文档名：电动汽车双层优化模型的充放电调度策略
摘要：传统的分时电价策略虽然一定程度上可以改善电动汽车无序充电所产生的电网日负荷峰谷差加大、负荷率降低等状况,但易产生新的负荷高峰,并且当前多目标优化等策略削峰填谷效果欠佳或用户参与度不高.针对上述问题,提出一种基于双层优化模型的调度策略以充分考虑电网和用户两侧需求.第1层模型以优化电网日负荷方差最小为目标函数;第2层优化模型建立以车主充电成本最小以及保证用户出行需求的目标函数,然后用改进的粒子群-模拟退火算法对双层优化模型进行循环迭代求解,并将第2层优化后的结果反馈给第1层,以此循环优化,输出最终结果.对比优化前后的负荷曲线,结果表明:与当前优化策略相比,所提出的基于双层优化模型的V2G调度策略能有效降低新的负荷高峰及负荷峰谷差,减少参与V2G的用户成本,实现两侧双赢.
Abstract：Againstthebackdropofcarbonpeakingandcarbonneutrality,electricvehicles(EV)havebecomeincreasinglypopularastheyusecleanerenergy,achievehigherefficiencyandhavemadebreakthroughsinenergystorage.Theypromisetoeffectivelyaddressthecurrentenergyshortageandenvironmentalpollutionproblems.However,disorderedEVchargingbringsmanychallengestothepowergrid,andgreatlyaffectsthesafetyandreliabilityofthepowergrid.Althoughthetraditionaltime-of-useelectricitypricestrategyhasimprovedthenegativeimpactsofdisorderlyEVcharging,suchastheincreaseofdailyloadpeak-valleydifferenceandthedecreaseofloadrate,itiseasytoproducenewloadpeaksandtheeffectofthecurrentmulti-objectiveoptimizationstrategyisunsatisfactory.ThecurrentVehicle-to-Grid(V2G)technologyisabletosolvethefundamentalproblem,requiringareasonableandefficientEVcharginganddischargingschedulingstrategy.Thispaperaimstoestablishamathematicalmodelofcharginganddischargingload,improvetheparticleswarmoptimization(PSO)algorithm,andstudytheorderlycharginganddischargingoptimizationstrategyanditseffect.First,basedonthenationalhouseholdtravelsurvey(NHTS)data,thisstudydeeplyanalyzestheEV'sdrivingrange,chargingstarttime,chargingendtimeandbatterystateofchargeatthebeginningofcharging,andestablishestheEVchargingloadmodel.TheEVchargingloadissimulatedandanalyzedbyMonteCarlomethod.ThesimulationresultsshowthatalargenumberofEVsareconnectedtothedistributionnetwork,andthepeak-to-valleydifferenceofthedailyloadcurveincreasessignificantly.Thedisorderlychargingloadwillhaveahugeimpactonthesafeoperationofthedistributionnetwork.Althoughtheloadcurveguidedbythetime-of-useelectricitypricehasacertainpeakclippingeffect,theeffectisnotgoodandanewloadpeakisgenerated.Second,inthecaseoflongparkingtimeandlargenumberofEVs,thePSOmethodeasilygivesrisetosuchproblemsaslocalextremum.ConsideringthecharacteristicsoftheEVbi-leveloptimizationmodelandtheadvantagesanddisadvantagesofparticleswarmandsimulatedannealing(SA)algorithm,thispaperusestheimprovedPSO-SAhybridalgorithmtosolvetheabovetwo-layermodel.ThePSO-SAalgorithmaddressestheproblemwhenthePSOalgorithmfallsintothelocaloptimumwhiletheSAalgorithmisemployedtoperturbandoptimizetheoptimalsolutionobtainedbythecurrentPSO,tryingtojumpoutofthelocaloptimalsearchforabettersolution.OurresultsshowtheimprovedPSO-SAachievesahigherefficiencyandbetteroptimizationaccuracy.Third,thispaperproposesaschedulingstrategybasedonEVbi-leveloptimizationmodelthatfullyconsiderstheneedsofboththepowergridandusers.BycontrollingthecharginganddischargingpowerofEVsindifferenttimeperiodswithinaday,thispaperconsidersvariousconstraintssuchastransformerandEVcharginganddischargingpower,andusestheimprovedPSO-SAalgorithmtooptimizethefirstlayermodeltoobtainthedailyloadcurveconsideringonlythedemandofthegridside.Takingthefirst-stageoptimizationresultsasconstraints,thispaperobtainsthesecond-leveloptimizationstrategybasedonuser-siderequirements.Takingtheminimumcharginganddischargingcostoftheownerastheoptimizationobjectiveandconsideringthevariousconstraintsofthelayermodel,thispaperobtainstheoptimizationresultsofthesecondlayermodel.ThecharginganddischargingpoweroptimizationresultsofEVineachtimeperiodobtainedbythesecondlayermodelarefedbacktotheupperlayerforthenextcycle.Theupperandlowermodelsiteraterepeatedlyuntiltheresultsmeettheterminationconditions.Finally,theoptimalsolutionofEVoptimalschedulingbasedonthetwo-layermodelisobtained.Comparedwiththeloadcurvesbeforeandafteroptimization,ourresultsshowtheproposedV2Gschedulingstrategybasedonthebi-leveloptimizationmodeleffectivelyreducestheloadpeak-valleydifferences,increasestheloadrateandreducesthepowercostsforEVowners.
作者：马永翔 王希鑫 闫群民 孔志战 淡文国 Author：MAYongxiang WANGXixin YANQunmin KONGZhizhan DANWenguo
作者单位：陕西理工大学电气工程学院,陕西汉中723001陕西省电力有限公司,西安710048乌兰察布电业局,内蒙古乌兰察布012000
刊名：重庆理工大学学报
Journal：JournalofChongqingInstituteofTechnology
年，卷(期)：2024, 38(3)
分类号：TM73
关键词：电动汽车 V2G技术 充放电优化调度 双层优化模型 改进粒子群-模拟退火算法
Keywords：electricvehicles V2G chargeanddischargeoptimizationscheduling bi-leveloptimizationmodel improvedparticleswarm-simulatedannealingalgorithm
机标分类号：TM73TP391U491
在线出版日期：2024年3月25日
基金项目：国家自然科学基金，陕西省教育厅重点科学研究计划项目电动汽车双层优化模型的充放电调度策略[
期刊论文] 重庆理工大学学报--2024, 38(3)马永翔 王希鑫 闫群民 孔志战 淡文国传统的分时电价策略虽然一定程度上可以改善电动汽车无序充电所产生的电网日负荷峰谷差加大、负荷率降低等状况,但易产生新的负荷高峰,并且当前多目标优化等策略削峰填谷效果欠佳或用户参与度不高.针对上述问题,提出一种...参考文献和引证文献
参考文献
引证文献
本文读者也读过
相似文献
相关博文
电动汽车双层优化模型的充放电调度策略 Charge and discharge scheduling strategies for electric vehicle double-layer optimization models
电动汽车双层优化模型的充放电调度策略.pdf
- 文件大小:
- 1.38 MB
- 下载次数:
- 60
-
高速下载
|
|
