Fault identification and location strategy of photovoltaic power station based on gated recurrent unit

GAO Tianlong; LIU Weiliang; ZHANG Wenjun

doi:10.19666/j.rlfd.202107124

English Version Reader LoginJournal Login

您当前的位置：

首页 >

文章列表页 >

Fault identification and location strategy of photovoltaic power station based on gated recurrent unit

更新时间：2025-09-28

- Fault identification and location strategy of photovoltaic power station based on gated recurrent unit
- Thermal Power Generation Vol. 51, Issue 3, Pages: 21-28(2022)
- 作者机构：
  
  1. 西北工业大学微电子学院
  2. 华北电力大学自动化系
- 作者简介：
- 基金信息：
- DOI：10.19666/j.rlfd.202107124
  CLC： TM615;TP183
- Published：2022
- 稿件说明：
移动端阅览
GAO Tianlong, LIU Weiliang, ZHANG Wenjun. Fault identification and location strategy of photovoltaic power station based on gated recurrent unit[J]. Thermal power generation, 2022, 51(3): 21-28.
DOI：

GAO Tianlong, LIU Weiliang, ZHANG Wenjun. Fault identification and location strategy of photovoltaic power station based on gated recurrent unit[J]. Thermal power generation, 2022, 51(3): 21-28. DOI： 10.19666/j.rlfd.202107124.

摘要

光伏电站大多地处恶劣环境，易出现多种类型故障，目前光伏阵列故障的精确识别与定位存在较大困难。为此，提出一种基于门控循环（GRU）神经网络的光伏阵列故障识别与定位策略。首先搭建了光伏电站仿真模型，采用扰动观测法实现光伏组件的最大功率跟踪；通过开展不同类型光伏阵列故障模拟试验，分析确定故障特征参数；然后采集不同故障的样本集，建立基于GRU神经网络的故障诊断模型；最后分别通过仿真模型与光伏实验平台与BP神经网络模型进行对比验证。验证结果表明，GRU神经网络故障诊断模型能够更加准确地识别与定位光伏阵列故障，可用于指导光伏电站日常维护，提高电站的经济效益。

Abstract

Photovoltaic power stations are mostly located in harsh environment and are prone to various types of failures. At present

there are great difficulties in accurate fault identification and location. For this reason

a fault identification and location strategy for photovoltaic power stations based on gated cycle network(GRU) is proposed.Firstly

the simulation model of photovoltaic power station is established

and the disturbance observation method is applied to realize the maximum power tracking of the photovoltaic modules. Via simulation tests for different types of photovoltaic array faults

the fault characteristic parameters are determined. Then

the sample sets of different faults are collected

and the fault diagnosis model based on the GRU neural network is established

which is verified by the simulation model and photovoltaic experimental platform. The verification results show that

the GRU neural network fault diagnosis model can accurately identify and locate photovoltaic faults

which can be used to guide the daily maintenance of photovoltaic power stations and improve the economic benefits of photovoltaic power stations.

关键词

Keywords

references

Views

307

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

MULTISENSOR INFORMATION MERGING TECHNOLOGY AND IT'S APPLICATION IN VIBRITION FAULT DIAGNOSIS OF ROTARY MACHINES

APPLICATION OF EXTENDED Petri NETWORK INTO FAULT DIAGNOSIS OF THERMAL POWER UNITS

APPLICATION OF MULTI-DIMENSIONAL BP NEURAL NETWORK IN FAULT DIAGNOSIS OF BOILERS

RESIDUAL SPACE OF KERNEL PRINCIPAL COMPONENT ANALYSIS BASED SENSOR FAULT DIAGNOSIS

Kernel principal component analysis based fault diagnosis for condenser systems

Related Author

JIN Fenghua

LI Luping

LUO Yi

CHEN Yu-fei

MA Zeng-huiNorth China Electric Power University

GAO Jianqiang

CHEN Peng College of Energy

ZHOU Weiqing1

Related Institution

Changsha University of Science and Technology

North China Electric Power University

College of Energy, Power and Mechanical Engineering, North China Electric Power University

Huaneng International Power Co., Ltd

AI问答

⁰