基于独立主成分分析的PWM逆变电路的故障诊断

PWM inverter circuit fault diagnosis based on independent principal component analysis

PWM逆变电路故障具有多发性和非线性相关性,对其准确检测的难度较大。为了提高PWM逆变电路的故障诊断性能,提出基于独立主成分分析的PWM逆变电路的故障诊断方法,采用电路输出信号测量方法进行故障状态特征检测,对采集的PWM逆变电路信号进行调理和抗混叠滤波处理,提取检测信号的独立主成分特征,实现故障识别。仿真结果表明,采用该方法进行PWM逆变电路故障诊断的准确检测性能较好,故障诊断可靠稳定。

The fault of the PWM inverter circuit has the characteristics of malfunction and nonlinear correlation, and it is difficult to detect its accuracy. In order to improve the fault diagnosis performance of the PWM inverter circuit, the PWM inverter circuit fault diagnosis method based on independent principal component analysis is proposed. The measuring method of the cir- cuit output signal is used to detect the fault state feature to perform the conditioning and anti-aliasing filtering for the acquired signal of the PWM inverter circuit, and extract the independent principal component characteristics of the signal, so as to recog- nize the fault. The simulation results show that the method has better accurate detection performance for PWM inverter circuit fault diagnosis, and the fault diagnosis is reliable and stable.