A Fuzzy Decision Variables Framework for Large-Scale Multiobjective Optimization

In large-scale multiobjective optimization, too many decision variables hinder the convergence search of evolutionary algorithms. Reducing range space will significantly alleviate this puzzle. With in mind, article proposes a fuzzy framework for optimization. The divides entire process into two main stages: 1) evolution and 2) precise evolution. evolution, we blur original solution to reduce algorithm so that population can quickly converge. degree fuzzification gradually decreases with process. Once approximately converges, turn actual are directly optimized increase diversity as be closer true Pareto optimal front. Finally, embeds some representative algorithms proposed verifies framework’s effectiveness through comparative experiments on various problems 500 5000 variables. experimental results show improve performance computational efficiency optimization