非饱和残坡积土强度随含水量变化的改进BP神经网络预测研究  

A prediction model of unsaturated residual slope soil with water content changes based on improved BP Neural Network

作　　者：肖治宇[1] 陈昌富[1] 彭钊 

机构地区：[1]湖南大学岩土工程研究所,长沙410082 [2]深圳高速工程顾问有限公司,深圳518034

出　　处：《水文地质工程地质》2011年第2期79-83,101,共6页Hydrogeology & Engineering Geology

基　　金：国家自然科学基金资助项目(50878082);交通部西部交通科技项目(200631880237);湖南省自然科学基金重点项目(09JJ3104)

摘　　要：水是导致残坡积土边坡滑坡的最主要原因,因此研究水对残坡积土力学性质的影响具有重要的理论和工程实际意义。本文在普通三轴仪上进行非饱和残坡积土的强度试验,并基于试验结果建立了非饱和残坡积土强度随含水量变化的改进BP神经网络预测模型。模型计算结果与试验结果对比分析表明,该神经网络预测模型具有较高的拟和精度和良好的泛化能力,能较好地预测试验含水量范围内任意含水量下对应的非饱和残坡积土应力-应变关系,从而能够弥补室内试验由于设计方案等自身缺陷引起的不足。研究结果可为工程应用提供参考。A study of the mechanical properties of an unsaturated residual slope soil is of the engineering significance.General laboratory triaxial tests are carried out to obtain the stress-strain relationship between the unsaturated residual slope soil and the changes in water content.Improved BP neural network prediction model is established based on test data.The results from comparing predicted values and measured values show that this network prediction model has good fitting precision and good generalization ability.The method can better predict the unsaturated residual slope soil strength under any corresponding water content.The results can also provide a reference for engineering applications.

关 键 词：非饱和残坡积土 吸力 含水量 三轴试验 BP神经网络 

分 类 号：TU411.3[建筑科学—岩土工程]