摘要
提出了基于人工神经网络预测钢筋混凝土柱峰值承载力的方法。该方法采用5个设计参数作为神经网络的输入:混凝土强度、轴压比、剪跨比、纵筋配筋率和纵筋屈服强度。为验证该方法的可行性与有效性,基于PEER 154组实验数据,利用神经网络模型对矩形混凝土柱的峰值承载力进行预测并与经验模型的预测结果进行比较。比较分析结果表明:神经网络模型预测结果与实验结果吻合度远高于其他经验模型;同时也表明神经网络为精确预测结构在地震作用下的性能提供了一种新方法。
An artificial neural network method for the prediction of the peak load capacity of RC columns was presented in this paper. In this method,5 design parameters: concrete compression stress,axial load ratio, aspect ratio, longitudinal reinforcement ratio and longitudinal reinforcement yield strength were chosen as inputs of the ANN. In order to demonstrate the feasibility and effectiveness of proposed method, the ANN models were applied to predict the peak load capacity of rectangular RC columns using 154 sets of experimental data provided by PEER. Furthermore, the predicted results were compared with empirical model results. Comparative analysis showed that prediction degree of agreement with the experimental results of ANN models is much better than that of other empirical prediction models. Meanwhile, the result reveals that the proposed method provides a novel way for accurately estimating structural performance under earthquake.
作者
林庄慧
唐和生
李大伟
薛松涛
LIN Zhuanghui;TANG Hesheng;LI Dawei;XUE Songtao(Disaster Migtigation for Structure,Tongji University,Shanghai 200092,China)
出处
《结构工程师》
北大核心
2019年第1期174-179,共6页
Structural Engineers
