Fault Diagnosis Theory: Method and Application Based on Multisensor Data Fusion

Abstract

Abstract This article discusses the elementary theory and method of condition monitoring and fault diagnosis in complex mechanical systems, using multisensor data fusion and integration technology. The basic concept of decision-layer multi-sensor data fusion, i.e., Dempster-Shafer evidence theory, is concisely described. The algorithms and implementation of decision-layer data fusion for condition monitoring and fault diagnosis, based on Dempster-Shafer evidence theory, are studied. We present an approach to diagnose multiple faults of a working diesel engine, based on the above algorithms. The analysis results show that this approach effectively improves the accuracy and reliability of fault diagnosis of the diesel engine.