Real Time Remote Monitoring and Anomaly Detection in Industrial Robots Based on Vibration Signals, Enabling Large Scale Deployment of Condition based Maintenance

Abstract

Rita Chattopadhyay, Mruthunjaya (Jay) Chetty, Eric XiaozhongJi, Stephanie Cope and Jeffrey E Davis

Loss of wafers and expensive repairs of process equipment are often caused by uncontrolled and unmonitored failures of critical components during semiconductor process. High volume manufacturing (HVM) of semiconductor chips employ large number of robots. Malfunctioning of these robots causes particle contamination, wafer slip and wafer breaks, resulting in production yield loss, equipment down time and factory throughput. Presently, wafer handling monitoring instruments diagnose vibrations of a robot at the end-effector. Detection of anomaly in these vibrations are performed manually during scheduled maintenance and are highly dependent on the experience of maintenance personnel. This not only is prone to human error, but also limits large scale deployment in semiconductor fabrication. The proposed solution automates this process by monitoring the vibration signal patterns, continuouslyin real-time, to proactively identify robots that are at risk of failure. The vibration signals are captured using tri-axial accelerometers placed near the bearings in the arms of the robot. The proposed method analyzes specific parameters of the vibration signal and generates alerts for maintenance, before the uncontrolled vibrations affect production. Identifying parameters which are correlated to failures ischallenging. This work presents four such indicative parameters, determined based on time and frequency domain analysis of the vibration data collected from good and faulty robots. The proposed method based on outlier detection is an Edge /Cloud architecture for remote monitoring and alerting

PDF