摘要
根据第二代非支配排序遗传算法(NSGAⅡ)的不足之处,提出了一种新的多目标遗传算法——非支配排序均匀遗传算法(NSUGA)。新算法采用了多父本多点交叉方式,同时将均匀设计的思想用于算法的交叉操作;新算法还对拥挤距离的计算过程和算法的终止条件进行了改进。通过两个多目标优化测试函数的仿真计算对比,显示NSUGA算法在求解精度、计算效率和避免算法陷于局部最优解方面均优于NSGA II算法。
According to the shortage of non-dominated sorting generic algorithm Ⅱ(NSGA Ⅱ),put forward one new multi-objective generic algorithm called non-dominated sorting uniform generic algorithm(NSUGA).In the new algorithm,adopted the crossing mode of multi-parent and multi-point,and used the idea of uniform design in the algorithm's crossing action,improved moreover the crowding-distance calculation and the algorithm's terminating condition.Simulation results on two test problems show that NSUGA is better in the precision of seeking pareto solutions,the efficiency of computing and the avoidance of gaining local optimal solutions than NSGA II.
出处
《计算机应用研究》
CSCD
北大核心
2011年第11期4020-4022,4025,共4页
Application Research of Computers
基金
湖南省教育厅资助科研项目(09C397)
国家自然科学基金资助项目(50975032)
关键词
多目标遗传算法
多目标优化
多父本多点交叉
非支配排序
均匀设计
multi-objective generic algorithm(MOGA)
multi-objective optimization
crossing mode of multi-parent and multi-point
non-dominated sorting
uniform design