改进的蚂蚁算法及其在暴雨强度公式参数优化中的应用

蚂蚁算法是解决组合优化问题的一种优秀的算法，但直接用于解决连续优化问题存在很大困难。本文通过引入带可变邻域搜索项的进化策略对蚂蚁算法进行了改进，然后将改进的蚂蚁算法应用到解决暴雨强度公式的参数优化这类连续问题中，并和其它优化方法得到的优化结果进行了比较。结果表明，改进的蚂蚁算法可以成功用于暴雨强度公式的参数优化，并且在实验采用的各种优化算法优化参数得到的暴雨强度公式拟合原始数据的效果比较中只有免疫进化算法在优化过程中迭代次数和迭代规模都要大得多的情况下才和改进的蚂蚁算法差不多，而比其它的优化方法都要好。

Improved Ant Colony Algorithm and its Application to Parameters Optimization in Storm Intensity Formula

Evolutionary strategy with searching in variable neighborhood parameter is taken to improve ant colony algorithm (ACA). After that, the improved ACA is applied to optimizing the parameter of storm intensity formula and the optimizing result is compared with those of other optimizing algorithms. The results show the improved ACA can be used to solve the continued problem, and the formula whose parameter is optimized by the improved ACA can simulate the primary data better than those whose parameter are optimized by other optimizing methods except the immune evolutionary algorithm (IEA), and the improved ACA can get almost the same result with less optimization scale and shorter optimization time than IEA.