基于神经网络的光伏阵列多峰MPPT的研究
|
摘要
为了减少神经网络训练数据的数量,根据局部阴影条件下光伏阵列的输出特性,提出基本阴影遮挡类型概念,使得神经网络仅需要训练少量数据,就可以准确地预测最大功率点电压。首先,通过实际光伏阵列数据测试仅训练基本阴影遮挡类型的BP神经网络对最大功率点电压的跟踪效果。然后,搭建光伏发电MPPT仿真系统,对比扰动法、固定电压法和BP神经网络结合扰动法在阴影类型、光照强度和温度三方面变化时对MPP的跟踪效果。最后,通过分析表明,经过基本阴影遮挡类型训练的BP神经网络结合扰动法能够有效地跟踪最大功率点,即基本阴影遮挡类型能够减少神经网络跟踪多峰MPP的训练数据获取量。
For reducing the number of neural network training data,according to the output characteristics of the photovoltaic array under partial shadow conditions,the concept is proposed about basic shadow occlusion type,so that the neural network only needs to be trained with a small amount of data and the voltage of MPP can be accurately predicted. Firstly,the actual PV array data is used to test the effect of BP neural network training on the tracking of the voltage of MPP only with the basic shadow shading type. Then,built the MPPT simulation system of photovoltaic power generation system and compared the perturbation method,fixed voltage method and BP neural network combined with the perturbation method to track the MPP. Finally,the analysis shows that the BP neural network combined with the perturbation method trained by the basic shadow occlusion type can effectively track the MPP,that is,the basic shadow occlusion type can reduce the acquisition of training data when the neural network tracks multi-peak MPPT.
For reducing the number of neural network training data,according to the output characteristics of the photovoltaic array under partial shadow conditions,the concept is proposed about basic shadow occlusion type,so that the neural network only needs to be trained with a small amount of data and the voltage of MPP can be accurately predicted. Firstly,the actual PV array data is used to test the effect of BP neural network training on the tracking of the voltage of MPP only with the basic shadow shading type. Then,built the MPPT simulation system of photovoltaic power generation system and compared the perturbation method,fixed voltage method and BP neural network combined with the perturbation method to track the MPP. Finally,the analysis shows that the BP neural network combined with the perturbation method trained by the basic shadow occlusion type can effectively track the MPP,that is,the basic shadow occlusion type can reduce the acquisition of training data when the neural network tracks multi-peak MPPT.
引文
[1] Liu Yihua,Chen Jinghsiao,Huang Jiawei. A review of maximum power point tracking techniques for use in partially shaded conditions[J]. Renewable&Sustainable Energy Reviews,2015,41(41):436-453.
[2] Liu Yihua,Liu Chunliang,Huang Jiawei,et al. Neural-network-based maximum power point tracking methods for photovoltaic systems operating under fast changing environments[J]. Solar Energy,2013,89(2):42-53.
[3] Bahgat ABG,Helwa NH,Ahmad GE,et al. maximum power point tracking controller for PV systems using neural networks[J]. Renewable Energy,2014,30(8):1257-1268.
[4]梁惺彦,张蔚.基于BP神经网络光伏发电系统MPPT的研究[J].制造业自动化,2012,34(24):112-114,130.Liang Xingyan,Zhang Wei. The research of maximum power point tracking in photovoltaic power system based on back propagation networks[J]. Manufacturing Automation,2012,34(24):112-114,130.
[5]李腾飞.复杂光照条件下光伏发电系统输出特性及最大功率点跟踪研究[D].太原:太原理工大学,2014.Li Tengfei. Study on the output characteristics and MPPT of PV system under complex illumination[D]. Taiyuan:Taiyuan University of Technology,2014.
[6]胥芳,张任,吴乐彬,等.自适应BP神经网络在光伏MPPT中的应用[J].太阳能学报,2012,33(3):468-472.Xu Fang,Zhang Ren,Wu Lebin,et al. Self-adaption BP neural network for the maximum-power-point tracking in photovoltaic[J]. Acta Energiae Solaris Sinica,2012,33(3):468-472.
[7]崔岩,白静晶.光伏阵列多峰最大功率点跟踪研究[J].电机与控制学报,2012,16(6):87-91.Cui Yan,Bai Jingjing. Research on multi-peak PV module maximum power point tracking[J]. Electric Machines and Control,2012,16(6):87-91.
[8]孟彦京,徐晓龙.光伏电池数学模型的LabVIEW仿真分析[J].电测与仪表,2015,52(2):38-42.Meng Yanjing,Xu Xiaolong. Simulation and analysis and of the photovoltaic cell mathematical model based on LabVIEW[J]. Electrical Measurement&Instrumentation,2015,52(2):38-42.
[9]朱正伟,郭枫,孙广辉,等.基于RBF-BP神经网络的光伏MPPT研究[J].计算机仿真,2015,32(2):131-134,160.Zhu Zhengwei,Guo Feng,Sun Guanghui,et al. Research on photovoltaic maximum power point tracking based on RBF-BP neural network[J].Computer Simulation,2015,32(2):131-134,160.
[10]林虹江,周步祥,冉伊,等.基于遗传优化BP神经网络算法的光伏系统最大功率点跟踪研究[J].电测与仪表,2015,52(5):35-40.Lin Hongjiang,Zhou Buxiang,Ran Yi,et al. Research on the maximum power point tracking of the photovoltaic system based on the genetic optimization BP neural network algorithm[J]. Electrical Measurement&Instrumentation,2015,52(5):35-40.
[11]闫超,倪福佳,刘嘉瑜,等.基于改进FOA优化BP神经网络算法的光伏系统MPPT研究[J].电测与仪表,2018,55(8):24-29,130.Yan Chao,Ni Fujia,Liu Jiayu,et al. Research of the photovoltaic system MPPT based on improved IFOA-BP neural network algorithm[J]. Electrical Measurement&Instrumentation,2018,55(8):24-29,130.
[2] Liu Yihua,Liu Chunliang,Huang Jiawei,et al. Neural-network-based maximum power point tracking methods for photovoltaic systems operating under fast changing environments[J]. Solar Energy,2013,89(2):42-53.
[3] Bahgat ABG,Helwa NH,Ahmad GE,et al. maximum power point tracking controller for PV systems using neural networks[J]. Renewable Energy,2014,30(8):1257-1268.
[4]梁惺彦,张蔚.基于BP神经网络光伏发电系统MPPT的研究[J].制造业自动化,2012,34(24):112-114,130.Liang Xingyan,Zhang Wei. The research of maximum power point tracking in photovoltaic power system based on back propagation networks[J]. Manufacturing Automation,2012,34(24):112-114,130.
[5]李腾飞.复杂光照条件下光伏发电系统输出特性及最大功率点跟踪研究[D].太原:太原理工大学,2014.Li Tengfei. Study on the output characteristics and MPPT of PV system under complex illumination[D]. Taiyuan:Taiyuan University of Technology,2014.
[6]胥芳,张任,吴乐彬,等.自适应BP神经网络在光伏MPPT中的应用[J].太阳能学报,2012,33(3):468-472.Xu Fang,Zhang Ren,Wu Lebin,et al. Self-adaption BP neural network for the maximum-power-point tracking in photovoltaic[J]. Acta Energiae Solaris Sinica,2012,33(3):468-472.
[7]崔岩,白静晶.光伏阵列多峰最大功率点跟踪研究[J].电机与控制学报,2012,16(6):87-91.Cui Yan,Bai Jingjing. Research on multi-peak PV module maximum power point tracking[J]. Electric Machines and Control,2012,16(6):87-91.
[8]孟彦京,徐晓龙.光伏电池数学模型的LabVIEW仿真分析[J].电测与仪表,2015,52(2):38-42.Meng Yanjing,Xu Xiaolong. Simulation and analysis and of the photovoltaic cell mathematical model based on LabVIEW[J]. Electrical Measurement&Instrumentation,2015,52(2):38-42.
[9]朱正伟,郭枫,孙广辉,等.基于RBF-BP神经网络的光伏MPPT研究[J].计算机仿真,2015,32(2):131-134,160.Zhu Zhengwei,Guo Feng,Sun Guanghui,et al. Research on photovoltaic maximum power point tracking based on RBF-BP neural network[J].Computer Simulation,2015,32(2):131-134,160.
[10]林虹江,周步祥,冉伊,等.基于遗传优化BP神经网络算法的光伏系统最大功率点跟踪研究[J].电测与仪表,2015,52(5):35-40.Lin Hongjiang,Zhou Buxiang,Ran Yi,et al. Research on the maximum power point tracking of the photovoltaic system based on the genetic optimization BP neural network algorithm[J]. Electrical Measurement&Instrumentation,2015,52(5):35-40.
[11]闫超,倪福佳,刘嘉瑜,等.基于改进FOA优化BP神经网络算法的光伏系统MPPT研究[J].电测与仪表,2018,55(8):24-29,130.Yan Chao,Ni Fujia,Liu Jiayu,et al. Research of the photovoltaic system MPPT based on improved IFOA-BP neural network algorithm[J]. Electrical Measurement&Instrumentation,2018,55(8):24-29,130.