哈尔滨商业大学学报（自然科学版）2025，Vol.41Issue(6)：716-722,7.
基于机器学习的RC框架隔震结构地震响应预测
Seismic response prediction of RC frame isolated structures based on machine learning
孙意武 1肖梅玲 1庞鑫慧 2高志超 1吴正 1汪厚明1
作者信息
- 1. 云南大学 建筑与规划学院,昆明 650091
- 2. 云南大学 信息学院,昆明 650091
- 折叠
摘要
Abstract
To address the issues of high computational cost and low efficiency in traditional elastoplastic time-history analysis for seismic response assessment of RC frame isolated structures,a machine learning-based method was proposed to achieve rapid and accurate prediction of key engineering demand parameters.A seismic response database encompassing a wide range of ground motions was established through nonlinear time-history analysis.Using ground motion characteristics,structural properties,and isolation bearing parameters as input variables,and the maximum inter-story displacement angle as the prediction target,two ensemble learning algorithms random forest and extreme gradient boosting were employed for model training and testing.The relative importance of input features was systematically analyzed,and the influence of training dataset size on model prediction performance was investigated.The results demonstrated that both random forest and extreme gradient boosting models exhibited strong generalization capability and high prediction accuracy.Furthermore,model performance improved significantly with the expansion of the dataset size,and the damping ratio of the isolated structure showed the highest relative importance score among all input parameters.关键词
机器学习/RC框架/隔震结构/地震响应/预测/特征重要性Key words
machine learning/RC frame/isolated structures/seismic response/prediction/feature importance分类
建筑与水利引用本文复制引用
孙意武,肖梅玲,庞鑫慧,高志超,吴正,汪厚明..基于机器学习的RC框架隔震结构地震响应预测[J].哈尔滨商业大学学报（自然科学版）,2025,41(6):716-722,7.
