Cavitation Detection Method for Pumps Based on Motor Current Signature Analysis

Abstract

Simple and cost-effective methods for cavitation detection are required for the stable operation of pump systems. Previous researches on cavitation detection systems have used microphones or vibration sensors, however they are costly and difficult to install. Recently, motor current signature analysis (MCSA) has been gaining attention with regard to motor fault detection. In this paper, we review the characteristics of motor current signals in a cavitation state and propose an MCSA technique for pump cavitation detection. We observed that the foot around the fundamental harmonic spectrum in the power spectral density (PSD) of current signals increased in a cavitation state. Based on these results, we propose defining an evaluation value that reflects the foot and using it for cavitation detection.