摘要
为了提高测量土壤水分Van Genuchten方程求解的精度,提出平行竞争PSO算法。首先在PSO算法基础上,当粒子群最大半径值小于某个阈值时,竞争才被触发,同时最差粒子被重置,粒子被重置的比例随着迭代次数增加而非线性减少;接着粒子群分成若干子群,子群的群平均适应度与原始粒子群平均适应度相差不能小于设定的阈值,引入不同的共享因子对子群与子群、粒子与子群之间进行信息共享动态调节;最后粒子适应度函数由Van Genuchten方程参数构成,给出了算法流程。实验仿真显示本文算法对测试函数求解具有收敛速度较快、解精度较高的特点,测量粉壤土脱湿数据的相对误差最大为5%,吸湿数据的相对误差最大为4%,相比其他算法都较小。
In order to improve the accuracy of the solution of the Van Genuchten equation of soil moisture measurement, a parallel competitive PSO algorithm is proposed. Firstly, based on the PSO algorithm, when the maximum radius value of the particle swarm is less than a certain threshold, the competition is triggered, and the worst particle is reset. Then, the particle swarm is divided into several subgroups, and the difference between the average fitness of the subgroup and the average fitness of the original particle swarm can not be less than the set threshold. Different sharing factors are introduced to dynamically adjust the information sharing between the subgroups and the subgroups and between the particles and the subgroups. Finally, the particle fitness function is composed by the parameters of Van Genuchten equation, and the algorithm flow is given. Experimental simulation shows that the algorithm in this paper has the characteristics of fast convergence speed and high solution accuracy in solving the test function, and the maximum relative error of the measured data of dehumidification of powdery loam is 5%, and the maximum relative error of the moisture absorption data is 4%, which is smaller than other algorithms.
作者
刘云潺
张慧宁
LIU Yun chan;ZHANG Hui ning(Yellow River Conservancy Technical Institute, Kaifeng 475004, Henan Province, China)
出处
《节水灌溉》
北大核心
2019年第7期82-86,共5页
Water Saving Irrigation
基金
河南省科技厅鉴定项目(预科鉴委字[2013]第201号)
关键词
平行
竞争
土壤
水分
parallel
competition
soil
moisture
