摘要
针对方钢管混凝土偏心受压柱钢管和混凝土材料受力状态复杂,各种因素对极限承载力的影响难以独立精确描述的问题,通过神经网络自学习、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,建立了预测钢管混凝土极限承载力的神经网络模型.以现有的方钢管偏心受压柱试验数据为样本,训练了一个四层BP网络模型,用模型计算了偏心受压方钢管混凝土柱的极限承载力.对6组实验数据进行了预测,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,对钢管混凝土偏心短柱进行承载力计算.
Both steel tube and filled concrete of concrete-filled square steel tube under eccentric load are in complicated stress status, and the influence of various factors on the ultimate beating capacity is difficult to be ascertained accurately. On the other hand, the relationship between input and output variables can be obtained by self-studying, self-organizing, self-adapting and nonlinear mapping of neural network. Thus, it is reasonable to build the neural network model for predicting the ultimate bearing capacity of concrete-filled square steel tube columns. A four-layer BP network model was trained based on the existing experimental data of square steel tube eccentrically loaded columns. The model was used to calculate he ultimate bearing capacity of concrete-filled square steel tube columns under eccentric load. The prediction was performed for six groups of experimental data. The results show that the predicted values are in good agreement with the measured ones, and the prediction precision is higher. The present method can be taken as an auxiliary means for actual structure design.
出处
《沈阳工业大学学报》
EI
CAS
2009年第1期116-120,共5页
Journal of Shenyang University of Technology
基金
教育部留学回国人员科研基金资助项目(20046293)
关键词
方钢管混凝土
神经网络
偏心受压柱
承载力
非线性映射
concrete-filled square steel tube
neural capacity
nonlinear mapping network
eccentrically loaded column
bearing
