Abstract: An adaptive ant colony algorithm is presented for the optimization of multi-minimum continuous function. By dividing the space of solution into subdomains and dynamically adjusting the strategy of selection of the paths and the strategy of the trail information updating based on the distribution of the solutions, the algorithm can find the subdomain in which the solution is located, and then determines the specific value of the solution within the subdomain. The simulation results show that the algorithm has many good performances such as avoiding local optimum, high precision solution, quick convergence and good reliability, which are better than those of simple genetic algorithm and clonal selection algorithm.

摘要： 针对多极值连续函数优化问题，提出一种自适应蚁群算法。该方法将解空间划分成若干子域，根据蚂蚁在搜索过程中所得解的分布状况动态的调节蚂蚁的路径选择策略和信息量更新策略，求出解所在的子域，然后在该子域内确定解的具体值。仿真结果表明该算法具有不易陷入局部最优、解的精度高、收敛速度快、稳定性好等优点，其性能优于基本遗传算法以及克隆选择算法。