摘要
针对光伏发电设备工作过程中存在的故障检测低下问题,设计了集采样、稳压、采样保持和故障采集于一体的集成电路,并提出了故障检测方法。通过分布式测控网络,使系统能够实时、准确地获得光伏发电设备的运行状态。故障检测装置自带计算处理模块智能对比检测数据,为实现故障检测算法提供数据支撑。为了更好地识别光伏发电设备的故障,提出了基于反向传播(BP)神经网络引入可信度检测度量参数的故障监测算法。该算法不仅能够消除误差,并且有很强的抗干扰性。试验结果证明,在不同的干扰信噪比下,检测故障的时间仅在5 ms左右,能够自主地对光伏发电设备运行过程中产生的故障进行识别,极大地提高了光伏发电设备故障检测的工作效率。
Aiming at the problem of low fault detection in the working process of photovoltaic power generation equipment,an integrated circuit integrating sampling,voltage stabilization,sample holding and fault acquisition is designed and a fault detection method is proposed.The system can obtain the operating status of photovoltaic power generation equipment in real time and accurately through a distributed measurement and control network.The fault detection device comes with a computational processing module to intelligently compare the detection data and provide data support for the implementation of the fault detection algorithm.In order to better identify the faults of photovoltaic power generation equipment,a fault monitoring algorithm based on back propagation(BP)neural network introducing plausibility detection metric parameters is proposed.The algorithm not only can eliminate errors but also has strong anti-interference property.The experimental results prove that the fault detection time is only about 5 ms under different interference signal-to-noise ratios,which can autonomously identify the faults generated during the operation of photovoltaic power generation equipment and greatly improve the efficiency of photovoltaic power generation equipment fault detection.
作者
王宁
王恩路
韩则胤
韩国强
苏宝定
WANG Ning;WANG Enlu;HAN Zeyin;HAN Guoqiang;SU Baoding(CGN New Energy Holdings Co.,Ltd.,Beijing 100070,China)
出处
《自动化仪表》
CAS
2023年第3期88-90,97,共4页
Process Automation Instrumentation
关键词
光伏发电设备
故障检测
采样电路
反向传播神经网络
可信度
归一化
控制芯片
干扰信噪比
Photovoltaic power generation equipment
Fault detection
Sampling circuit
Back propagation(BP)neural network
Credibility
Normalization
Control chip
Interference signal-to-noise ratio
