资源描述：
《基于改进foa优化bp神经网络算法的》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。
1、第44卷第504期电测与仪表Vol.44No.5042017年第12期ElectricalMeasurement&InstrumentationDec.2017基于改进FOA优化BP神经网络算法的光伏系统MPPT研究*闫超1,2，倪福佳1，刘嘉瑜1,2，贺诗明2，高振远2，王少帅1（1.中国矿业大学江苏省煤矿电气与自动化工程实验室，江苏徐州221116；2.中国矿业大学电气与动力工程学院，江苏徐州221116）摘要：针对基于BP神经网络的光伏系统MPPT策略在光照强度突变时存在较大误差的问题，本文提出了一种改进的果蝇优化算
2、法用于BP神经网络的权值和阈值优化，并建立了基于IFOA-BP神经网络算法的光伏系统MPPT控制的仿真模型。测试和仿真结果表明，IFOA的收敛速度和求解精度较改进前均有明显提升；IFOA优化后的BP神经网络收敛速度加快，预测误差减少；较之于电导增量法，IFOA-BP神经网络的MPPT策略在稳态条件下能明显抑制功率波动，在外界条件发生突变时，能迅速准确地追踪到最大功率点，具有良好的稳态精度和动态特性。关键词：光伏电池；最大功率点跟踪；BP神经网络；改进果蝇优化算法中图分类号：TM933文献标识码：B文章编号：1001-139
3、0（2018）00-0000-00ResearchofonthephotovoltaicsystemMPPTbasedonimprovedIFOA-BPneuralnetworkalgorithmYanChao1,2,NiFujia1,LiuJiayu1,2,HeShiming1,GaoZhenyuan1,WangShaoshuai1(1.JiangsuProvinceLaboratoryofElectricalandAutomationEngineeringforCoalMining,ChinaUniversityofM
4、ining&Technology,Xuzhou221116,Jiangsu,China.2.SchoolofElectricalandPowerEngineering,ChinaUniversityofMining&Technology,Xuzhou221116,Jiangsu,China)Abstract：:WhentheBPneuralnetworkisadoptedtopredictthevoltageatthemaximumpowerpoint,thereisabigerrorifthelightintensity
5、changesdrastically.Aimingatthisproblem,anovelimprovedfruitflyoptimizationalgorithm(IFOA)determiningtheoptimalBPneuralnetworkparameters(weightandthreshold)isproposed,andasimulationmodelofthephotovoltaicsystemMPPTcontrolstrategybasedontheIFOA-BPneuralnetworkalgorith
6、mwasisestablished.Thetestandsimulationresultsshowthatthe,IFOAhasagreatadvantageinconvergencesearchspeedandsolutionaccuracythanFOA;IFOA-BPneuralnetworkcaneffectivelyincreasestheconvergencespeedandreducesthepredictionerror;.comparedComparedwiththeincrementalconducta
7、nce(INC)method,theproposedphotovoltaicsystemMPPTcontrolalgorithmbasedonIFOA-BPneuralnetworkcouldsuppresstheoscillationaroundthemaximumpowerpoint(MPP)understeady-stateconditionsandtrackdowntheMPPquicklyandaccuratelywhenlightintensityandtemperaturechangedrastically,
8、whichverifiesthestability,precisionandrapidityoftheproposedMPPTmethod.Keywords：:photovoltaiccell,maximumpowerpointtracking,BPneuralnetwork,improvedfruit
