基于振动信号时频分解-样本熵的受电弓裂纹故障诊断

构建了基于时频分解-样本熵测度的受电弓振动信号故障特征提取模型。对振动信号进行聚合经验模态分解,接着对分解得到的本征模态函数计算参数优化后的样本熵特征。将获取的故障特征输入基于粒子群参数优化的支持向量机(PSO-SVM)进行受电弓故障识别分析。结果发现,基于受电弓顶管振动信号的EEMD样本熵故障诊断效果较好,而碳滑板振动信号诊断效果较差。针对这一特点,利用二代小波样本熵进行优化,进一步提高了碳滑板振动信号故障诊断结果,验证了现代时频分析算法与信息熵联合的诊断方法在受电弓振动信号特征提取与故障诊断的可行性与有效性。

Fault diagnosis for pantograph cracks based on time-frequency decomposition and sample entropy of vibration signals

A fault feature extraction model of pantograph vibration signal based on time-frequency decomposition and sample entropy was constructed. Firstly, ensemble empirical mode decomposition was conducted for vibration signal, then sample entropy was calculated after optimizing parameters for the intrinsic modal function obtained by EEMD. The sample entropy features were input to the support vector machine (PSO-SVM) based on particle swarm optimization (PSO) to identify the pantograph fault identification. The results show that the EEMD sample entropy fault diagnosis based on the pantograph panhead top pipe vibration signal has good accuracy, and the carbon contact strip vibration signal has poor diagnosis effect. According to this, the second-generation wavelet sample entropy was used to optimize and further improve the fault diagnosis results of carbon contact strip vibration signal. It verified the feasibility and effectiveness of the combination of the modern time-frequency analysis algorithm and the information entropy in the feature extraction and fault diagnosis of pantograph fault vibration signals.