AI-based damage detection method for acceleration vibration of Rotating Machinery

Abstract

In recent years, condition-based maintenance (CBM) is becoming popular for machinery equipment. CBM uses sensors to detect anomalies and performs maintenance when anomalies are detected. However, when the observed object is vibration of rotating machinery such as motors, there is a risk of overlooking anomalies using only a threshold value because it is necessary to find anomaly features in the time series data. Therefore, we developed three damage detection methods (two of them AI-based) for vibration waveforms obtained from an acceleration sensor attached to a rotating machine.