基于EEMD样本熵的高速列车转向架故障特征提取

Feature Extraction of High Speed Train Bogie Based on Ensemble Empirical Mode Decomposition and Sample Entropy

为了监测高速列车转向架关键部件的工作状态,提出了采用聚合经验模态分解和样本熵信息测度理论相结合的方法提取信号特征.以转向架正常、空气弹簧失气、横向减振器故障和抗蛇行减振器故障4种典型工况下车体及转向架的振动信号为研究对象,将信号进行聚合经验模态分解,得到一系列成分简单的固有模态函数,分别计算样本熵值构成高维特征矢量,最后采用支持向量机进行故障状态的分类识别.实验结果表明,列车在200 km/h速度下,故障识别率可以达到88%,证明了该特征提取算法的有效性.

To monitor the working condition of key components of high speed train bogie in time, a novel method for feature extraction is proposed by combination of ensemble empirical mode decomposition （EEMD） and sample entropy theory. Vibration signals are obtained from train body and bogie under four typical working conditions, such as normal condition, air spring fault, lateral damper fault, and yaw damper fault. After EEMD, signals have been decomposed into a series of intrinsic mode functions （IMFs）, and the sample entropies of these IMFs constitute a high dimensional characteristic vector. Finally, the support vector machine is used to identify the fault conditions based on the characteristic vector. The experimental result shows that the recognition rate is 88% at the speed of 200 km/h. Therefore, this feature extraction method is effective for high speed train bogie fault signals