Research on the condition based maintenance decision of equipment in power plant

Aiming at the problem that the condition information of the equipment in power plant is seldom considered during making maintenance decision and as a result insufficiency or surplus of maintenance occurs often, a model of condition based maintenance decision is proposed. Based on system division, the criticality of equipment in same level is evaluated and maintenance strategies of equipment are decided based on the criticality index. To the equipment using condition based maintenance strategy, the condition is evaluated and predicted by fully using the operating parameters, condition monitoring parameters and the residual life is estimated. To minimize the maintenance cost of the equipment in same subsystem, a decision tree algorithm is used, in which equipment condition, logistic support constraints, life cycle profit, and equipment dependencies are considered. The method is applied to the feed water subsystem of a thermal power plant. It is shown by the instance that this method is feasible and effective, and it can be used as a technical support next in realizing condition-based maintenance.