黄土场地大型浸水试验湿陷沉降预测的Elman方法  

The Elman Method for Forecasting the Large Water Immersion Collapsibility Test in the Loess Field

作　　者：韩晓萌[1] 王家鼎[1] 谷天峰[1] 薛建功[2] 

机构地区：[1]西北大学地质学系大陆动力学国家重点实验室,陕西西安710069 [2]陕西省公路局,陕西西安710068

出　　处：《地下水》2008年第6期111-113,共3页Ground water

基　　金：国家自然科学基金(4057215-7);高等学校博士学科点专项科研基金(20050697013)

摘　　要：收集郑西高速铁路湿陷性黄土场地现场大型浸水试验资料,运用MATLAB建立了黄土场地湿陷沉降的Elman神经网络预测模型。通过对样本的训练和预测,表明该模型预测的结果与实际湿陷沉降比较接近,是一种比较理想的预测方法。The paper collects data about the field large water immersion test in the wet sedimentation loess of ZhengXi High - Speed Railway, making of the MATLAB to establish a model of Etman neural network to forecast loess field collapsibility settlement. Through to training and forecasting the samples, compared with the forecasted result, it is shown that the forecasted results are approach, so it＇s one perfect method.

关 键 词：ELMAN神经网络 黄土湿陷 沉降 高速铁路地基 浸水试验 

分 类 号：P642.131[天文地球—工程地质学]