Bearing Capacity Prediction of Inclined Loaded Strip Footing on Reinforced Sand by Ann
Bearing Capacity Prediction of Inclined Loaded Strip Footing on Reinforced Sand by Ann
复制标题
Ann 法对加筋砂倾斜加载条形基础承载力预测
DOI：
10.1007/978-3-319-63570-5_9
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
B. Das
中科院分区：
文献类型：
--
作者：
R. Sahu;C. Patra;N. Sivakugan;B. Das
对位于多层土工格栅加筋密砂和松散砂上的条形基础进行了倾斜荷载作用下的室内模型试验。基于室内模型试验结果，建立了承载力折减系数的神经网络模型。利用人工神经网络计算得到的折减系数，可以由条形基础在中心竖向荷载作用下的极限承载力估算出条形基础在中心倾斜荷载作用下的极限承载力。进行了全面的敏感性分析，以找出影响折减系数的重要参数。重点给出了神经解释图的构造，基于神经网络模型中开发的权重，以确定输入参数对输出的直接或反向影响。基于神经网络模型的训练权值，建立了神经网络模型方程。人工神经网络（ANN）的结果与室内模型试验结果进行了比较，这些结果是一致的。
Laboratory model tests have been conducted on a strip foundation resting over multi-layered geogrid-reinforced dense and loose sand subjected to inclined load. Based on the laboratory model test results, a neural network model is developed to estimate the reduction factor for bearing capacity. The reduction factor obtained by ANN can be used to estimate the ultimate bearing capacity of a strip foundation subjected to centric inclined load from the ultimate bearing capacity of the same foundation under centric vertical loading. A thorough sensitivity analysis was carried out to find out the important parameters affecting the reduction factor. Emphasis was given on the construction of neural interpretation diagram, based on the weights developed in the neural network model, to determine the direct or inverse effect of input parameters to the output. An ANN model equation is developed based on trained weights of the neural network model. The results from artificial neural network (ANN) were compared with the laboratory model test results and these results are in good agreement.