Fault Diagnosis of Photovoltaic Array Based on Deep Belief Network

The environment of the PV array is harsh and severe, resulting in frequent faults. In order to improve the accuracy of PV array fault diagnosis, a deep belief networks (DBN) based fault diagnosis method is proposed for the common fault types of PV arrays. The experimental feature parameters was obta...

Full description

Saved in:
Bibliographic Details
Main Authors: Caixia TAO, Xu WANG, Fengyang GAO
Format: Article
Language:zho
Published: State Grid Energy Research Institute 2019-12-01
Series:Zhongguo dianli
Subjects:
Online Access:https://www.electricpower.com.cn/CN/10.11930/j.issn.1004-9649.201901066
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The environment of the PV array is harsh and severe, resulting in frequent faults. In order to improve the accuracy of PV array fault diagnosis, a deep belief networks (DBN) based fault diagnosis method is proposed for the common fault types of PV arrays. The experimental feature parameters was obtained by Matlab simulation and the fault diagnosis model with the five operating states of the PV array is established. According to the characteristics of the DBN, the impacts of training sets, training periods and restricted boltzmann machine (RBM) layers on the model performance are analyzed through recognition experiments. Compared with the fuzzy C-means clustering (FCM), the support vector machine (SVM) and the back-propagation neural network (BPNN) method from the overall diagnostic accuracy and different types of fault diagnostic accuracy. The results show that the method is suitable for fault classification of photovoltaic arrays, and it improves the accuracy of fault identification effectively compared with other diagnostic models.
ISSN:1004-9649