Exploration of online diagnostic system to support organizational learning for plant maintenance work -problem setting and proposal of system architecture-
Hidekazu Yoshikawa,Ming Yang,Yangping Zhou
2009-01-01
Abstract:A distributed online maintenance support system is proposed by the notion that the plant maintenance work is an intelligent creative work that, based on (i)multi-layered versatile knowledge of "objective, function, behavior and structure" on the target systems and equipments to be maintained, with applying (ii)broad expertise knowledge such as cause-consequence relations of equipment failures, symptoms of failure, how to find such symptoms, how to predict failure progression, and by processing (iii)many sensor signals from plant instrumentation, to reduce risk information (iv) whether or not the levels of defense-in-depth (DiD) are well maintained. Among the above mentioned four elements of maintenance work, the item (i) is central because it will be further expanded to the items (ii) and (iii), finally to the item (iv). In this exploratory study, the designing method of the whole system architecture is first discussed as a distributed online diagnostic system, and then the formation method of elementary client module is discussed where the client module will support the individual maintenance workers for individual equipment diagnosis that (i)evaluating the soundness of the equipment at the present time by utilizing the time records of various sensors data, (ii)evaluating failure modes in the future, and (iii)reducing time vs. reliability curve of the equipment failure modes in the future, and (iii)reducing time vs. reliability curve of the equipment until the next scheduled reactor shutdown.