Reducing power plant maintenance costs is an important way to improve the power enterprises production efficiency, for the current situation that maintenance assessment index is too single and the problem of high complexity is difficult to solve by common maintenance optimization methods, power plant multi-objective maintenance projects optimization model is established and solved by a method of fast-non-dominated-sorting genetic algorithm with elitist strategy (NSGA-II) in this paper. This method overcomes the previous problem of single optimizing assessment index and improves the efficiency and robustness of the problem solving efficiency. By the case study of a 300MW power unit electric feed...