2025 International Conference on Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD 2025)

Vibration sensing is a crucial means for condition monitoring and fault prediction of rotating machinery, which could potentially ensure their safety and reliability. The existing methods including housing-mounted vibration sensors and laser-based shaft displacement sensors are restricted by signal propagation attenuation and environmental factors, respectively. Therefore, it is still a challenge to obtain high quality vibration signals and the sensing systems can hardly perform real-time operation state observation. This study presents a Field Programmable Gate Array (FPGA)-based small outline shaft-end mounted on rotor vibration sensing solution, which allows real-time extraction of multi-domain feature and in-situ rotor condition monitoring. A novel design concept for shaft-end mounted wireless vibration sensing is proposed, which employs wireless power supply for small outline and lightweight construction to reduce its impact on rotor operation. A multi-domain feature extraction method for rotor fault analysis and real-time observation is developed with in-sensor computing enhanced by FPGA. Then, research of rotor imbalance and impact detection is conducted, which validates the feasibility and effectiveness of the ORS sensor for real-time fault monitoring at a sampling rate of 4000Hz. This investigation provides a lightweight, high-precision, real-time sensing approach for the online monitoring of rotating machinery.

Rotor condition monitoring, on rotor sensing (ORS), In-sensor computing, multi-domain feature extraction, wireless power supply