摘要
设计了一种新颖的基于差分进化算法和NSGA-Ⅱ的混合进化算法用来解决多目标优化问题。在此算法中,根据算法的搜索情况设计相应的自适应变异算子,以便在突变操作中找到Pareto解。同时,选择操作将基于NSGA-Ⅱ快速非优超排序和拥挤机制将父代与子代的双种群进行截短,确保最优解不会丢失并保证解的多样性。三个经典测试函数的仿真结果表明,文中算法在实现多目标优化问题的两个目标(获得收敛于真实Pareto前沿的解和解沿着前沿均匀扩展)方面表现出良好的综合性能。
A novel hybrid differential evolution algorithm for multi-objective optimization problem named DE-NSGAⅡ is proposed based on differential evolution and NSGA-Ⅱ.In DE-NSGAⅡ,the mutation operator parameter is adaptive for the searching effect in every generation to find Pareto solutions.The selective operator based on fast ranking mechanisms and crowing distance sorting of NSGA-Ⅱtruncates the population set formed by the parents and the new points to ensure the optimal solution not be lost and to ensure the diversity of optimal solution.DE-NSGAⅡ is implemented on three classical multi-objective optimization problems,and the results illustrate the good comprehensive performance of DE-NSGAⅡ in achieving two goals： they find the solutions converge to the true Pareto-front and uniform spread along the front.
出处
《运筹与管理》
CSCD
北大核心
2010年第6期58-64,共7页
Operations Research and Management Science
基金
国家自然科学基金资助项目(70801030)
中央高校基本科研业务费专项资金项目(2010MS133)
关键词
运筹学
混合进化算法
自适应差分进化算法
NSGA-Ⅱ
多目标优化
仿真
operations research
hybrid evolutionary algorithm
adaptive differential evolution algorithm
NSGA-Ⅱ
multi-objective optimization
simulation
