A BP neural network-based prediction method for seismic response of circular tunnel linings and parameter analysis
YU Haitao1, ZHU Chenyang2
1. Key Laboratory of Geotechnical and Underground Engineering of Ministry of Education, Tongji University, Shanghai 200092, China; 2. Department of Geotechnical Engineering, Tongji University, Shanghai 200092, China
Abstract: A prediction method of seismic response of circular tunnel based on artificial neural network was proposed. Taking the peak value of bedrock ground motion, the buried depth of tunnel, the relative stiffness ratio of stratum to structure and the contact condition between tunnel and stratum as the basic input parameters, the key response indexes of lining stress and deformation including bending moment, axial force, shear force and diameter change rate of circular tunnel under different ground motions were predicted. Through the response acceleration method, 320 sets of numerical calculation models for seismic response of tunnel structures composed of different input parameter values were established. The results were extracted to obtain the data set of this research and used for the establishment and test of the prediction model. The results showed that the mean square error and correlation coefficient of the response prediction value and the reference value of each group of models performed well, which verified the feasibility of the analysis model. Based on the prediction model after data training, the seismic response of tunnel structure under arbitrary ground motion input could be obtained, and then the prediction formula of seismic response of circular tunnel could be fitted. The validity of the method was verified by comparing with the classical analytical solution of seismic response of deep buried circular tunnel under degraded conditions. This method could also conveniently and quickly analyze the sensitivity influence ranking and effect of basic input parameters, which provided a new analysis method for seismic design and analysis of underground structures.
禹海涛, 朱晨阳. 基于BP神经网络的圆形隧道地震响应预测方法及参数分析[J]. 隧道与地下工程灾害防治, 2023, 5(3): 19-26.
YU Haitao, ZHU Chenyang. A BP neural network-based prediction method for seismic response of circular tunnel linings and parameter analysis. Hazard Control in Tunnelling and Underground Engineering, 2023, 5(3): 19-26.
[1] YUKIO T, IKUO T. Seismic soil structure interaction of cross sections of flexible underground structures subjected to soil liquefaction[J]. Soils and Foundations, 2003, 43(2):69-87. [2] 袁勇,禹海涛,燕晓,等. 超长沉管隧道多点振动台试验模拟与分析[J]. 中国公路学报, 2016, 29(12):157-165. YUAN Yong, YU Haitao, YAN Xiao, et al. Multi-point shaking table test simulation and analysis of a super-long immersed tunnel[J]. China Journal of Highway and Transport, 2016, 29(12):157-165. [3] GOMES R C, GOUVEIA F, TORCATO D, et al. Seismic response of shallow circular tunnels in two-layered ground[J]. Soil Dynamics and Earthquake Engineering, 2015, 75:37-43. [4] 黄忠凯,张冬梅. 软土地层浅埋盾构隧道地震易损性分析[J]. 现代隧道技术, 2020, 57(5):51-60. HUANG Zhongkai, ZHANG Dongmei. Seismic vulnerability analysis of the shallow-buried shield tunnel in soft soil stratum[J]. Modern Tunnelling Technology, 2020, 57(5):51-60. [5] WANG J N. Seismic design of tunnels: a simple state-of-the-art design approach[M].New York,USA: Parsons Brinckerhoff Quade and Douglas Inc, 1993. [6] BOBET A. Drained and undrained response of deep tunnels subjected to far-field shear loading[J]. Tunnelling and Underground Space Technology, 2010, 25(1):21-31. [7] SAJEDI S O, LIANG X. A data-driven framework for near real-time and robust damage diagnosis of building structures[J]. Structural Control and Health Monitoring, 2020, 27(3):e2488.1-e2488.22. [8] XUE Y D, LI Y C. A fast detection method via region-based fully convolutional neural networks for shield tunnel lining defects[J]. Computer-Aided Civil and Infrastructure Engineering, 2018, 33(8):638-654. [9] HUANG P F, CHEN Z Y. Deep learning for nonlinear seismic responses prediction of subway station[J]. Engineering Structures, 2021, 244(5):112735. [10] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors[J]. Nature, 1986, 323(6088):533-536. [11] KATAYAMA Ikuo, ADACHI Masanobu, SHIMADA Minoru, et al. A proposal for a practical quasi-dynamic response to seismic intensity method for buried structures[C] //Proceedings of the 40th Annual Lecture of the Civil Society Section 1. Tokyo, Japan: Japan Society of Civil Engineers, 1985, 40:737-738. [12] 禹海涛,张正伟,李攀.地下结构抗震设计的改进等效反应加速度法[J].岩土力学,2020,41(7):2401-2410. YU Haitao, ZHANG Zhengwei, LI Pan. Improved equivalent response acceleration method for seismic design of underground structures[J]. Rock and Soil Mechanics, 2020, 41(7):2401-2410. [13] 中华人民共和国住房和城乡建设部.城市轨道交通结构抗震设计规范[M]. 北京:中国计划出版社,2014. [14] POWER M S, ROSIDI D, KANESHIRO J Y. Seismic vulnerability of tunnels and underground structures revisited[C] //Proceedings of North American Tunneling.Newport Beach, USA:[s.n] , 1998:243-250. [15] WANG Z Z, ZHANG Z. Seismic damage classification and risk assessment of mountain tunnels with a validation for the 2008 Wenchuan Earthquake[J]. Soil Dynamics and Earthquake Engineering, 2013, 45:45-55. [16] CHEN Z Y, SHI C, LI T B, et al. Damage characteristics and influence factors of mountain tunnels under strong earthquakes[J]. Natural Hazards, 2012, 61(2):387-401. [17] WANG J N. Seismic design of tunnels: a simple state-of-the-art design approach[M].New York,USA: Parsons Brinckerh-off Quade and Douglas, Inc., 1993. [18] 中华人民共和国原城乡建筑环境保护部.建筑抗震设计规范:GB50011—2010[S].北京:中国建筑工业出版社, 2009. [19] 同济大学,上海申通轨道交通研究咨询有限公司.地下铁道建筑结构抗震设计规范:DG/TJ08—2064—2009[S].上海:上海市建筑建材业市场管理总站, 2009. [20] 杨喻声,禹海涛,袁勇.地下结构抗震分析中若干土层模拟方法对比研究[J].防灾减灾工程学报,2019,39(2):244-249. YANG Yusheng, YU Haitao, YUAN Yong. Comparative study of several soil layer simulation methods in seismic analysis of underground structures [J]. Journal of Disaster Prevention and Mitigation Engineering, 2019, 39(2):244-249. [21] 陈国兴, 庄海洋, 徐烨. 软弱地基浅埋隧洞对场地设计地震动的影响[J].岩土工程学报,2004,26(6):739-744. CHEN Guoxing, ZHUANG Haiyang, XU Ye. A study on influence of excavated shallow tunnel on design parameters of ground motion in the soft site[J]. Chinese Journal of Geotechnical Engineering, 2004, 26(6):739-744. [22] JALAL F E, XU Y, IQBAL M, et al. Predictive modeling of swell-strength of expansive soils using artificial intelligence approaches: ANN, ANFIS and GEP[J]. Journal of Environmental Management, 2021, 289:112420. [23] HISHAM M, HAMDY G, EL-MAHDY O O. Prediction of temperature variation in FRP-wrapped RC columns exposed to fire using artificial neural networks[J]. Engineering Structures, 2021, 238:112219. [24] 赵弘, 周瑞祥, 林廷圻. 基于Levenberg-Marquardt算法的神经网络监督控制[J]. 西安交通大学学报, 2002, 36(5):523-527. ZHAO Hong, ZHOU Ruixiang, LIN Tingqi. Neural network supervised control based on levenberg-marquardt algorithm[J]. Journal of Xi'an Jiaotong University, 2002, 36(5):523-527. [25] 张峤,邓贵仕. Levenberg-Marquardt神经网络在煤矿作业人员人因可靠性评价中应用研究[J].大连理工大学学报,2015, 55(4):424-430. ZHANG Qiao, DENG Guishi. Investigation on application of Levenberg-Marquardt neural networks to human reliability evaluation of coalmine workers[J]. Journal of Dalian University of Technology, 2015, 55(4):424-430. [26] LI X L, LU J F. Energy-saving feature extraction method for urban buildings with near-zero energy-consuming based on SVR[J]. International Journal of Global Energy Issues, 2020, 42(5/6):375. [27] ARGYROUDIS S A, PITILAKIS K D. Seismic fragility curves of shallow tunnels in alluvial deposits[J]. Soil Dynamics and Earthquake Engineering, 2012, 35:1-12. [28] SOTIRIOS Argyroudis, GRIGORIOS Tsinidis, FILIPPO Gatti, et al. Effects of SSI and lining corrosion on the seismic vulnerability of shallow circular tunnels[J]. Soil Dynamics and Earthquake Engineering, 2017, 98:244-256. [29] MIAH M I, AHMED S, ZENDEHBOUDI S, et al. Machine learning approach to model rock strength: prediction and variable selection with aid of log data[J].Rock Mechanics and Rock Engineering, 2020, 53(10):4691-4715. [30] HUANG Z, ARGYROUDIS S A, PITILAKIS K, et al. Fragility assessment of tunnels in soft soils using artificial neural networks[J].Underground Space, 2022, 7(2):242-253. [31] 闫长斌,汪鹤健,周建,等.基于Bootstrap-SVR-ANN算法的TBM施工速度预测[J]. 岩土工程学报, 2021, 43(6):1078-1087. YAN Changbin, WANG Hejian, ZHOU Jian, et al. Prediction of TBM advance rate based on Bootstrap method and SVR-ANN algorithm[J]. Journal of Geotechnical Engineering, 2021, 43(6):1078-1087. [32] 马莎,肖明,黄志全.PLS-ANN在岩体力学参数优选中的应用[J]. 武汉大学学报(工学版), 2008, 41(3):43-46. MA Sha, XIAO Ming, HUANG Zhiquan. PLS-ANN method and its application to select rock mass mechanical para meters[J]. Journal of Wuhan University(Engineering), 2008, 41(3):43-46. [33] GÖKHAN A, GÜZELLER C O, ESER M T. The effect of the normalization method used in different sample sizes on the success of artificial neural network model[J]. International Journal of Assessment Tools in Education, 2019, 6(2):170-192.
2023, Vol. 5 
