Compared with general engineering projects, electric power projects have higher requirements for security and stability. The achieve of time, cost and quality are necessary conditions to ensure that the electric power project can be high-quality finished. In order to ensure the smooth implementation and completion of the project, it is necessary to optimize the interest relationship among the targets, and to achieve the overall optimal state. Traditional multi-objective optimization methods have great limitations in practical project optimization process. But genetic algorithm is simple, easy to operate, universal and suitable for parallel processing and it can achieve good convergence and operability in solving multi-objective optimization problem.This paper studies the application of genetic algorithm in multi-objective optimization problems in electric power projects. Firstly, the general mathematical model for multi-objective optimization problems are studied and the traditional multi-objective optimization methods are analyzed in detail. Secondly, the meaning of schema theorem and puzzle block hypothesis are studied and the main components and computational steps of genetic algorithms are analyzed. The principles and characteristics of several representative multi-objective genetic algorithms are studied, including Vector Evaluation Genetic Algorithm, Multi-Objective Genetic Algorithm and Non-dominated Sorting Genetic Algorithm NSGA-II, the improved Non-dominated Sorting Genetic Algorithm was studied in detail. Finally, a three-dimensional time-cost-quality multi-objective optimization model is established and a series of Pareto optimal solutions are obtained.10 samples selected from those Pareto optimal solutions were analyzed. The results show that the multi-objective optimization algorithm of this study can effectively solve the multi-objective optimization problems in electric power projects. |