Evolutionary Multi-objective Optimization of Substation Maintenance using Markov Model

Improving the reliability and reducing the overall cost are two important but often conflicting objectives for substations. Proper scheduling of preventive maintenance provides an effective means to tradeoff between the two objectives. In this paper, Pareto-based multi-objective evolutionary algorithms are proposed to optimize the maintenance activities because of their abilities of robust search towards best-compromise solutions for large-size optimization problems. Markov model is proposed to predict the deterioration process, maintenance operations, and availability of individual components. Minimum cut sets method is employed to identify the critical components by evaluating the overall reliability of interconnected systems. Pareto-fronts are generated for comparisons with other substation configurations. Results for four different substation configurations are presented to demonstrate potentials of the proposed approach for handling more complicated configurations.