相位编码量子蚁群算法及在连续优化中的应用  

Phase encoded-based quantum ant colony algorithm with application to continuous optimization

作　　者：李盼池[1,2] 杨雨[2] 张巧翠[2] 

机构地区：[1]东北石油大学石油与天然气工程博士后科研流动站 [2]东北石油大学计算机与信息技术学院,黑龙江大庆163318

出　　处：《计算机应用研究》2010年第12期4450-4453,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(60773065);中国博士后科学基金资助项目(20090460864);黑龙江省博士后科学基金资助项目(LBH-Z09289);黑龙江省教育厅科学技术研究资助项目(11551015;11551017)

摘　　要：针对蚁群算法只适用于离散优化问题的局限性和收敛速度慢的问题,提出一种适合连续优化的量子蚁群算法。该方法直接采用量子位的相位对蚂蚁编码。首先根据基于信息素强度和可见度构造的选择概率,选择蚂蚁的前进目标;然后采用量子旋转门更新描述蚂蚁位置的量子比特,完成蚂蚁移动,并采用Pauli-Z门实现蚂蚁的变异增加位置的多样性;最后根据移动后的新位置完成蚁群信息素强度和可见度的更新。由于优化过程统一在空间[0,2π]n进行,而与具体问题无关,对不同尺度空间的优化问题具有良好的适应性。以函数极值优化和控制器参数优化为例,仿真结果表明该方法的搜索能力和优化效率明显优于连续量子蚁群算法和标准遗传算法。Aiming at the shortcoming of ant colony optimization which was only suitable for the discrete optimization and the problem of slow convergence,this paper proposed a suitable quantum ant colony optimization algorithm for continuous optimization. The locations of ant were directly encoded by the phase of qubits in the proposed algorithm. First,determined the destination to move according to the select probability constructed by the pheromone information and heuristic information,and then updated the qubits of ant by quantum rotation gates to achieve ant moving,and mutated by quantum Pauli-Z gates to increase diversity of ants. Finally,updated the pheromone information and the heuristic information in the new location of ants. As performed the optimization in [0,2π]n ,which had nothing to do with the specific issues,hence,the proposed method had good adaptability for a variety of optimization problems. With applications of function extremum and controller parameters optimization,the simulation results show that the proposed algorithm is superior to the continuous quantum ant colony algorithm and the standard genetic algorithm in both search capability and optimization efficiency.

关 键 词：量子计算 蚁群算法 相位编码 连续优化 优化算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]