Improved K-Means Clustering Method Based on Complex Network for Rolling Bearing Fault Diagnosis

Abstract

According to the selection difficulties of initial clustering center of k-means clustering algorithm, this paper proposes a method that is to use complex network degree to improve k-means clustering algorithm for fault pattern recognition method, and to improve the accuracy of clustering. Use network to represent fault data structure, with joint connecting matrix to express similarity between nodes, according to the complex concepts of networks degree, calculate the size of every node degree, and select the maximum degree of node as k-means clustering initial center. This method is applied to the rolling bearing clustering diagnosis example, achieving good fault diagnosis effect. This study provides a new method for the selection of initial cluster centers of K-means clustering.