基于支持向量机方法的土壤水分特征曲线预测模型  

Prediction Model of Soil Moisture Characteristic Curve Based on Support Vector Machine

作　　者：李彬楠 樊贵盛[1] LI Bin-nan;FAN Gui-sheng(Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学水利科学与工程学院,太原030024

出　　处：《节水灌溉》2019年第1期108-111,117,共5页Water Saving Irrigation

基　　金：国家自然科学基金资助项目"区域尺度上土壤入渗参数多元非线性传输函数研究"(40671081);山西省农田节水技术开发服务推广站项目"农艺节水措施对作物产量及ET的影响效果研究"

摘　　要：在山西省黄土高原区进行野外试验获取土壤样品,经室内试验测定,最终获得土壤样品的水分特征曲线以及理化参数,建立了基于支持向量机的Van-Genuchten预测模型。研究与分析的结果:输入变量选用了5个土壤基本理化参数(土壤黏粒、粉粒、密度、有机质和全盐量),输出变量为Van-Genuchten模型的参数α、n,对土壤水分特征曲线Van-Genuchten模型的参数进行预测并取得良好的结果。所建立的支持向量机预测模型下,Van-Genuchten模型参数α、n的预测值与检验值的平均相对误差都小于4%,建模与检验样本都具有较高的精确度。研究成果有助于丰富黄土地区的土壤水分特征曲线理论研究。Field experiments were carried out in the Loess Plateau of Shanxi Province to obtain soil samples.The soil characteristics curves and physical and chemical parameters were obtained by laboratory tests.The Van-Genuchten prediction model based on support vector machine was established.The input variables were selected from five basic soil physical and chemical parameters(soil clay,silt,bulk density,organic matter and total salt),and the output variables were parameters α and n of the Van-Genuchten model.The parameters of the curve Van-Genuchten model were predicted and good results were obtained.Under the support vector machine prediction model established in this paper,the average relative error between the predicted values and the test values of Van-Genuchten model parameters αand n are less than 4%,and the modeling and test samples have high precision.The research results will help to enrich the theoretical study of soil water characteristic curves in loess areas.

关 键 词：土壤理化参数 土壤水分特征曲线 支持向量机 Van-Genuchten模型参数 

分 类 号：S152[农业科学—土壤学] TV93[农业科学—农业基础科学]