In this paper,three novel fuzzy clustering models are proposed based on the principle of cluster center separation.

New structure algorithm of clustering center and category determination method;

With the Kohonen network clustering in neural network employed, the degree of relationship of the universal joint axle of the rolling mill was input to Kohonen network as the training sample, studied and clustered by the network to generate different clustering centers according to the different depth and different degree of relationship among the cracks.

With the characteristics of the Kohonen network clustering in neural network,the degree of relationship of universal joint axis of rolling mill is input to Kohonen network as training sample,and is studied and clustered by the network to generate different clustering centers owing to the different depth and different degree of relationship among severity of crack.

Optimizing initial cluster center of K-means algorithm

Since each single fault diagnosis method has its advantages and disadvantages,so after the immune system is applied to choosing hidden layer neural network data and getting a cluster center,then a suitable layer weight is selected,and neural network output is calculated,the output result shows that the diagnosis is effective.

The iterative expressions for cluster center and fuzzy membership are deduced respectively.

RBF design method based on center clustering and PSO

Then,a cluster centroid splitting strategy is proposed based on maximum distance to resolve the problem of selecting cluster centroid.

Migration strategy of parallel migration strategy for effect of K-Median Cluster;

Application and Research of Parallel Genetic Algorithm in Data Mining of K-Medians;