冻融土壤Philip入渗模型参数的BP预报模型

为提高季节性冻土区冬、春季储水灌溉的灌水质量和效果,在季节性冻土区冻融期间进行了大田冻融土壤的系列入渗试验,获取了自然冻融条件下的大量土壤入渗试验数据,拟合得到了不同冻融条件下的Philip入渗模型参数,建立了入渗模型参数与土壤基本理化参数间的BP神经网络的预报模型,实现了基于土壤体积含水率、黏粒含量、土壤密度、土壤温度以及灌溉用水水温等基本理化参数对Philip入渗模型参数稳渗率A、吸渗率S的预报。所得到的预测值与实测值之间相对误差的平均值控制在7%以内。研究表明,利用冻融土壤条件下土壤常规理化参数对Philip入渗模型参数进行预报是可行的,可为季节性冻融土壤灌溉技术参数的确定提供有力支撑。

The BP Forecasting Model Based on the Philip Infiltration Model Under Freezing Thawing Condition

In order to improve the quality and effect of the seasonal frozen distractions in spring and winter storage irrigation, a series of experiment during the seasonal frozen distractions has been conducted, and a lot of soil infiltration data under nature conditions are obtained.By fitting the Philip infiltration model parameters, a predictive model based on BP neural network is established, and the goal of using basic physical and chemical parameters of the surface soil, including bulk density of soil, clay content, silt particle content, moisture content, irrigation water temperature and soil temperature, to forecast the parameters of Philip model, which are steady infiltration rate and sorptivety parameters, is achieved.The average relative error between the predicted and measured values is less than 7%.The study results show that it is feasible to use the soil conventional physical and chemical parameters to predict the soil water infiltration parameter in Philip model.The research can provide strong support for obtaining parameters of the seasonal freezing and thawing of soil irrigation techniques.