计算机工程与应用 ›› 2013, Vol. 49 ›› Issue (11): 58-60.

• 理论研究、研发设计 • 上一篇    下一篇

基于拟合与粒子群优化的VG方程参数估计

曹怀火1,欧阳艾嘉2,艾海男3   

  1. 1.池州学院 数学与计算机科学系,安徽 池州 247000
    2.湖南大学 计算机与通信学院,长沙 410082
    3.重庆大学 三峡库区生态环境教育部重点实验室,重庆 400045
  • 出版日期:2013-06-01 发布日期:2013-06-14

Parameters estimation of VG equation based on fitting method and particle swarm optimization

CAO Huaihuo1, OUYANG Aijia2, AI Hainan3   

  1. 1.Department of Mathematics and Computer Science, Chizhou College, Chizhou, Anhui 247000, China
    2.School of Computer and Communication, Hunan University, Changsha 410082, China
    3.Laboratory of Three Gorges Reservoir Region’s Eco-Environment, Ministry of Ducation, Chongqing University, Chongqing 400045, China
  • Online:2013-06-01 Published:2013-06-14

摘要: 考虑到粒子群算法受初值影响,易于产生局部最优解的缺陷,将lsqcurvefit拟合方法与粒子群算法相结合,提出一种新的混合型粒子群优化算法,用于Van Genuchten方程参数估计得到了较好的结果。数值实验结果分析表明,该算法在参数估计中求解精度高、收敛速度快、寻优能力强,而且不需要给出参数的初始值,是一种值得推广的方法。

关键词: lqcurvefit拟合方法, 粒子群算法, Van Genuchten方程, 参数估计

Abstract: Taking into the particle swarm optimization is affected by the initial values and produces easily the local optimal solutions, based on lsqcurvefit fitted method and particle swarm optimization, a new hybrid algorithm of particle swarm optimization is proposed to estimate the parameters for Van Genuchtem equation and get better results. The result of a case study shows the algorithm is characterized with high accuracy, fast convergence and high robustness in the parameters estimation, moreover, it does not need the initial parameter value. Therefore, this method should be promoted.

Key words: lsqcurvefit fitting method, particle swarm optimization, Van Genuchten equation, parameters estimation