基于连续小波Tsallis奇异熵的航空交流电弧故障检测

AC Arc-fault Detection Based on Continuous Wavelet Tsallis Singular Entropy in Airplane

为了有效提取航空交流电弧故障的电流信号特征,提出连续小波变换Tsallis奇异熵与极限学习机相结合的航空交流电弧故障判别新方法。对发生电弧故障线路的电流信号进行连续小波变换分解,对获得的时频系数矩阵进行奇异值分解(SVD),得到被测线路电流信号的Tsallis奇异熵,构建特征向量,采用极限学习机对TSE特征向量进行训练,得到适用于航空交流电弧故障检测的分类模型,应用该模型对不同负载下的特征数据进行识别分类。根据试验结果可以看出,连续小波变换Tsallis奇异熵结合极限学习机能够准确识别电弧故障状态与正常运行状态。

In order to effectively extract the current signal characteristics of aviation AC arc fault,a new fault identification method based on continuous wavelet transform Tsallis singular entropy(TSE)and extreme learning machine(ELM)was proposed.The continuous wavelet transform was performed on the current signal of the arc fault line,and the obtained time-frequency coefficient matrix was subjected to singular value decomposition(SVD)to obtain the Tsallis singular entropy of the measured line current signal,and the eigenvector was constructed.The extreme learning machine was used for the TSE.The eigenvectors were trained to obtain a classification model suitable for aeronautical AC arc fault detection.The model was used to identify and classify the feature data under different loads.According to the experimental results,it can be seen that the continuous wavelet transform Tsallis singular entropy combined with the limit learning machine can accurately identify the arc fault state and normal operating state.