方钢管砼柱轴心受压承载力的神经网络方法研究

Investigation on bearing capacity of square CFT columns under axial load based on neural network

方钢管混凝土轴心受压柱的钢管和混凝土材料的受力状态复杂,各种因素对极限承载力的影响难以独立地精确描述.神经网络通过自学习、自组织、自适应和非线性映射,可找到输入、输出变量之间的关系,因此适用于极限承载力的预测.以现有的方钢管混凝土柱轴心受压试验数据为样本,训练了一个三层的BP网络模型,建立了预测轴心受压方钢管混凝土柱极限承载力的神经网络模型.对8组实验数据进行预测的结果表明:预测值与试验值吻合良好,精度较高,该方法可作为实际结构设计的一种辅助手段.

Steel tube and filled concrete of condition. The influence of various factors on square CFT columns under axial load are in complicated stress mechanical performance is difficult to ascertain accurately. On the other hand, neural network is good at obtaining the relationship between input and output variables by self-learning, self-organizing, self-adapting and nonlinear mapping. Therefore, it is suitable to use neural network for calculating the bearing capacity of square CFT columns. In this paper a three-layer back-propagation model of network was trained according to experimental data of square CFT columns under axial load. A neural network model for axially loaded square CFT columns was set up. The model was verified by eight sets of experimental data. Results show the predicted values are in good agreement with test values. Precision in calculation is good enough to be used as an auxiliary method for structure design.