基于高斯变异和自适应参考点的MOPSO优化算法

MULTI-OBJECTIVE PARTICLE SWARM OPTIMIZATION BASED ON GAUSSIAN MUTATION AND ADAPTIVE REFERENCE POINT

针对MOPSO优化算法在优化多目标问题当中收敛程度较差和容易进入部分最优的缺点,提出一种基于高斯变异和自适应参考点融合的MOPSO优化算法。利用高斯变异位置更新方法改善解集提前停止寻优现象,提高MOPSO优化算法在寻找最优过程中寻找解集的多样性;采用自适应参考点的外部档案维护策略,将收敛性较差的粒子剔除,提高算法的收敛性。实验结果表明:改进的MOPSO算法同传统的MOPSO算法相比,反向代距离和超体积比有了明显的改善,具有更好的解集多样性和收敛性。

Aiming at the problem that multi-objective particle swarm optimization was slower and easy to fall into local optimum in solving multi-objective optimization problems, we proposed a multi-objective particle swarm optimization algorithm based on Gaussian mutation and adaptive reference point fusion. The Gaussian mutation location update method was used to improve the premature phenomenon of the solution, and the diversity of the search solution in the optimization process of the multi-objective particle swarm optimization algorithm was improved. The external file maintenance strategy of the adaptive reference point was used to eliminate the particles with poor convergence and improve the convergence of the algorithm. The experimental results show that the improved multi-objective particle swarm optimization has significantly improved reverse generation distance and super volume ratio compared with the traditional multi-objective particle swarm optimization algorithm, and has better diversity and convergence.