Faults Diagnosis for a centrifugal machine using the Mahalanobis distance

Abstract

The paper proposes a fault diagnosis procedure based on a model-free approach and the use of pattern recognition techniques. In particular this paper aims to improve the isolation performance of a Fuzzy Faults Classifier (FFC) previously proposed by the author by the use of the Mahalanobis distance as metric for identifying the most probable fault. The proposed approach is applied to an industrial multishaft centrifugal compressor located in an Air Separation Unit (ASU). In particular faults due to the wear and tear of the thrust bearing and to fouling of the compressor stage are considered. The case study confirms the goodness of the overall procedure in the detections of both single as well as multiple faults and shows the improvements of the proposed approach in terms of the FDI system promptness.