冻融土壤Philip入渗模型参数的BP预报模型

发布时间：2018-03-08 20:12

本文选题：冻融土壤　切入点：Phillip入渗模型　出处：《节水灌溉》2017年05期 　论文类型：期刊论文

【摘要】：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。
[Abstract]:In order to improve the irrigation quality and effect of winter and spring storage irrigation in seasonal frozen soil region, a series of soil infiltration tests were carried out during the freezing and thawing period in seasonal frozen soil region, and a large number of soil infiltration test data under natural freezing and thawing conditions were obtained. The parameters of Philip infiltration model under different freezing and thawing conditions were obtained, and the prediction model of BP neural network between infiltration model parameters and basic physical and chemical parameters of soil was established. The prediction model was realized based on soil volume moisture content, clay content and soil density. The prediction of Philip infiltration model parameters such as soil temperature and water temperature of irrigation water. The average relative error between the predicted value and the measured value is controlled within 7%. It is feasible to predict the parameters of Philip infiltration model by using the conventional physical and chemical parameters of soil under the condition of freeze-thawing soil, which can provide strong support for the determination of technical parameters of seasonal freeze-thaw soil irrigation.
【作者单位】：太原理工大学水利科学与工程学院;
【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)
【分类号】：S152.7

【相似文献】