水文模型参数优选的改进粒子群优化算法

Improved particle swarm optimization algorithm for parameter optimization of hydrologic models

针对标准粒子群算法的早熟收敛问题,提出了一个提高算法性能的改进途径,即引入动态改变惯性权重策略和混沌思想,在两个方面同时改进以提高粒子群算法的收敛速度和克服局部极值的能力.对两个函数进行寻优测试表明,改进后的粒子群算法收敛速度、精度以及全局搜索能力均优于标准粒子群算法.最后将提出的改进粒子群算法应用于新安江模型进行参数优选,应用结果表明,该算法具有较强的可行性与实用性.

Aiming at the problem of premature convergence in the particle swarm optimization（PSO） algorithm,an improved algorithm is put forward.In the algorithm,the dynamic inertia weight is proposed and the chaos theory is introduced.By combining these two methods,the convergence rate of the algorithm and the capability of overcoming local extreme value are increased.Experiments on two functions show that the improved algorithm is prior to traditional PSO in convergence rate,precision and global searching ability.The improved PSO is applied to optimize parameters of XAJ model.Application results show that this algorithm has good feasibility and practicability.