Research On Multi-Preference Based Many-Objective Evolutionary Algorithm And Its Applicatioin

Evolutionary algorithms(EAs) is a kind of population-based optimization algorithm that simulated evolution mechanism, now it has been getting more and more attention for it’s strong ability of global optimization. In recent years, it has been a hot research in the field of multi-objective optimization research while using evolutionary algorithms to solve multi-objective optimization problems. However, on the one hand, most existing multi-objective evolutionary algorithms use Pareto-based sorting method to choice better solutions, this leads to losing selective pressure for solutions in the high-dimensional space, and influence the convergence; On the other hand, the irregular shape of the Pareto front of the many-objective optimization problems make currently algorithms hard to maintain the distribution uniformity of the optimal solution sets. Based on the two difficult problems, multi-preference was used in this paper to improve the Individual selection pressure and the distribution uniformity of solution set. The research content in this paper is as follows:1. Research on multi-preference based many-objective evolutionary algorithm is proposed. Firstly, focusing on how to use multi-preference to guide the solutions, a collaborative evolutionary strategy is built to implement the co-evolutionary of the preferences and the solutions. Secondly, the statistic analysis of the frequency of multi-preference is analysed to acquire the optimal synergy model between multi-preference and solutions. In the algorithm performance evaluation experiments, the proposed algorithm is been used to solve2-10objectives testing functions, the experimental results verify it’s effectiveness and superiority. 2. A method of antenna array optimization based on bipolar preferences many-objective optimization algorithms is been proposed. Firstly, combining decision makers’ preference on excellent solutions and their disgust on worse solutions, the TOPSIS method is been introduced to compared the similarity between solutions, and a stricter dominant relationship is built to guide the solutions to the higher directional radiation pattern and lower zero in value. Secondly, a high-dimensional space diagonal technology method is been used to visualize the solutions in the many-objective space. In the algorithm performance evaluation experiments, the proposed method is compared with other three currently algorithms, the experimental results show the superiority in the convergence precision.