Simulation of a Cylindrical Roller Bearing with an Embedded Piezoelectric Sensor for Local Fault Detection

Local fault detection in bearings through accelerometer has been one of the most fundamental condition monitoring techniques for three decades. Nevertheless, the sensitivity of accelerometers to surrounding noises from other components/machines and a long transmission path between the sensor and bearing has attracted more attention toward using embedded sensors and measurement of local strain rather than acceleration. These days, utilizing embedded smart materials, such as low-cost piezoelectric sensors, can benefit different industries to detect abnormal conditions in machines or structures. In some cases, these abnormalities show up in the form of sudden strain changes that is detectable by piezoelectric materials. Therefore, in this research, using embedded piezoelectric sensors in a bearing housing with a short transmission path is proposed to detect abnormalities due to a local fault. Through numerical simulation in ANSYS APDL, the dynamic of a cylindrical roller bearing in the healthy and defective conditions is investigated. According to the results, the existence of a local fault affects the radial strain changes and proportionally changes the generated voltage signal. These changes (fault symptoms) are investigated in the time and frequency domains for fault detection.