基于支持向量机方法的土壤水分特征曲线预测模型
|
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.
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.
引文
[1]赵雅琼,王周锋,王文科.不同粒径下土壤水分特征曲线的测定与拟合模型的研究[J].中国科技论文,2015,10(3):287-290.
[2]谭霄,伍靖伟,李大成,等.盐分对土壤水分特征曲线的影响[J].灌溉排水学报,2014,33(z1):228-232.
[3]高惠嫣,杨路华.不同质地土壤的水分特征曲线参数分析[J].河北农业大学学报,2012,35(5):000129-132.
[4]Garder W R,Hillel D,Benyamini Y.Post irrigation movement of soil water:1.Redistribution[J].Water Resocurce Research,1970,6(3):851-861.
[5]Russo D.Determining soil hydraulic properties by Parameter estimation:On the selection of a model for the hydraulic properties[J].Water Resour.Research.,1988,24(3):453-459.
[6]Fredlund D G,Xing A.Equations for the soil-water characteristic curve[J].Canadian Geotechnical Journal,1994,31(4):521-532.
[7]Van Genuchten M T.A closed-form equation for predicting the hydraulic conductivity of unsaturated soils[J].Soil Science Society of A-merica Journal,1980,44(5):892-898.
[8]李晓婷,刘勇,王平.基于支持向量机的城市土壤重金属污染评价[J].生态环境学报,2014,(8):1 359-1 365.
[9]刘文霞,徐晓波,周樨.基于支持向量机的纯电动公交车充/换电站日负荷预测[J].电力自动化设备,2014,34(11):41-47.
[10]黄华坚.支持向量机法计算高土石坝坝坡稳定可靠度[J].人民黄河,2015,37(12):125-127.
[2]谭霄,伍靖伟,李大成,等.盐分对土壤水分特征曲线的影响[J].灌溉排水学报,2014,33(z1):228-232.
[3]高惠嫣,杨路华.不同质地土壤的水分特征曲线参数分析[J].河北农业大学学报,2012,35(5):000129-132.
[4]Garder W R,Hillel D,Benyamini Y.Post irrigation movement of soil water:1.Redistribution[J].Water Resocurce Research,1970,6(3):851-861.
[5]Russo D.Determining soil hydraulic properties by Parameter estimation:On the selection of a model for the hydraulic properties[J].Water Resour.Research.,1988,24(3):453-459.
[6]Fredlund D G,Xing A.Equations for the soil-water characteristic curve[J].Canadian Geotechnical Journal,1994,31(4):521-532.
[7]Van Genuchten M T.A closed-form equation for predicting the hydraulic conductivity of unsaturated soils[J].Soil Science Society of A-merica Journal,1980,44(5):892-898.
[8]李晓婷,刘勇,王平.基于支持向量机的城市土壤重金属污染评价[J].生态环境学报,2014,(8):1 359-1 365.
[9]刘文霞,徐晓波,周樨.基于支持向量机的纯电动公交车充/换电站日负荷预测[J].电力自动化设备,2014,34(11):41-47.
[10]黄华坚.支持向量机法计算高土石坝坝坡稳定可靠度[J].人民黄河,2015,37(12):125-127.