方钢管混凝土短柱轴心受压承载力的神经网络模拟

由于方钢管混凝土的侧向约束机构复杂,对方钢管混凝土柱强度承载力的计算至今仍没有一种统一的方法。本文拟采用神经网络方法对轴心受压方钢管混凝土短柱的承载力进行模拟。以混凝土抗压强度、钢管的屈服强度、套箍指标、截面尺寸和宽厚比等五个参数为网络输入,以构件的极限承载力为网络输出,构建多层前馈神经网络来描述它们之间的非线性关系。利用55组试验数据对网络进行训练和测试,并将其预测值与三种承载力计算模型的预测值进行比较。对比结果表明本文建立的神经网络模型对55组试验数据给出了最好的模拟精度,可作为预测方钢管混凝土柱承载能力的一种新方法。

Neural network modeling of bearing capacity of axially loaded concrete-filled square steel tubular short columns

Due to the complexity of the confinement mechanism in concrete-filled square steel tubes(CFST),there is still no unified method for calculating the bearing capacity of CFST columns.The application of artificial neural network to predict the ultimate bearing capacity of CFST short columns under axial loading is explored.Input parameters consisted of concrete compressive strength,yield strength of steel tube,confinement index,sectional dimension and width-tothickness ratio.The ultimate bearing capacity was the only output parameter.A multi-layer feed-forward neural network was used to describe the nonlinear relationships between the input and output variables.Fifty-five experimental data of CFST short columns under axial loading were used to train and test the neural network.A comparison study between the neural network model and three analytical models was also carried out.The study shows that neural network model possesses good accuracy and it can be a new method for predicting the ultimate strength of axially loaded CFST short columns.