动态多目标优化进化算法研究进展

动态多目标优化问题(Dynamic multi-objective optimization problems, DMOPs)已成为工程优化的研究热点, 其目标函数, 约束函数和相关参数都可能随时间不断变化, 如何利用搜索到的历史最优解对新的环境变化做出快速响应, 是设计动态多目标优化进化算法(Dynamic multi-objective optimization evolutionary algorithm, DMOEA)的重点和难点. 本文在介绍DMOEA的基础上, 分析了近年来基于个体和种群级别的环境响应策略, 多策略混合等的DMOEA主要研究进展, 并介绍了DMOEA的性能测试函数, 评价指标以及在工程优化领域中的应用, 分析了DMOEA研究中仍面临的主要问题, 展望了未来的研究方向.

Research Progress of Dynamic Multi-objective Optimization Evolutionary Algorithm

Dynamic multi-objective optimization problems (DMOPs) have become a research focus on the engineering optimization, the objective function, constraint functions and related parameters are likely to be changing over time, How to make rapid response to new environment changes by using the historical optimal solution is the key and difficulty of designing dynamic multi-objective optimization evolutionary algorithm (DMOEA). Based on the introduction to DMOEA, this paper analyzes the main research progress of DMOEA based on individual and population level environmental response strategy and multi-strategy mixing in recent years, introduces the performance test function, evaluation index and application of DMOEA in the field of engineering optimization, analyzes the main problems still faced in DMOEA research and giving an outlook to the future research.