Study On Decision Models And Its Application Of Condition-Based Maintenance For Power Plant Equipment

Several types of decision-making models in the Condition-Based Maintenance aiming at resolving the lack of quantitative analysis are established to optimize the inspection interval and maintenance time and strategy for the equipment in the power plant. According to the delay time theory, Two-Stage Function Test Model based on the statistical analysis of failure law is built to determine the optimal detection interval. Markov Process is often used to describe the development of the equipment degradation, and in this paper, Markov Decision Model is built up for equipment maintenance strategy determination with using the Policy Iteration to find out the solution, which is the optimal maintenance strategy under the specific target. Evaluation indicators are determined by considering both the impact and consequence of equipment failure and the characteristics of maintenance program. Entropy Weight Multi-Objective Evaluation Model for maintenance policy is established to rank the optional maintenance policies. Taking the major and auxiliary equipment of turbine generating unit as the real examples, the established decision-making models in the paper are tested and verified to be practical and efficient and will be widely used in the power plant.