论文部分内容阅读
为有效提取机车轴承故障特征,开展信号自适应分解方法对比研究。分析了经验模态分解、局域均值分解和局部特征尺度分解3种常用方法的局部均值计算、分解成分和分解能力。针对局域均值分解存在的问题,提出了改进方案并有效验证。进一步提出了先做改进局域均值分解,再采用1(1/2)维谱处理得到的乘积分量的机车轴承诊断的方法,成功用于DF_4型机车的故障诊断。
In order to effectively extract locomotive bearing fault characteristics, a comparative study of signal adaptive decomposition methods is carried out. The local mean calculation, decomposition component and decomposition ability of three commonly used methods of empirical mode decomposition, local average decomposition and local feature scale decomposition are analyzed. Aiming at the existing problems of local average decomposition, an improved scheme is proposed and validated. The method to diagnose the locomotive bearing of the DF_4 locomotive is successfully put forward by using the method of improving the local average decomposition and then using the product components obtained from the 1 (1/2) dimension spectrum processing.