基于非劣排序遗传算法的核电厂维修决策多目标优化方法研究

Research on Multi-Objective Optimization Method for Maintenance Decision of Nuclear Power Plants

提出核电厂系统级、不完全、非定期逐次预防维修模型。该模型引入役龄回退因子衡量维修效果,并且通过邻接矩阵法与结构函数矩阵法得到系统可靠度;同时将每个元件的更换对整个系统可靠度及成本的影响考虑在内,推导出可靠度与总成本的计算公式。采用非劣排序遗传算法(NSGA2)对该模型进行多目标优化,得到核电厂某系统整个寿期的多目标维修方案。通过算例阐述计算步骤,进行维修方案的经济分析与多目标优化。

This paper proposes a new kind of maintenance model which characteristic is in the system level, incompletely maintenance and sequential maintenance. The model introduces age reduction factor measuring the maintenance effect and obtains the system reliability by the adjacency matrix method and the structural function matrix method. At the same time, the influence of the replacement of each component on the system reliability and total cost is taken account, and the formula for calculating the reliability and the total cost is derived. Then, using the NSGA-II method to make the multi-objective optimization of the model, obtaining the multi-objective maintenance plan about a system for the whole life of a nuclear power plant and making suggestions for maintenance decision making. Finally, an example is presented to verify the calculation steps, the economic analysis and multi-objective optimization of the maintenance program.