论文部分内容阅读
对腹板易发生后屈曲的简支蜂窝钢梁的承载力进行研究,讨论分析梁承载力和失效模型的非线性有限元法的准确性。由于非线性有限元计算量很大,故基于有限元进行参数研究,提出蜂窝钢梁腹板的后屈曲临界荷载的经验公式。另采用传统的反向传播神经网路和自适应神经模糊推理系统方法进行求解,并对比传统有限元分析结果,验证经验公式、反向传播神经网络法、自适应神经模糊推理系统法的准确性。结果表明:反向传播神经网络法和自适应神经模糊推理系统法比经验方程更准确。
The bearing capacity of simply supported honeycomb steel beams with post-buckling webs is studied, and the accuracy of the nonlinear finite element method for analyzing the bearing capacity and failure modes of beams is discussed. Due to the large amount of nonlinear finite element calculation, the empirical formula of the post-buckling critical load of honeycomb girder webs is proposed based on the finite element method. In addition, the traditional backpropagation neural network and adaptive neuro-fuzzy inference system are used to solve the problems. Compared with the traditional finite element analysis, the empirical formula, the backpropagation neural network and the accuracy of the adaptive neuro-fuzzy inference system . The results show that the backpropagation neural network method and the adaptive neuro-fuzzy inference system method are more accurate than the empirical equation.