Fault Diagnosis for Reciprocating Air Compressor Valve Using P-V Indicator Diagram and SVM

Abstract

This paper presents a method of fault diagnosis for the reciprocating air compressor valve based on the indicator diagram and the support vector machine (SVM). This paper strikes 7 invariant moments of the indicator diagram of reciprocating air compressor, using image processing methods, according to the same moment theory. Then the method can be used to extract effective features as feature vectors for training the support vector machine, and achieve fault diagnosis for reciprocating air compressor valve. Finally, the paper simulate 5 kinds of working conditions of valve to identify using the fault monitoring system of reciprocating compressor valve, in order to verify the feasibility and effectiveness of the method.