电气技术  2021, Vol. 22 Issue (10): 57-64    DOI:
电气设备检修与故障诊断 |
一种基于极限学习机和皮尔逊相关系数的光伏阵列故障快速诊断方法
陈世群1, 高伟2, 陈孝琪2, 涂彦昭2, 杨艳3
1.国网福建省电力有限公司福州市长乐区供电公司,福州 350202;
2.福州大学电气工程与自动化学院,福州 350108;
3.淮阴工学院自动化学院,江苏 淮安 223003
A fast fault diagnosis method for photovoltaic array via extreme learning machine and Pearson's correlation coefficient
CHEN Shiqun1, GAO Wei2, CHEN Xiaoqi2, TU Yanzhao2, YANG Yan3
1. Fuzhou Changle District Electric Power Supply Branch of State Grid Fujian Electric Power Co., Ltd, Fuzhou 350202;
2. College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108;
3. Faculty of Automation, Huaiyin Institute of Technology, Huaian, Jiangsu 223003
全文: PDF (5390 KB)   HTML (1 KB) 
输出: BibTeX | EndNote (RIS)      
摘要 

光伏阵列故障会造成能量损失,甚至引发供电中断或火灾事故。因此,对故障的快速识别至关重要。本文提出一种光伏阵列故障快速诊断方法,用于快速感知故障及故障发生的时刻。通过分析光伏阵列常见故障的信号变化规律,提出利用正常运行时的功率波形训练一个极限学习机预测模型,用于预测短时功率的变化;接着计算实测波形和预测波形的皮尔逊相关系数;若光伏阵列发生故障,相关系数将低于一定的阈值,从而识别故障的发生。实测实验验证了所提方法具有很强的故障辨识能力,准确率达到99.37%。所提方法的故障辨识时间仅为4.355ms,亦可作为光伏阵列故障录波的启动方法使用。

服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
陈世群
高伟
陈孝琪
涂彦昭
杨艳
关键词 光伏阵列故障录波极限学习机(ELM)皮尔逊相关系数    
Abstract

The failure of the photovoltaic (PV) array can cause energy loss and even cause power supply interruption or fire accident. Therefore, it is very important to identify the fault quickly. A new fast fault diagnosis method for PV array is proposed to quickly perceive the fault and the time of fault occurrence in this paper. Based on the analysis of the signal variation rule of PV array traditional faults, a prediction model of extreme learning machine (ELM) is investigated to train the power waveform during normal operation, which is adopted to predict the short-time power variation. Then Pearson's correlation coefficient of measured waveforms and predicted waveforms are calculated. If the PV array fails, the correlation coefficient will be below the threshold, thus identifying the occurrence of the fault. Experimental results show that the proposed method has strong ability of fault identification, and the accuracy reaches 99.37%. The fault identification time of the proposed method is only 4.355ms, and it can be used as the start-up method of PV array fault recording.

Key wordsphotovoltaic array    fault recording    extreme learning machine (ELM)    Pearson's correlation coefficient   
收稿日期: 2021-02-26     
基金资助:

江苏省住房和城乡建设厅指导项目(2019ZD001287); 江苏淮安市科技局自然科学研究计划项目(HAB201905)

作者简介: 陈世群(1995—),男,福建福州人,硕士,研究方向为新能源故障诊断研究。
引用本文:   
陈世群, 高伟, 陈孝琪, 涂彦昭, 杨艳. 一种基于极限学习机和皮尔逊相关系数的光伏阵列故障快速诊断方法[J]. 电气技术, 2021, 22(10): 57-64. CHEN Shiqun, GAO Wei, CHEN Xiaoqi, TU Yanzhao, YANG Yan. A fast fault diagnosis method for photovoltaic array via extreme learning machine and Pearson's correlation coefficient. Electrical Engineering, 2021, 22(10): 57-64.
链接本文:  
https://dqjs.cesmedia.cn/CN/Y2021/V22/I10/57