基于KPCA和NSGAⅡ优化CNN参数的电动汽车充电站短期负荷预测
|
摘要
为提升电动汽车充电站短期负荷预测的效率和精度,提出了基于核主成分分析(kernel principal component analysis,KPCA)和非劣排序遗传算法II(non-dominated sorting genetic algorithm II,NSGAII)优化卷积神经网络(convolutional neural network,CNN)的充电站短期负荷预测方法。应用KPCA对模型输入变量进行降噪处理,简化了网络结构,加快了预测速度;通过多次负荷预测测试比较误差的方式确定卷积神经网络模型中卷积层和子采样层的最佳神经元个数,保证了预测方法的准确性;利用NSGAII对卷积神经网络的参数进行优化,提高了预测方法的运算速度和预测精度。通过算例分析以及和其他方法的对比,验证了文中方法具有较高的效率和精度。
In order to improve the short-term load forecasting efficiency and precision of electric vehicle charging station,this paper proposes a short-term load forecasting method for charging station based on kernel principal component analysis( KPCA) and non-dominated sorting genetic algorithm Ⅱ( NSGAⅡ). The KPCA is used to reduce the noise of the model input variables,which simplifies the network structure and accelerates the prediction speed. Through the comparison of the load forecasting error to define the convolutional neural network( CNN) model in convolution layers and sub sampling the top layer neurons number,the accuracy of the model is guaranteed. By using the NSGAⅡ to optimize the parameters of the CNN,the operation speed and precision of the prediction method are improved. Through example analysis and comparison with other methods,it is proved that the method has high efficiency and precision.
In order to improve the short-term load forecasting efficiency and precision of electric vehicle charging station,this paper proposes a short-term load forecasting method for charging station based on kernel principal component analysis( KPCA) and non-dominated sorting genetic algorithm Ⅱ( NSGAⅡ). The KPCA is used to reduce the noise of the model input variables,which simplifies the network structure and accelerates the prediction speed. Through the comparison of the load forecasting error to define the convolutional neural network( CNN) model in convolution layers and sub sampling the top layer neurons number,the accuracy of the model is guaranteed. By using the NSGAⅡ to optimize the parameters of the CNN,the operation speed and precision of the prediction method are improved. Through example analysis and comparison with other methods,it is proved that the method has high efficiency and precision.
引文
[1]刘满平.新电改方案的核心、着力点及影响[J].宏观经济管理,2015(6):20-22.LIU M anping.The core,the focus and the impact of the new electric pow er reform program[J].M acroeconomic M anagement,2015(6):20-22.
[2]曾鸣,张晓春,王丽华.以能源互联网思维推动能源供给侧改革[J].电力建设,2016,37(4):10-15.ZENG M ing,ZHANG Xiaochun,WANG Lihua.Energy supply side reform promoting based on energy internet thinking[J].Electric Pow er Construction,2016,37(4):10-15.
[3]张艳娟,苏小林,闫晓霞,等.基于电动汽车时空特性的充电负荷预测[J].电力建设,2015,36(7):75-82.ZHANG Yanjuan,SU Xiaolin,YAN Xiaoxia,et al.A method of charging load forecast based on electric vehicle time-space characteristics[J].Electric Pow er Construction,2015,36(7):75-82.
[4]杨少兵,吴命利,姜久春,等.电动汽车充电站负荷建模方法[J].电网技术,2013,37(5):1190-1195.YANG Shaobing,WU M ingli,JIANG Jiuchun,et al.An Approach for load modeling of electric vehicle charging station[J].Pow er System Technology,2013,37(5):1190-1195.
[5]常德政,任杰,赵建伟,等.基于RBF-NN的电动汽车充电站短期负荷预测研究[J].青岛大学学报(工程技术版),2014,29(4):44-48.CHANG Dezheng,REN Jie,ZHAO Jianw ei,et al.Study on short term load forecasting of electric vehicle charging station based on RBF-NN[J].Journal of Qiingdao University(Engineering and Technology Edition),2014,29(4):44-48.
[6]常德政.智能电网中电动汽车充电站短期负荷预测模型研究[D].青岛:青岛大学,2015.CHANG Dezheng.Short-term load forecasting of electric vehicle charging station in the smart grid[D].Qiingdao:Qiingdao University,2015
[7]DAI Qian,CAI Tao,DUAN Shanxu,et al.Stochastic modeling and forecasting of load demand for electric bus battery-sw ap station[J].IEEE Transactions on Power Delivery,2014,29(4):1909-1917.
[8]徐晓波.电动汽车充/换电站短期负荷预测方法研究[D].北京:华北电力大学,2015.XU Xiaobo.Study on short-term load forecasting method for electric vehicle charging/exchange pow er station[D].Beijing:North China Electric Pow er University,2015
[9]张维戈,颉飞翔,黄梅,等.快换式公交充电站短期负荷预测方法的研究[J].电力系统保护与控制,2013,41(4):61-66.ZHANG Weige,XIE Feixiang,HUANG M ei,et al.Research on short-term load forecasting methods of electric buses charging station[J].Power System Protection and Control,2013,41(4):61-66.
[10]刘文霞,龙日尚,徐晓波,等.考虑数据新鲜度和交叉熵的电动汽车短期充电负荷预测模型[J].电力系统自动化,2016,40(12):45-52.LIU Wenxia,LONG Rishang,XU Xiaobo,et al.Forecasting model of short-term EV charging load based on data freshness and cross entropy[J].Automation of Electric Pow er Systems,2016,40(12):45-52.
[11]ASHTARI A,BIBEAU E,SHAHIDINEJAD S,et al.PEV charging profile prediction and analysis based on vehicle usage data[J].IEEE Transactions on Smart Grid,2012,3(1):341-350.
[12]黄小庆,陈颉,陈永新,等.大数据背景下的充电站负荷预测方法[J].电力系统自动化,2016,40(12):68-74.HUANG Xiaoqing,CHEN Jie,CHEN Yongxin,et al.Load forecasting method for electric vehicle charging station based on big data[J].Automation of Electric Pow er Systems,2016,40(12):68-74.
[13]刘畅,刘天琪,陈振寰,等.基于KPCA和BP神经网络的短期负荷预测[J].电测与仪表,2016,53(10):57-61.LIU Chang,LIU Tianqi,CHEN Zhenhuan,et al.Short-term load forecasting based on KPCA and BP neural netw ork[J].Electrical M easurement&Instrumentation,2016,53(10):57-61.
[14]王茜,张粒子.采用NSGA-Ⅱ混合智能算法的风电场多目标电网规划[J].中国电机工程学报,2011,31(19):17-24.WANG Qian,ZHANG Lizi.M ulti objective grid planning of w ind farm using NSGA-II hybrid intelligent algorithm[J].Proceedings of the CSEE,2011,31(19):17-24.
[15]贠汝安,董增川,王好芳.基于NSGA2的水库多目标优化[J].山东大学学报(工学版),2010,40(6):124-128.YUN Ru’an,DONG Zengchuan,WANG Haofang.M ultiobjective optimization of a reservoirbased on NSGA2[J].Journal of Shandong University(Engineering Science),2010,40(6):124-128.
[16]孙建龙,吴锁平,陈燕超.基于改进NSGA2算法的配电网分布式电源优化配置[J].电力建设,2014,35(2):86-90.Sun Jianlong,Wu Suoping,Chen Yanchao.Optimal configuration of distributed generation in distribution netw ork based on improved NSGA2[J].Electric Pow er Construction,2014,35(2):86-90.
[17]李思琴,林磊,孙承杰.基于卷积神经网络的搜索广告点击率预测[J].智能计算机与应用,2015,5(5):22-25,28.LI Siqin,LIN Lin,SUN Chengjie.Search advertisement click through rate prediction based on convolutional neural netw ork[J].Intelligent Computer and Applications,2015,5(5):22-25,28.
[18]丛瑜,肖怀铁,付强.基于核主分量分析的高分辨雷达目标特征提取与识别[J].电光与控制,2008,15(2):31-35,38.CONG Yu,XIAO Huaitie,FU Qiang.Target feature extraction and recognition for high-range recognition radar based on the KPCA method[J].Electronics Optics&Control,2008,15(2):31-35,38.
[19]杨冰,王丽芳,廖承林,等.不确定充电习惯对电动汽车充电负荷需求及充电负荷调节的影响[J].电工技术学报,2015,30(4):226-232.YANG Bing,WANG Lifang,LIAO Chenglin,et al.The influence of uncertain charging habits on electric vehicle charging load demand and charging load regulation[J].Transactions of China Electrotechnical Society,2015,30(4):226-232.
[20]王哲,代兵琪,李相栋.基于PSO-SNN的电动汽车充电站短期负荷预测模型研究[J].电气技术,2016(1):46-50.WANG Zhe,DAI Bingqi,Li Xiangdong.Research on short term load forecasting model of electric vehicle charging station based on PSO-SNN[J].Electrical Engineering,2016(1):46-50.
[21]郑牡丹.基于云计算的充电站充电负荷预测体系结构研究[D].北京:华北电力大学,2015.ZHENG M udan.Research on the structure of charging load forecasting system based on cloud computing[D].Beijing:North China Electric Pow er University,2015.
[22]牛东晓,曹树华,赵磊.电力负荷预测技术及其应用[M].北京:中国电力出版社,1998.
[2]曾鸣,张晓春,王丽华.以能源互联网思维推动能源供给侧改革[J].电力建设,2016,37(4):10-15.ZENG M ing,ZHANG Xiaochun,WANG Lihua.Energy supply side reform promoting based on energy internet thinking[J].Electric Pow er Construction,2016,37(4):10-15.
[3]张艳娟,苏小林,闫晓霞,等.基于电动汽车时空特性的充电负荷预测[J].电力建设,2015,36(7):75-82.ZHANG Yanjuan,SU Xiaolin,YAN Xiaoxia,et al.A method of charging load forecast based on electric vehicle time-space characteristics[J].Electric Pow er Construction,2015,36(7):75-82.
[4]杨少兵,吴命利,姜久春,等.电动汽车充电站负荷建模方法[J].电网技术,2013,37(5):1190-1195.YANG Shaobing,WU M ingli,JIANG Jiuchun,et al.An Approach for load modeling of electric vehicle charging station[J].Pow er System Technology,2013,37(5):1190-1195.
[5]常德政,任杰,赵建伟,等.基于RBF-NN的电动汽车充电站短期负荷预测研究[J].青岛大学学报(工程技术版),2014,29(4):44-48.CHANG Dezheng,REN Jie,ZHAO Jianw ei,et al.Study on short term load forecasting of electric vehicle charging station based on RBF-NN[J].Journal of Qiingdao University(Engineering and Technology Edition),2014,29(4):44-48.
[6]常德政.智能电网中电动汽车充电站短期负荷预测模型研究[D].青岛:青岛大学,2015.CHANG Dezheng.Short-term load forecasting of electric vehicle charging station in the smart grid[D].Qiingdao:Qiingdao University,2015
[7]DAI Qian,CAI Tao,DUAN Shanxu,et al.Stochastic modeling and forecasting of load demand for electric bus battery-sw ap station[J].IEEE Transactions on Power Delivery,2014,29(4):1909-1917.
[8]徐晓波.电动汽车充/换电站短期负荷预测方法研究[D].北京:华北电力大学,2015.XU Xiaobo.Study on short-term load forecasting method for electric vehicle charging/exchange pow er station[D].Beijing:North China Electric Pow er University,2015
[9]张维戈,颉飞翔,黄梅,等.快换式公交充电站短期负荷预测方法的研究[J].电力系统保护与控制,2013,41(4):61-66.ZHANG Weige,XIE Feixiang,HUANG M ei,et al.Research on short-term load forecasting methods of electric buses charging station[J].Power System Protection and Control,2013,41(4):61-66.
[10]刘文霞,龙日尚,徐晓波,等.考虑数据新鲜度和交叉熵的电动汽车短期充电负荷预测模型[J].电力系统自动化,2016,40(12):45-52.LIU Wenxia,LONG Rishang,XU Xiaobo,et al.Forecasting model of short-term EV charging load based on data freshness and cross entropy[J].Automation of Electric Pow er Systems,2016,40(12):45-52.
[11]ASHTARI A,BIBEAU E,SHAHIDINEJAD S,et al.PEV charging profile prediction and analysis based on vehicle usage data[J].IEEE Transactions on Smart Grid,2012,3(1):341-350.
[12]黄小庆,陈颉,陈永新,等.大数据背景下的充电站负荷预测方法[J].电力系统自动化,2016,40(12):68-74.HUANG Xiaoqing,CHEN Jie,CHEN Yongxin,et al.Load forecasting method for electric vehicle charging station based on big data[J].Automation of Electric Pow er Systems,2016,40(12):68-74.
[13]刘畅,刘天琪,陈振寰,等.基于KPCA和BP神经网络的短期负荷预测[J].电测与仪表,2016,53(10):57-61.LIU Chang,LIU Tianqi,CHEN Zhenhuan,et al.Short-term load forecasting based on KPCA and BP neural netw ork[J].Electrical M easurement&Instrumentation,2016,53(10):57-61.
[14]王茜,张粒子.采用NSGA-Ⅱ混合智能算法的风电场多目标电网规划[J].中国电机工程学报,2011,31(19):17-24.WANG Qian,ZHANG Lizi.M ulti objective grid planning of w ind farm using NSGA-II hybrid intelligent algorithm[J].Proceedings of the CSEE,2011,31(19):17-24.
[15]贠汝安,董增川,王好芳.基于NSGA2的水库多目标优化[J].山东大学学报(工学版),2010,40(6):124-128.YUN Ru’an,DONG Zengchuan,WANG Haofang.M ultiobjective optimization of a reservoirbased on NSGA2[J].Journal of Shandong University(Engineering Science),2010,40(6):124-128.
[16]孙建龙,吴锁平,陈燕超.基于改进NSGA2算法的配电网分布式电源优化配置[J].电力建设,2014,35(2):86-90.Sun Jianlong,Wu Suoping,Chen Yanchao.Optimal configuration of distributed generation in distribution netw ork based on improved NSGA2[J].Electric Pow er Construction,2014,35(2):86-90.
[17]李思琴,林磊,孙承杰.基于卷积神经网络的搜索广告点击率预测[J].智能计算机与应用,2015,5(5):22-25,28.LI Siqin,LIN Lin,SUN Chengjie.Search advertisement click through rate prediction based on convolutional neural netw ork[J].Intelligent Computer and Applications,2015,5(5):22-25,28.
[18]丛瑜,肖怀铁,付强.基于核主分量分析的高分辨雷达目标特征提取与识别[J].电光与控制,2008,15(2):31-35,38.CONG Yu,XIAO Huaitie,FU Qiang.Target feature extraction and recognition for high-range recognition radar based on the KPCA method[J].Electronics Optics&Control,2008,15(2):31-35,38.
[19]杨冰,王丽芳,廖承林,等.不确定充电习惯对电动汽车充电负荷需求及充电负荷调节的影响[J].电工技术学报,2015,30(4):226-232.YANG Bing,WANG Lifang,LIAO Chenglin,et al.The influence of uncertain charging habits on electric vehicle charging load demand and charging load regulation[J].Transactions of China Electrotechnical Society,2015,30(4):226-232.
[20]王哲,代兵琪,李相栋.基于PSO-SNN的电动汽车充电站短期负荷预测模型研究[J].电气技术,2016(1):46-50.WANG Zhe,DAI Bingqi,Li Xiangdong.Research on short term load forecasting model of electric vehicle charging station based on PSO-SNN[J].Electrical Engineering,2016(1):46-50.
[21]郑牡丹.基于云计算的充电站充电负荷预测体系结构研究[D].北京:华北电力大学,2015.ZHENG M udan.Research on the structure of charging load forecasting system based on cloud computing[D].Beijing:North China Electric Pow er University,2015.
[22]牛东晓,曹树华,赵磊.电力负荷预测技术及其应用[M].北京:中国电力出版社,1998.