基于快速傅里叶变换和改进T-S模糊神经网络集成模型的逆变器开路故障诊断方法研究  

Research on open‑circuit fault diagnosis method for inverter transistor based on FFT and improved T‑S FNN ensemble model

作　　者：田广强 乔珊珊 侯奥 王福忠[3] TIAN Guangqiang;QIAO Shanshan;HOU Ao;WANG Fuzhong(School of Intelligent Engineering,Huanghe Jiaotong University,Jiaozuo 454950,China;Henan Power Transmission and Transformation Construction Co.,Ltd.,Zhengzhou 450000,China;School of Electrical Engineering and Automation,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]黄河交通学院智能工程学院,河南焦作454950 [2]河南送变电建设有限公司,河南郑州450000 [3]河南理工大学电气工程与自动化学院,河南焦作454000

出　　处：《电力科学与技术学报》2023年第6期76-86,共11页Journal of Electric Power Science And Technology

基　　金：国家自然科学基金(U1804143);河南省科技攻关(212102210146)。

摘　　要：针对受负载扰动和测量噪声影响,逆变器开路时的故障边界间、故障与特征间存在交叠和模糊性问题,在对逆变器功率管开路故障特征的分析基础上,提出基于快速傅里叶变换和改进T‑S(Takagi‑Sugeno)模糊神经网络集成模型的逆变器开路故障诊断模型。首先,依据快速傅里叶变换分析逆变器的三相输出电流波形,提取功率管发生不同类型开路故障时的故障特征;其次,采用规则自分裂技术和模糊C均值设计T‑S模糊神经网络的前件网络的隶属函数层;然后,依托自适应Levenberg‑Marquardt算法对T‑S网络参数进行训练;最后,利用训练后的T‑S网络实现逆变器功率管的多种故障类型与位置的诊断。实验结果表明,所提出模型的诊断准确率高达96%,能够显著改善逆变器功率管开路故障诊断时所存在的问题。Aiming at overlap and fuzziness between fault boundaries,faults,and characteristics under load disturbances and measurement noise influence when the inverter is in an open-circuit state,,an inverter open circuit fault diagnosis model built upon the fast Fourier transform(FFT)and improved Takagi-Sugeno(T-S)fuzzy neural network(FNN)integration model is proposed based on the analysis of the characteristics of the inverter power tube open circuit fault.Firstly,fault characteristics are extracted when different types of open-circuit faults occur in the power tubes according to the three-phase output current waveforms of the inverter analyzed by the FFT.Secondly,the membership function layer of the antecedent network of the T-S fuzzy neural network is designed by using the rule self-splitting technology and fuzzy C-means,and the parameters of the T-S network are trained by leveraging the adaptive Levenberg-Marquardt algorithm.The trained T-S network is used to realize the diagnosis of multiple fault types and positions of the inverter power tubes.The example results show that the diagnostic accuracy of the proposed model is up to 96%,which can significantly improve the problems existing in the open-circuit fault diagnosis of inverter power tubes.

关 键 词：逆变器 开路故障诊断 快速傅里叶变换 改进T‑S模糊神经网络 自适应LM算法 

分 类 号：TM464[电气工程—电器]