Expert system for induction motor fault detection

Abstract

The paper describes the development and implementation of an expert system for fault detection of induction motors. Early detection of induction motors failures may prevent the occurrence of malfunction and ensure timely replacement and servicing. In order to make timely and accurate fault detection an expert system is designed and developed. The expert system is a software system aimed at 'simulating' human-expert knowledge and assistance in making decisions in a particular area. This paper presents a conceptually new type of expert system based on the artificial neural networks. Inputs into the expert system are the results of the algorithms to detect faults based on different physical quantities (current and vibration). The expert system performs analysis of input data and, based on them, it finally defines the type of fault or faults in the case of multiple failures.

Keywords

• induction machine
• fault detection
• expert system