Diagnostics of synchronous motor based on sound recognition with application of Linear Predictive Cepstrum Coefficients and fuzzy classifier

Abstract

This document provides the concept of investigations of acoustic signals of imminent failure conditions of synchronous motor. Measurements were made by re-corder OLYMPUS WS-200S. Sound recognition software has been implemented. Algorithms of signal processing and analysis have been used. The system is based on the LPCC algorithm and fuzzy classifier with triangular membership function. Results confirm the correct operation of the system of sound recognition of synchronous motor.