基于差分进化算法的模糊核聚类算法及其在故障诊断中的应用

摘要

In allusion to the determination of the kernel parameters and the effective evaluation of the clustering results of Fuzzy Kernel-clustering Algorithm (FKCA), differential evolution algorithm (EA) is used to search the optimal kernel parameter and the clustering centers. Furthermore, the Xie-Beni index is promoted to the kernel space, and a new fitness function is designed to improve the clustering performance. The proposed method is applied in the standard benchmark as well as the motor bearing fault dataset. The results shows that the proposed method is a promising clustering method for fault diagnosis.%针对模糊核聚类方法中，核函数参数的确定问题以及聚类结果的有效评价问题，提出采用差分进化算法进行核函数参数和聚类中心的同时寻优策略。并将Xie-Beni指标推广至核空间，设计了有效的适应度函数以实现聚类效果的提升。对所提出的方法进行数值试验，同时应用在电机轴承的故障诊断中，取得了不错的效果，验证了方法的可行性。