Fault diagnosis of the continuous stirred tank heater using fuzzy-possibilistic c-means algorithm

Abstract

This paper mainly introduces a practical algorithm called fuzzy-possibilistic c-means (FPCM) clustering algorithm. It is based on fuzzy c-means (FCM) clustering algorithm and possibilistic c-means (PCM) clustering algorithm. FPCM algorithm figures out the existing problems of the above two algorithms and produces both memberships and possibilities simultaneously. For example, FPCM algorithm works out the inconsistency problem of FCM algorithm and overcomes the coincident clusters problem of PCM algorithm. Then this paper applies FPCM algorithm to the fault detection and diagnosis of the continuous stirred tank heaterCSTH). The effect of the fault diagnosis approach is demonstrated on the CSTH benchmark.