论文部分内容阅读
提出一种基于空间自适应划分的多目标优化算法.为了增强种群的收敛性和多样性,多维搜索空间被划分成多个网格,网格内的粒子通过共享“引导”粒子的经验信息调整自身的速度和位置,并引入年龄观测器实时记录引导粒子对Pareto解集所做的贡献,及时更新引导粒子,以增强算法的全局搜索能力.对多目标测试函数以及环境经济调度问题进行了仿真实验,实验结果表明,所提出算法能对解空间进行更加全面、充分的探索,快速找到一组分布具有较好的逼近性、宽广性和均匀性的最优解集合.
A multi-objective optimization algorithm based on spatial adaptive partitioning is proposed.In order to enhance the convergence and diversity of the population, the multidimensional search space is divided into a number of grids, and the particles in the grid share the experience of “guiding” the particles Information is used to adjust its speed and position and the age observer is introduced to record the contributions of the guided particle to the Pareto solution set in real time and to update the guide particle in time to enhance the global search ability of the algorithm.Multiple objective test functions and environmental economic dispatch The experimental results show that the proposed algorithm can comprehensively and fully explore the solution space and quickly find a set of optimal solution sets with good approximation, broadness and uniformity.