改进的粒子群算法求解Van Genuchten方程参数

Improved particle swarm optimizer for solving parameters of Van Genuchten model

Van Genuchten模型是研究土壤水力学性质应用最广泛的模型.运用该模型的关键是4个参数的求解.为了精确地求解这些参数,提出一种基于动态邻居和局部搜索的粒子群算法(DNLPSO)该算法具有如下特征:根据粒子的运行,为每个粒子动态的构建邻居;每个粒子的速度不是由它的邻居中最好运行的粒子来调整,而是由它的所有邻居共同调整;引入拟牛顿(Quasi-Newton)算法来提升局部搜索能力,以加快收敛速度.实验结果表明DNLPSO求解Van Genuchten模型参数的精度高于其他方法.

Van-Genuchten model is widely implemented when researching soil hydraulic characteristics. Its four parameters are the key to application of this model. In order to solve these parameters exactly, this paper presents an improved PSO based on dynamic neighbor topology and local search （DNLPSO） to solve these parameters. The algorithm shows the following characteristics： The neighborhood of each particle is dynamically changed in terms of the performances of the particles; the velocity of each particle is not adjusted by the best performing particle in its neighborhood, but by all particles in its neighborhood including itself; the Quasi-Newton method is combined to improve its local searching ability to speed convergence velocity. The results demonstrated good performance of DNLPSO in solving the parameters compared with the other methods.