A proposal of hybrid fuzzy clustering algorithm with application in condition monitoring of industrial processes

Abstract

In this chapter a hybrid algorithm using fuzzy clustering techniques is presented. The algorithm is applied in a condition monitoring scheme with online detection of novel faults and automatic learning. The proposal, initially identifies the outliers based on data density. Later, the outliers are removed and the clustering process is performed. To extract the important features and improve the clustering, the maximum-entropy-regularized weighted fuzzy c-means is used. Then, the use of kernel functions is performed for clustering the data, where there is a non-linear relationship between the variables. Thus, the classification accuracy can be improved because better class separability is achieved. Next, the regulation factor of the resulting partition fuzziness (parameter m) and the Gaussian Kernel bandwidth (parameter σ) are optimized. The feasibility of the proposal is demonstrated by using the DAMADICS benchmark.