韩素敏, 周孟, 郑书晴.基于BP神经网络的三相电压源型逆变器开路故障诊断[J].河南理工大学学报（自然科学版）,2021,40(6):126-131.

HAN S M, ZHOU M, ZHENG  S Q.Open circuit fault diagnosis of three-phase voltage source inverter based on BP neural network[J].Journal of Henan Polytechnic University(Natural Science) ,2021,40(6):126-131.

基于BP神经网络的三相电压源型逆变器开路故障诊断

韩素敏, 周孟, 郑书晴

河南理工大学 电气工程与自动化学院，河南 焦作454000

摘要:为实现三相电压源型逆变器（three-phase voltage source inverter，TP-VSI）开路故障的诊断与精确定位，提高故障诊断效率，提出基于BP神经网络的TP-VSI开路故障诊断方法。首先，研究TP-VSI正常工作原理、开路故障机理，确定线电压作为检测量；其次，采用快速傅里叶变换（fast Fourier transform ,FFT）提取不同故障下的特征值，构建故障特征向量；最后，建立一个3 层BP神经网络，进行故障诊断，实现对开路故障的精确定位。该方法通过获取包含故障信息的故障特征值，经过FFT把时域信号变换到频域信号，再通过BP神经网络建立TP-VSI开路故障与特征值的一一对应关系。与其他智能诊断方法进行对比，证明了 BP神经网络对TP-VSI开路故障诊断的可行性和较高的诊断准确度，实现了 BP神经网络对TP-VSI开路故障的诊断与精确定位。

关键词:BP神经网络;三相电压源型逆变器;开路;故障机理;故障诊断;精确定位

doi:10.16186/j.cnki.1673-9787.2020090023

基金项目:国家重点研发计划项目（2016YFC0600906）；河南省重点研发与推广专项项目（科技攻关）（20210220094）

收稿日期:2020/09/04

修回日期:2021/01/14

出版日期:2021/11/15

Open circuit fault diagnosis of three-phase voltage source inverter based on BP neural network

HAN Sumin, ZHOU Meng, ZHENG shuqing

School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo 454000 ,Henan,China

Abstract:in order to diagnose and locate the open-circuit fault of three-phase voltage source inverter（TP-VSI）accurately and to improve the fault diagnosis efficiency ，a method of TP-VSl open-circuit fault diagnosis based on BP neural network was proposed. Firstly the normal working principle of 'TP-VSI and the mechanism of open-circuit fault were studied ,and the line voltage was taken as the detection quantity. Secondly ,the fast Fourier transform（FFT）was used to extract the eigenvalues under different faults to construct the fault eigenvectors. Finally ,a three-layer BP neural network was established for fault diagnosis and accurate positioning of open circuit faults. In this method ,the fault eigenvalues containing fault information were obtained ,the time do-main signals were transformed into frequency domain signals through FFT ,and then the relationship between TP-VSI open-circuit faults and eigenvalues was established through BP neural network. Compared with other intelligent diagnosis methods,the feasibility of BP neural network for TP-VSI open-circuit fault diagnosis and the higher fault diagnosis accuracy were realized. BP neural network was realized for TP-VSI open-circuit fault diagnosi sand accurate location.

Key words:BP neural net work;three-phase voltage inverter;open circuit;fault mechanism;fault diagnosis;accurate location

 基于BP神经网络的三相电压源型逆变器开路故障诊断_韩素敏.pdf