Mw 0.95Mwh 7Mw 0

Table 3. Optimization results with approximate LOLE(LossOfLoadExpectaGon)

When 7Mw Tradition units is set, 1.4Mw Virtual units Or 1.9Mw Tradition units to grid can achieve the equal reliability. The demand of credit capacity of combinedpowergeneraGonsystemofwindturbineandphotovoltaicis1.4Mw.

credit capacity of combinedpowergeneraBons

set the credit capacity of combinedpowergeneraGon.CalculatealltheconfiguraGon�

Forexample,whencapacityofstorageenergyis1.2MWh、0.95MWh、0.85MWh，thecapacityofwindturbineandPVwillbeasFollowingfigure:�

different capacity of tradition units with sameLOLE

0�

10�

20�

30�

40�

50�

60�

70�

80�

90�

100�

4.5� 5.5� 6.2� 6.5� 6.8� 7� 7.2� 8�

capa

city(M

W)�

capacityoftradiBonalunits(MW)�

traditional units� wind capacity�

pv capacity� energy storage�

Different traditional units-wind-PV-Storage configuration is as following

Case 2 can be seen composed by lots of case 1, then search the optimal point. and set power dispatching sequence as: wind-solar-energy storage-traditional units, We can see, when take 6.8 traditional units, the result is optimal.�

Case3: Take the fuel price trend into concern�

in the future,cost of renewable resources will reduce, the fuel price will increase, operaGoncostswillbeasfollowing:�

Takethelowcarbonscenarioforexample�

Inthelowcarbonscenario,thetrendsofopGmalconfiguraGonareasfollowingFig. �

Traditional units-wind-PV -Storage configuration Capacity �

0�

10�

20�

30�

40�

50�

60�

70�

80�

90�

100�

2011� 2015� 2020� 2025� 2030�

capa

city(M

W)�

capacityoftradiBonalunits(MW)�

traditional units� wind capacity� pv capacity� energy storage�

4.conclusion�

(1).Obtainingthecrediblecapacityofwind‐solar‐banerypowersystemisveryimportant,aprobabilisGcmethodisproposed.

(2).Thismethodcouldconsidertheuncertaintyfactors,includetherandomnesschangesofunit’soutputcausedbyfluctuaGonsofwindspeedandilluminaGonintensity.

(3)ThisproposedopGmalsizingofhybridwind‐solar‐banerypowerwasselectedsothattheenGrelifecycleofthetotalinvestmentcostlowest.