中文摘要：

如何使算法快速收敛到真正的Pareto前沿,并保持解集在前沿分布的均匀性是多目标优化算法重点研究解决的问题.提出一种基于云模型的改进NSGA-Ⅱ算法,利用正态云模型云滴的随机性和稳定倾向性特点,分别对交叉、变异、拥挤距离算子进行改进.使算法既具有传统的趋势性和满足快速寻优能力,又具有随机性.在提高收敛速度与保持种群多样性之间做了个很好的权衡.通过求解多目标背包问题,对本文算法的多目标优化性能进行了考察,并与NSGA-Ⅱ算法进行比较,结果表明本文算法在整个解空间内能快速搜索到Pareto最优解,使搜索到的Pareto最优解在前沿均匀分布.

英文摘要：

How fast the algorithm converges to the true Pareto frontier solution set and maintains the uniformity in the forefront of multi-objective optimization algorithm is the key research issues.This paper presents an improved NSGA-Ⅱ algorithm based on cloud model,using the normal cloud model cloud droplets is random and bias stability characteristics make the algorithm trends in both traditional and fast search capability to meet,but also has random.Improving the convergence rate and maintain the population diversity,made a good trade-off.By solving the multi-objective optimization problem,multi-objective optimization algorithm performance was investigated,and compare to the NSGA-Ⅱ,the results show the algorithm can Quick Search Pareto optimal solution in the solution space and the Pareto optimal solution to the uniform distribution in the front.