基于复合多尺度排列熵与FO-SVM的滚动轴承故障诊断方法

Fault Diagnosis for Rolling Bearings Based on Composite Multi-scale Permutation Entropy and FO-SVM

MPE算法中不充分的时间序列粗粒化过程会造成原始振动信号在时间序列中信息的缺失。为优化这种不成熟的粗粒化过程,相关学者创新地采用复合粗粒化的思想,提出了复合多尺度排列熵(CMPE)。为了实现滚动轴承的智能故障诊断,提出一种基于CMPE与萤火虫优化支持向量机(FO-SVM)的滚动轴承智能故障诊断方法。首先使用CMPE表征滚动轴承的原始故障信息,然后构建FO-SVM多故障分类器,实现对滚动轴承故障类型和程度的智能识别。通过仿真信号分析验证了CMPE相对于MPE在信号稳定性方面的优越性;实验数据分析结果表明:相比于基于MPE与FO-SVM的滚动轴承故障诊断方法,所提故障诊断方法不仅能够准确诊断滚动轴承的故障类型和程度,而且识别率达到了100%。

In the MPE algorithm,the insufficient process of course-grained time series can cause the loss of the original vibration signal in the time series information.In order to optimize this immature of coarse-grained process,the relevant scholars have innovatively adopted the idea of composite coarse granulation and put forward composite multi-scale permutation entropy(CMPE).In this paper,an intelligent fault diagnosis method for rolling bearings based on CMPE and firefly optimized support vector machine(FO-SVM)is proposed.Firstly,the CMPE is used to extract the original fault information of the rolling bearing.Then the FO-SVM multi-fault classifier is constructed to realize the intelligent identification of the fault types and degrees of the rolling bearing.The superiority of CMPE relative to MPE in signal stability is verified by signal simulation analysis.The experimental data shows that compared with the fault diagnosis method based on MPE and FOSVM,the proposed fault diagnosis method can accurately diagnose the types and degrees of failure of the rolling bearing,and the recognition accuracy is up to 100%.