基于卷积神经网络的光伏系统直流串联电弧故障检测
|
摘要
本文提出一种新颖的基于卷积神经网络的光伏系统直流串联电弧故障检测方法。首先采用短时傅里叶变换提取电流信号的时频信息,以能量谱密度作为电流的时频联合能量函数,构造电流的时频谱图,然后以时频谱图中各时频点的能量谱密度作为卷积神经网络的输入,设计卷积神经网络算法实现电弧故障检测。经实验验证,所提出方法可清晰区分电弧故障电流特征和正常工作电流特征;在实验室测试中,所提出方法可准确地检测出光伏系统直流串联电弧故障。
A novel DC series arc-fault detection method for photovoltaic systems based on convolutional neural network is proposed. Firstly, the short-time Fourier transform is used to derive the time-frequency information of the current. Then energy spectral density, acted as the time-frequency joint function, is used to construct the time-frequency spectrogram of the current. The coordinate information on the time-frequency spectrogram of the current is inputted to the convolutional neural network. And the convolutional neural network is trained to discriminate the arc-fault current and normal operation current. Experiments verified that the proposed method can make a clear distinction between the arc-fault current and normal operation current. In the experiments, the DC series arc-fault current of the photovoltaic system can be detected accurately.
A novel DC series arc-fault detection method for photovoltaic systems based on convolutional neural network is proposed. Firstly, the short-time Fourier transform is used to derive the time-frequency information of the current. Then energy spectral density, acted as the time-frequency joint function, is used to construct the time-frequency spectrogram of the current. The coordinate information on the time-frequency spectrogram of the current is inputted to the convolutional neural network. And the convolutional neural network is trained to discriminate the arc-fault current and normal operation current. Experiments verified that the proposed method can make a clear distinction between the arc-fault current and normal operation current. In the experiments, the DC series arc-fault current of the photovoltaic system can be detected accurately.
引文
[1] 牟龙华,王伊健,蒋伟,等(Mu Longhua,Wang Yijian,Jiang Wei,et al.).光伏系统直流电弧故障特征及检测方法研究(Study on characteristics and detection method of DC arc fault for photovoltaic system) [J].中国电机工程学报(Proceedings of the CSEE),2016,36(19):5236-5244.
[2] 吴春华,冯夏云,袁同浩,等(Wu Chunhua,Feng Xiayun,Yuan Tonghao,et al.).基于BP神经网络的光伏故障电弧检测方法研究(Photovoltaic arc fault detection method based on BP neural network) [J].太阳能学报(Acta Energiae Solaris Sinica),2016,37(11):2958-2964.
[3] He C,Mu L,Wang Y.The detection of parallel arc fault in photovoltaic systems based on a mixed criterion[J].IEEE Journal of Photovoltaics,2017,7:1717-1724.
[4] Lu S,Phung B T,Zhang D.A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems[J].Renewable and Sustainable Energy Reviews,2018,89:88-98.
[5] 张士文,张峰,王子骏,等(Zhang Shiwen,Zhang Feng,Wang Zijun,et al.).一种基于小波变换能量与神经网络结合的串联型故障电弧辨识方法(Series arc fault identification method based on energy produced by wavelet transformation and neural network) [J].电工技术学报(Transactions of China Electrotechnical Society),2014,29(6):290-295.
[6] Shibo L,Tharmakulasingam S,Phung B T.DA-DCGAN:An effective methodology for DC series arc fault diagnosis in photovoltaic systems[J].IEEE Journal of Photovoltaics,2019,7:45831-45840.
[7] Georgijevic N,Jankovic M,Srdic S,et al.The detection of series arc-fault in photovoltaic systems based on the arc current entropy[J].IEEE Transactions on Power Electronics,2015,31(8):5917-5930.
[8] Chen S,Li X,Xiong J.Series arc fault identification for photovoltaic system based on time-domain and time-frequency-domain analysis[J].IEEE Journal of Photovoltaics,2017,7:1105-1114.
[9] 吴春华,徐文新,李智华,等(Wu Chunhua,Xu Wenxin,Li Zhihua,et al.).光伏系统直流电弧故障检测方法及其抗干扰研究(Study on detection method and its anti-interference of DC arc fault for photovoltaic system) [J].中国电机工程学报(Proceedings of the CSEE),2018,38(12):3546-3555.
[10] Shekhar A,Ramirez-Elizondo L,Bandyopadhyay S,et al.Detection of series arcs using load side voltage drop for protection of low voltage DC systems[J].IEEE Transactions on Smart Grid,2018,9(6):6288-6297.
[11] Telford R,Galloway S,Stephen B,et al.Diagnosis of series DC arc faults:A machine learning approach[J].IEEE Transactions on Industrial Informatics,2017,13(4):1598-1609.
[12] 董伟,张俊民,桂美景,等(Dong Wei,Zhang Junmin,Gui Meijing,et al.).基于小波包分解的航空故障电弧的识别(Identification of aviation arc fault based on wavelet packet decomposition) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2018,37(11):76-81.
[13] 陈照,李奎,张洋子,等(Chen Zhao,Li Kui,Zhang Yangzi,et al.).基于电弧电磁辐射的故障电弧识别(Arc fault detection based on electromagnetic radiation) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2017,36(3):70-74.
[14] 林方圆,苏建徽,赖纪东(Lin Fangyuan,Su Jianhui,Lai Jidong).光伏系统直流故障电弧识别方法研究(Study on detection method of DC arc fault in PV systems) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2015,34(12):7-13.
[15] 黄佳平(Huang Jiaping.).基于BP神经网络的直流电弧故障检测技术研究(Research on DC arc fault detection technology based on BP neural network) [D].杭州:杭州电子科技大学(Hangzhuou:Hangzhou Dianzi University),2017.
[16] 张宏建,孙志强(Zhang Hongjian,Sun Zhiqiang).现代检测技术(Modern detection technology) [M].北京:化学工业出版社(Beijing:Chemical Industry Press),2007.9.
[17] Zhivomirov H.On the development of STFT-analysis and ISTFT-synthesis routines and their practical implementation[J].TEM Journal,2019,8(1):56-64.
[2] 吴春华,冯夏云,袁同浩,等(Wu Chunhua,Feng Xiayun,Yuan Tonghao,et al.).基于BP神经网络的光伏故障电弧检测方法研究(Photovoltaic arc fault detection method based on BP neural network) [J].太阳能学报(Acta Energiae Solaris Sinica),2016,37(11):2958-2964.
[3] He C,Mu L,Wang Y.The detection of parallel arc fault in photovoltaic systems based on a mixed criterion[J].IEEE Journal of Photovoltaics,2017,7:1717-1724.
[4] Lu S,Phung B T,Zhang D.A comprehensive review on DC arc faults and their diagnosis methods in photovoltaic systems[J].Renewable and Sustainable Energy Reviews,2018,89:88-98.
[5] 张士文,张峰,王子骏,等(Zhang Shiwen,Zhang Feng,Wang Zijun,et al.).一种基于小波变换能量与神经网络结合的串联型故障电弧辨识方法(Series arc fault identification method based on energy produced by wavelet transformation and neural network) [J].电工技术学报(Transactions of China Electrotechnical Society),2014,29(6):290-295.
[6] Shibo L,Tharmakulasingam S,Phung B T.DA-DCGAN:An effective methodology for DC series arc fault diagnosis in photovoltaic systems[J].IEEE Journal of Photovoltaics,2019,7:45831-45840.
[7] Georgijevic N,Jankovic M,Srdic S,et al.The detection of series arc-fault in photovoltaic systems based on the arc current entropy[J].IEEE Transactions on Power Electronics,2015,31(8):5917-5930.
[8] Chen S,Li X,Xiong J.Series arc fault identification for photovoltaic system based on time-domain and time-frequency-domain analysis[J].IEEE Journal of Photovoltaics,2017,7:1105-1114.
[9] 吴春华,徐文新,李智华,等(Wu Chunhua,Xu Wenxin,Li Zhihua,et al.).光伏系统直流电弧故障检测方法及其抗干扰研究(Study on detection method and its anti-interference of DC arc fault for photovoltaic system) [J].中国电机工程学报(Proceedings of the CSEE),2018,38(12):3546-3555.
[10] Shekhar A,Ramirez-Elizondo L,Bandyopadhyay S,et al.Detection of series arcs using load side voltage drop for protection of low voltage DC systems[J].IEEE Transactions on Smart Grid,2018,9(6):6288-6297.
[11] Telford R,Galloway S,Stephen B,et al.Diagnosis of series DC arc faults:A machine learning approach[J].IEEE Transactions on Industrial Informatics,2017,13(4):1598-1609.
[12] 董伟,张俊民,桂美景,等(Dong Wei,Zhang Junmin,Gui Meijing,et al.).基于小波包分解的航空故障电弧的识别(Identification of aviation arc fault based on wavelet packet decomposition) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2018,37(11):76-81.
[13] 陈照,李奎,张洋子,等(Chen Zhao,Li Kui,Zhang Yangzi,et al.).基于电弧电磁辐射的故障电弧识别(Arc fault detection based on electromagnetic radiation) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2017,36(3):70-74.
[14] 林方圆,苏建徽,赖纪东(Lin Fangyuan,Su Jianhui,Lai Jidong).光伏系统直流故障电弧识别方法研究(Study on detection method of DC arc fault in PV systems) [J].电工电能新技术(Advanced Technology of Electrical Engineering and Energy),2015,34(12):7-13.
[15] 黄佳平(Huang Jiaping.).基于BP神经网络的直流电弧故障检测技术研究(Research on DC arc fault detection technology based on BP neural network) [D].杭州:杭州电子科技大学(Hangzhuou:Hangzhou Dianzi University),2017.
[16] 张宏建,孙志强(Zhang Hongjian,Sun Zhiqiang).现代检测技术(Modern detection technology) [M].北京:化学工业出版社(Beijing:Chemical Industry Press),2007.9.
[17] Zhivomirov H.On the development of STFT-analysis and ISTFT-synthesis routines and their practical implementation[J].TEM Journal,2019,8(1):56-64.