摘要
混凝土板的抗冲切承载力是结构设计中的重点与难题,基于所收集的475组钢筋混凝土板柱节点抗冲切承载力数据,建立了粒子群算法优化参数选择的支持向量机(PSO-SVR)预测模型,进行多个给定影响因素作为输入变量的RC板柱节点抗冲切承载力预测。将该模型与其他理论经验计算式的预测结果进行对比,结果表明,PSO-SVR模型的误差最小,预测结果最好。最后,利用SHAP解释模型对影响抗冲切承载力的不同指标进行敏感性分析,在给定的6个输入参数中,板的有效深度是最主要的影响因素,钢筋屈服强度的影响最弱。研究结果可以为钢筋混凝土板柱节点抗冲切承载力的预测提供新的思路与方法。
Punching shear bearing capacity of concrete slab is the key and difficult problem in structural design.Based on the collected 475 sets of reinforced concrete slab-column joint punching shear bearing capacity data,a support vector machine(PSO-SVR)prediction model is established for particle swarm optimization parameter selection.This prediction model is used to predict the punching shear bearing capacity of RC slab-column joints with multiple given influencing factors as input variables.The model is compared with the prediction results of other theoretical empirical formulas.The results show that the PSO-SVR model has the smallest error and the best prediction result.Finally,the SHAP interpretation model is used to analyze the sensitivity of different indexes affecting the punching shear bearing capacity.Among the given six input parameters,the effective depth of the slab is the most important factor,and the influence of the yield strength of the steel bar is the weakest.The research results can provide new ideas and methods for predicting the punching shear bearing capacity of reinforced concrete slab-column joints.
作者
蔡怡晴
费正玉
梁诗雪
CAI Yiqing;FEI Zhengyu;LIANG Shixue(School of Civil Engineering and Architecture,Zhejiang Sci-Tech University,Hangzhou 310018,China)
出处
《混凝土》
CAS
北大核心
2023年第2期34-39,共6页
Concrete
基金
国家自然科学基金(51808499)
浙江省自然科学基金(LY22E080016)。
关键词
钢筋混凝土板
抗冲切承载力
机器学习
经验计算式
敏感性分析
reinforced concrete slab
punching shear capacity
machine learning
empirical formula
sensitivity analysis
