【作者】 许昆；

【导师】 李智勇；

【作者基本信息】 湖南大学 ， 计算机应用技术， 2008， 硕士

【摘要】 常用的多目标优化方法自身的不足及其在实际应用中存在的诸多困难,一直阻碍着多目标优化方法的发展。在20世纪80年代中期,进化算法开始应用于解决多目标优化问题。目前涌现了多种多目标进化算法,其中一些已成功应用到实际应用中,从而形成了一个热门研究领域。量子粒子群算法是将量子计算与粒子群算法相结合的一种崭新的优化方法,具有很强的生命力和极大的研究价值。量子算法中融入了量子力学的许多基本特性,极大地提高了计算的效率,已逐步成为一种崭新的计算模式。量子粒子群算法大大提高了搜索效率且能弥补粒子群算法容易早熟的不足,具有广泛的研究前景。本文的主要工作和研究成果如下:1.在分析当前多目标优化算法的优缺点的基础上,针对求解多目标优化中存在的收敛性不够好,分布不均匀的问题,本文将量子理论引入粒子群算法,提出一种基于量子衍生方法的多目标粒子群算法,该算法采用Pareto支配关系来更新粒子的个体最优值和全局最优值,通过定义极大极小距离,并使用该距离方法来裁减非支配解。2.将该算法应用于多维0-1背包问题,实验结果表明该算法具有较强的搜索能力和寻优效率,与NSGA2算法和SPEA2算法相比在Pareto解集的收敛性指标上有提高,尤其适用于高维复杂函数的优化。3.将提出的算法应用于军队任务调度和指派的高级逻辑问题。并根据高级逻辑问题中参数多,约束条件加法和为1的特点,提出一种面向和约束的方法,采用三角公式转化约束条件,节约了存储空间的同时也提高了搜索效率。结果表明该方法的可行性和有效性。更多还原

【Abstract】 The deficiencies of ordinary multi-objective optimization algorithms, together with their implement problems in reality, always prevent the objective optimization methods from development. In the mid-1980s, Evolutionary Algorithm was introduced to solve the Multi-objective Optimization Problem. A variety of Multi-objective Evolutionary Algorithms are proposed at present. Some of them have been successfully applied to practical applications. More and more researchers begin to engage in this field. Quantum Particle SwarmOptimization algorithm (QPSO) is a new optimum method that combines quantum computation with Particle Swarm Optimization (PSO). It appears strong life-force and be valuable for research. Quantum computation absorbed many essential characters of quantum mechanics, which improved the computation efficiency, and become a brand new model of computation.QPSO has greatly enhanced the efficiency of search and can prevent premature of PSO and it has a wide research foreground. The following are the major contributions:1. On the basis of analyzing the advantages and disadvantages of Multi-objective Optimization algorithms, this article presents a Quantum-bit Particle Swarm Optimization (QBPSO) algorithm for Multi-objective Optimization Problems so as to improve the convergence and it is well-distributed. QBPSO adopts the non-dominated storing method for solutions population and use a new population diversity preserving strategy which is based on the Pareto max-min distance.2. The multidimensional 0-1knapsack problems are tested and the results show that the proposed method can efficiently find Pareto optimal solutions that are closer to Pareto font and better on distribution. Compared with NSGA2 and SPEA2, the value of S increases 11.10% and 11.06% on avenge. Especially, this proposed method is outstanding on more complex high-dimensional optimization problems.3. Apply our method on Advanced Logistics Problem (ALP) of army’s assignment and scheduling. Generally speaking, there are many parameters in ALP and sum of them is always1. According to this character, we proposed a new sum constraint oriented method, in which trigonometric formulas are used to transform constraint condition, so as to save storage space and improve the search efficiency. The experimental results indict its feasibility and effectiveness.更多还原