基于交叉小波变换和主元分析的电力电子电路故障特征提取

针对现有电力电子电路故障特征提取特征量精确度不足、分类差异性不明显以及故障提取过程易受到噪声的影响等缺点,提出一种基于交叉小波变换和主元分析的电力电子电路故障特征提取方法。该方法首先采用交叉小波变换分析故障信号,然后得出表征交叉小波谱图特性的特征量矩阵,最后利用主元分析方法降低特征量矩阵维数,剔除特征向量中的冗余信息。通过BP神经网络进行的故障诊断仿真测试,其诊断准确率达98.2%,证明了该方法的准确性。

Feature extraction of power electronics circuits based on Cross -Wavelet Transform and Principle Component Analysis

Aiming at drawbacks of current methods for power electronics circuits feature extraction.there is not enough accuracy and not obvious classification, and the process of feature extraction is easily affected by noise.Firstly, the faulty signal of circuit information feature is analyzed and extracted by cross-wavelet transform.Then, the initial feature matrix is obtained representing cross-wavelet spectrum.Finally, principle component analysis is applied for reducing the dimension of initial feature matrix and those which redundant information are eliminated.The back propagation (BP) neural network classifiers are utilized for fault diagnosis simulation test.The results show that the fault detection accuracy is up to 98.2%.Simulation results demonstrate the accuracy of the proposed method.