A Systematic Approach to Intelligent Maintenance of Production Systems with a Framework for Embedded Implementation

摘要

To achieve worry-free production and prevent unexpected downtime, predictive maintenance has become a prevailing strategy, supported by prognostics and health management techniques. In this paper, a systematic data-driven approach for intelligent maintenance is discussed, as well as an embedded system architecture for fault detection and prediction. Two case studies are presented, including performance prediction for wire-twisting machine based on temperature data, and an embedded solution for electric valve actuator monitoring.