基于PCA-BLS的逆变器故障诊断

为了确保电气设备的安全可靠运行，提出基于主成分分析法与宽度学习系统的逆变器故障诊断方法。利用主成分分析法对逆变器输出的电流信号进行处理，提取信号特征；构建宽度学习系统，并编写不同故障模式下的故障编码；利用不同故障模式下的信号特征对宽度学习系统进行训练，利用网络输出编码实现故障分类。仿真结果表明，该研究方法在诊断准确率及训练时间方面优于传统的神经网络故障诊断方法。

Inverter Fault Diagnosis Based on Broad Learning System

In order to ensure the safety and reliability of the electrical equipment during operating,the paper proposes a new inverter fault diagnosis method based on principal component analysis (PCA) and broad learning system (BLS).Firstly,the current signal of the inverter was processed by using PCA method,and the signal characteristics were extracted.Secondly,the BLS was constructed and the fault coding under different fault modes were obtained.Finally,the BLS network was trained by using the signal characteristics with different fault modes,and the fault classification was realized by using the network output coding.The simulation results show that the method proposed in the paper is superior to the traditional neural network in terms of diagnosis accuracy and training time.