多阶自复位钢桁框结构非线性位移比概率模型  

Probability model for nonlinear displacement ratio of multi-stage self-centering steel truss frame structures

作　　者：刘艳芝 施涛涛 张还阳 颜阳 周忠发[3] 柯珂 LIU Yanzhi;SHI Taotao;ZHANG Huanyang;YAN Yang;ZHOU Zhongfa;KE Ke(College of Civil Engineering,Hunan University,Changsha 410082,China;China Railway Shanghai Design Institute Group Co.,Ltd.,Shanghai 200070,China;Beijing Institute of Architectural Design,Beijing 100045,China;College of Civil Engineering,Chongqing University,Chongqing 400045,China)

机构地区：[1]湖南大学土木工程学院,长沙410082 [2]中铁上海设计院集团有限公司,上海200070 [3]北京市建筑设计研究院有限公司,北京100045 [4]重庆大学土木工程学院,重庆400045

出　　处：《东南大学学报(自然科学版)》2025年第2期444-451,共8页Journal of Southeast University:Natural Science Edition

基　　金：国家自然科学基金资助项目(52178111);湖南省自然科学基金资助项目(2023JJ30142);长沙市自然科学基金资助项目(kq2208033).

摘　　要：为提升钢桁框结构的抗震韧性,提出了多阶自复位钢桁框结构的概念,并量化了结构在地震作用下的非线性位移需求。基于多阶自复位钢桁框结构的等效单自由度(SDOF)体系,结合182条速度脉冲型地震动,采用非线性时程分析建立了结构的非线性位移比谱,讨论了周期、阻尼比和滞回参数对非线性位移比的影响。利用机器学习中的XGBoost算法建立了非线性位移比的概率预测模型。同时,为解决机器学习中的“黑箱”问题,利用SHAP和PDP方法对概率模型的预测结果进行了解释。结果表明,非线性位移比对周期、阻尼比和滞回参数敏感。概率预测模型不仅具有较高的预测精度,还能直观反映各参数的影响规律,可用于构建多阶自复位钢桁框结构基于位移的抗震设计方法。The multi⁃stage self⁃centering steel truss frame structure is proposed to enhance the seismic resil⁃ience of steel truss frame structures,and the nonlinear displacement demand of the novel structure is quanti⁃fied.Based on the equivalent single degree of freedom(SDOF)system representing the novel structure and 182 velocity pulse⁃like ground motions,the nonlinear displacement ratio spectrum is developed through nonlin⁃ear dynamic analyses.The influences of period,damping ratio,and hysteretic parameters on the nonlinear dis⁃placement ratio are discussed.A probabilistic prediction model for the nonlinear displacement ratio is devel⁃oped using the XGBoost algorithm in machine learning.In addition,to address the“black⁃box”issue in ma⁃chine learning,the prediction results of the model are interpreted based on shapley additive explanations(SHAP)and partial dependence plots(PDP)methods.The results indicate that the nonlinear displacement ra⁃tio is sensitive to period,damping ratio,and hysteretic parameters.The probabilistic prediction model not only exhibits high predictive accuracy but also intuitively reflects the influence patterns of various parameters.This model can be applied to develop a displacement⁃based seismic design method for the multi⁃stage self⁃centering steel truss frame structure.

关 键 词：多阶自复位钢桁框结构 抗震韧性 非线性位移比 脉冲型地震动 机器学习 

分 类 号：TU391[建筑科学—结构工程]