Fault Classification and Diagnosis of Industrial Application Motor Drives using Soft Computing Techniques

Abstract

Most of the modern industry drives uses an induction motor as the main drive system. The properties like compact size, low cost, and wide range of speed control makes induction motor a universally acceptable electromechanical device. Due to the wide usage, they are prone to different faults. The presence of faults affects the operation of the Induction motor by reducing its efficiency. If these faults are not diagnosed at the proper time, they can lead to the shutdown of the entire system under operation. Thus there is a constant need for the reliable and safe operation of Induction motors. Condition monitoring is required through which presence of various faults occurring in induction motor can be diagnosed beforehand and necessary precautions and preventive works can be performed. The proposed method deals with the fault diagnosis and detection through continuous monitoring based on Motor Current Signature Analysis (MCSA). Among the various methods used for fault diagnosis, Soft Computing techniques form a promising option. The proposed algorithm is implemented using Fuzzy Logic soft computing technique. The paper discusses the results obtained by simulating and detecting various faults in induction motor using Fuzzy logic, in C#. Experimental results have verified the effectiveness of the proposed method.