摘要
耗能能力是衡量钢筋混凝土柱(RC柱)抗震性能的重要指标,有限元模拟和试验法存在耗时耗力、成本高等问题。为此,文中提出了基于GRNN神经网络的RC柱耗能能力预测方法。从PEER数据库选取212组矩形柱,采用等效粘滞阻尼系数作为评价耗能能力的指标,利用MATLAB软件计算RC柱的滞回环面积和等效粘滞阻尼系数,通过SPSS显著性试验的方法评估影响RC柱等效粘滞阻尼系数的主要因素,建立RC柱等效粘滞阻尼系数神经网络预测模型。结果表明GRNN神经网络具有很好的表现,模型在测试集上的平均相对误差为21.33%,决定系数为0.792,等效粘滞阻尼系数预测值曲线与真实值曲线接近,具有良好的应用价值。
Energy dissipation capacity is an important indicator of the seismic performance of reinforced concrete columns(RC columns),and finite element simulation and test methods are time-consuming and costly.To this end,this paper proposes a method for predicting the energy dissipation capacity of RC columns based on GRNN neural networks.The equivalent viscous damping coefficient is used as an index to evaluate the energy dissipation capacity.The hysteresis loop area and equivalent viscous damping coefficient of RC columns are calculated by using MAT⁃LAB software,and the main factors affecting the energy dissipation capacity of RC columns are evaluated by means of SPSS significance experiments to establish a neural network prediction model for the energy dissipation capacity of RC columns.The results show that the GRNN neural network has good performance,the average relative error of the model on the test set is 21.33%,the coefficient of determination is 0.792,and the curve of the predicted value of the equivalent viscous damping coefficient is close to the curve of the true value,which has good application value.
作者
周祥
曾森
董一韩
ZHOU Xiang;ZENG Sen;DONG Yihan(School of Civil Engineering,Qingdao University of Technology,Shandong Qingdao 266525,China)
出处
《低温建筑技术》
2023年第4期116-120,共5页
Low Temperature Architecture Technology
关键词
钢筋混凝土柱
GRNN神经网络
耗能能力预测
等效粘滞阻尼系数
reinforced concrete column
GRNN neural network
energy consumption capacity prediction
equiva⁃lent viscous damping coefficient
作者简介
周祥(1997-),男,山东青岛人,硕士,现从事土木工程信息化方向研究;通信作者:曾森(1983-),男,广州人,副教授,现从事土木工程信息化、计算力学方向研究。
