Multi-state Aperiodic Inspection-based Maintenance Strategy Optimization using Real-time Data

Le résumé fourni par la source

For complex engineering systems, proper equipment maintenance is one of the most important means to ensure the system’s safety, reliability, and availability. To further reduce the maintenance cost of the system, this paper proposes a method to optimize the inspection and maintenance strategy based on real-time condition monitoring data. The method relies on a data-driven predictive model to construct a system Remaining Useful Life (RUL) distribution, converted into a system reliability index related to the system operating state. A clustering algorithm classifies the classes of degradation states during system operation. Different degradation state classes correspond to different inspection decision variables, which flexibly determine the optimal inspection interval and maintenance decision according to the system operation status. The performance of the proposed dynamic predictive maintenance strategy measures the maintenance cost rate (MCR) using the NASA aircraft engine dataset. The experimental results show that the method effectively reduces the inspection and maintenance cost of the system.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.