小样本条件下基于SGMM模型的滚动轴承故障诊断研究

Research on Mechanical Fault Diagnosis Based on SGMM Model under Small Sample Condition

摘要:

针对小样本条件下传统机械故障诊断诊断算法准确率偏低的问题,提出一种基于SGMM模型的故障诊断算法。先确定与故障建模策略相关的提取任务,预估潜在的机械故障状态变化；对故障信号进行变分模态分解,得到最小熵解卷积结果,并满足端点效应的处理需求,实现对机械故障位置的精确定位与诊断。实验结论表明,SGMM模型更注重对故障脉冲成分的连续检测,在以峭度作为衡量标准的条件下,该方法增强故障冲击力的作用更强,能更早诊断出轴承类机械元件的早期故障状态。

Abstract:

Aiming at the problems of difficulty and low accuracy of mechanical fault diagnosis under the condition of small samples, a fault diagnosis method based on sgmm model was proposed. Firstly, the extraction task related to the fault modeling strategy was determined, and the change of potential mechanical fault state was estimated; The minimum entropy deconvolution result was obtained by the variational mode decomposition of the fault signal, which meeted the processing requirements of the endpoint effect and realizes the location and diagnosis of the mechanical fault. The analysis results showed that sgmm model payed more attention to the continuous detection of fault pulse components. If kurtosis was taken as the measurement standard, this method could enhance the fault impact force more effectively, and could diagnose the early fault state of bearing mechanical components earlier.