摘要
探讨了用神经网络模型来预测组合梁中栓钉连接件抗剪强度的可行性。应用神经网络的ＢＰ模型，选择混凝土的抗压强度和弹性模量、栓钉的极限抗拉强度、栓钉的直径和长度作为关键的影响因素，提出了基于神经网络理论的栓钉连接件抗剪强度的预测方法。分析结果表明，神经网络方法比其他的经验计算方法更接近试验结果。
This paper investigates the feasibility of using neural networks to evaluate the ultimate shear strength of stud connectors.Based on the back propagation model of the neural network, details of the neural network methodology are presented to predict ultimate shear strength of stud connectors, where the compressive strength and modulus of the concrete, the tensile strength of the stud material, the diameter and height of the shank of the stud are chosen as the critical factors of the nonlinear variables. The analysed results show that the neural network pedictions have a better agreement with the experimental data of the pushout tests than the predictions using other conventional methods.
出处
《南京建筑工程学院学报》
1997年第1期7-12,共6页
Journal of Nanjing Architectural and Civil Engineering Institute(Natural Science)
关键词
组合梁
连接件
神经网络
抗剪强度
composite beams
stud connectors
neural networks
shear strength
