拥挤距离排序的多目标文化粒子群优化算法

Cultural based multi-objective particle swarm optimization algorithm using crowding distance sorting method

为解决工程优化设计问题,引入文化进化框架,提出一种拥挤距离排序的多目标文化粒子群算法,采用拥挤距离排序算子,并删除密集区域的多余粒子,以保证Pareto前沿的分布均匀性;基于拥挤距离值,从精英知识和条件知识中选择处于最分散区域的粒子,并将其分别作为全局和局部最优,以增强算法全局寻优能力;依据拥挤距离的变化,动态调整粒子群飞行参数,以提高算法收敛效率,通过标准测试问题以及与其他算法的对比,表明了所提出算法的有效性和鲁棒性。

To solve the engineering design problems, by introducing cultural evolution framework, a cultural based multiobjective particle swarm optimization algorithm with crowding distance sorting is proposed. The redundant particles in the crowded area are deleted with the distance sorting operator to guarantee the elitism＇s uniform distribution. With the distance value, the global and local best of the particles are selected from the most disperse region in the elitism and situational knowledge, respectively, so as to enhance its global searching capability. The evolution parameters are adjusted dynamically according to the changing of distance to improve the convergence speed. Some standard test problems and the comparison with other algorithms show the effectiveness and robustness of the algorithm.