基于随机森林和最小角回归的结构地震需求重要性度量分析

PDF(1483 KB)

振动与冲击 ›› 2019, Vol. 38 ›› Issue (4) : 115-120.

论文

王秀振1，钱永久1，宋帅2

作者信息 +

An analysis of the importance of structural earthquake demand based on random forest and least angle regression

WANG Xiuzhen1，QIAN Yongjiu1，SONG Shuai2

Author information +

文章历史 +

摘要

对带粘滞阻尼器的钢筋混凝土框架结构在El Centro地震波的作用下，采用OpenSEES软件进行了动力非线性时程分析，考虑了粘滞阻尼器的阻尼系数和刚度、钢筋的弹性模量和屈服强度、阻尼比、混凝土的抗压强度和弹性模量以及结构质量8个输入随机变量的影响，得到了框架结构的顶点位移、最大层间位移角以及基底剪力3种结构地震需求。提出将随机森林算法和最小角回归算法应用到结构地震需求的重要性度量分析中，得到了各个输入随机变量对3种结构地震需求的重要性性排序，并用Monte-Carlo数值模拟法进行了对比。结果表明，基于随机森林算法和最小角回归算法的结构地震需求重要性度量分析方法结果与Monte-Carlo数值模拟法基本一致，这两种方法是准确高效的方法，可以大大减少样本的数量。

Abstract

The nonlinear time history analyses of a reinforced concrete frame structure with viscous dampers under El Centro seismic wave were performed by OpenSEES software.The effect of damping coefficient and stiffness of viscous dampers,elastic modulus and yield strength of steel bars,damping ratio,compressive strength and elastic modulus of concrete and structure quality were considered.Three kinds of earthquake demand for reinforced concrete frame structures with viscous dampers were obtained respectively,namely,top displacement,the most story drift angle,and base shear.The application of random forest algorithm and least angle regression algorithm to the analysis of the importance of structural earthquake demand was proposed.The importance ranking to the three kinds of earthquake demand of each input random variable was obtained.The results were compared with the Monte-Carlo numerical simulation method.The results of structural seismic demand importance measure method based on random forest algorithm and least angle regression algorithm were basically the same as the Monte-Carlo numerical simulation method.These two methods are accurate and efficient,which can greatly reduce the number of samples.

导出引用

王秀振1，钱永久1，宋帅2. 基于随机森林和最小角回归的结构地震需求重要性度量分析[J]. 振动与冲击, 2019, 38(4): 115-120

WANG Xiuzhen1，QIAN Yongjiu1，SONG Shuai2. An analysis of the importance of structural earthquake demand based on random forest and least angle regression[J]. Journal of Vibration and Shock, 2019, 38(4): 115-120

参考文献

[1] 陈亮, 李建中.大跨径桥梁结构概率地震需求分析中地面运动强度参数的优化选择[J]. 振动与冲击, 2011, 30(10):91-97.
Chen Liang，Li Jianzhong. Optimal selection of ground motion intensity measures for probabilistic seismic demand analysis of long-span bridge structures[J]. JOURNAL OF VIBRATION AND SHOCK, 2011, 30(10):91-97.
[2] 刘骁骁,吴子燕，王其昂. 基于多维性能极限状态的概率地震需求分析[J]. 振动与冲击, 2017, 36(1): 181-187.
Liu Xiaoxiao，Wu Ziyan，Wang Qiang. Multidimensional performance limit state for probabilistic seismic demand analysis[J]. JOURNAL OF VIBRATION AND SHOCK, 2017, 36(1): 181-187.
[3] 周世军, 吴云丹, 江瑶,等. 高速铁路简支箱梁桥的概率地震需求模型及易损性分析[J]. 土木建筑与环境工程, 2017, 39(6): 12-21.
Zhou Shijun, Wu Yundan, Jiang Yao, and so on. Probabilistic seismic demand model and vulnerability analysis of simple supported box girder bridge of high speed railway [J]. civil construction and environmental engineering, 2017, 39(6): 12-21.
[4] 吕震宙，宋述芳，李洪双等.结构机构可靠性及可靠性灵敏度分析[M]. 科学出版社，北京, 2009.
Lü Zhenzhou, Song Shufang, Li Hongshuang and so on. Structural reliability and reliability sensitivity analysis[M]. Science Press, Beijing, 2009.
[5] Borgonovo E. A new uncertainty importance measure[J]. Reliability Engineering & System Safety, 2007, 92(6):771-784.
[6] HyunSeokSeo, ByungManKwak. Efficient statistical tolerance analysis for general distributions using three-point information[J]. International Journal of Production Research, 2002, 40(4):931-944.
[7] Ratto M, Pagano A, Young P. State Dependent Parameter meta-modelling and sensitivity analysis[J]. Computer Physics Communications, 2007, 177(11):863-876.
[8] Andrea Saltelli. Sensitivity Analysis for Importance Assessment[J]. Risk Analysis, 2002, 22(3):579–590.
[9] 宋帅，钱永久，钱聪. 桥梁地震需求中随机参数的重要性分析方法研究[J]. 工程力学，2018，35(3):106-114.
Song Shuai, Qian Yong-jiu, Qian Cong. Research on methods for importance analysis of random parameters in bridge seismic demand [J]. engineering mechanics, 2018, 35 (3): 106-114.
[10] 尹犟, 胡其高, 李鹏. 混凝土结构地震需求参数敏感性分析[J]. 中南大学学报(自然科学版), 2012, 43(5): 1955-1962.
Yin Jiang, Hu Qigao, Li Peng. Sensitivity analysis of earthquake demand parameters [J]. Journal of Central South University (NATURAL SCIENCE EDITION), 2012, 43 (5): 1955-1962.
[11] 董现，王湛. 基于参数相关性和混合神经网络的结构随机灵敏度分析方法[J]. 建筑结构学报, 2015, 36(4): 149-157.
DONG Xian, WANG Zhan. Stochastic sensitivity analysis method based on correlation parameters and hybrid neural network. JOURNAL OF BUILDING STRUCTURES, 2015, 36(4): 149-157.
[12] 秦玉华, 宫会丽, 宋楠, 等. 改进随机森林的波长选择用于烟叶近红外稳健校正模型的建立[J]. 烟草科技，2014, 323(6):64-67.
QIN Yuhua, GONG Huili, SONG Nan, et al. Wavelength selection base on modified random forest for establishing robust near infrared calibration model of tobacco[J]. Tobacco Science&Technology, 2014, 323(6):64-67.
[13] 秦玉华, 丁香乾, 宫会丽. 高维特征选择方法在近红外光谱分类中的应用[J]. 红外与激光工程, 2013, 42(5):1355-1359.
QIN Yuhua, DING Xiangqian, GONG Huili. High dimensional feature selection in near infrared spectroscopy classification[J]. Infrared and Laser Engineering, 2013, 42(5):1355-1359.
[14] Efron B, Hastie T, Johnstone L, Tibshirani R. Least Angle Regression[J]. The Annals of Statistics, 2004, 32(2): 407-451.
[15] 陈善雄, 刘小娟, 陈春蓉, 等. 针对Lasso问题的多维权重求解算法[J]. 计算机应用, 2017, 37(6):1674-1679.
Chen Shanxiong, Liu Xiaojuan, Chen Chunrong, et al. Multidimensional weight solving algorithm for Lasso problems [J]. computer applications, 2017, 37 (6): 1674-1679.
[16] Sobol I.M. Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates[J]. Mathematics and Computers in Simulation, 2001, 55(1): 271-280.
[17] Porter K A, Beck J L, Shaikhutdinov R V. Sensitivity of Building Loss Estimates to Major Uncertain Variables[J]. Earthquake Spectra, 2002, 18(4):719-743.
[18] GB50010-2010，钢筋混凝土结构设计规范[S].
GB 50010-2010，Code for design of reinforced concrete frame structure[S].
[19] Mirza S A, Macgregor J G. Variability of mechanical properties of reinforced bars[J]. Journal of Structural Division, 1979, 105(5):921-937.
[20] Mirza S A, Hatzinikolas M, Macgregor J G. Statistical Descriptions of Strength of Concrete[J]. Journal of the Structural Division, 1979, 105(6):1021-1037.