Abstract：Research purposes: The foundation pit deformation threats the safety of the foundation pit and surrounding buildings and structures．Application of neural networks in prediction of the foundation pit deformation in advance and giving alarm is a new way to control the risk．
Research conclusions: The neural networks can effectively predict the deformation and stability of the deep foundation pit in advance．and the stability based on which a judgement call be made on the stability of the deep excavation．The prediction value is much more closed to the measured value by using time series as an input layer．This shows the time series method can reflect the internal non—linear relation between the deformation of bracing structure of foundation pit and the time．In practice，the transfer of construction processing sequence can induce a sudden change of the measured value of the deformation．However，the big deviation of the predicted value from the mesaured value may result from ineffective training due to the small sample range of neural networks．
郑知斌,刘 勇,刘继尧, 赵智涛. 基于神经网络的基坑变形预测及安全控制研究[J]. 铁道工程学报, 2010, 27(9): 64-68.
ZHENG Zhi—bin，LIU Yong，LIU Ji—Yao，ZHAO Zhi—tao. Research on Prediction of Foundation Pit Deformation and Safety Control Based on Neural Networks. 铁道工程学报, 2010, 27(9): 64-68.
Wang Jinsan. Study on Structure Design and Ground Settlement in Deep Foundation Pit Based on Artificial Neural Networks [J]. Shenyang: Journal of Shenyang University, 2001 (1): 79 -82.
Ma Fenghai, eta.l Application of BP Neural Network Model in Supporting Pattern of Pile Foundation [J ]. The Chinese Journal of Geological Hazard and Control, 2002(13): 97 -99.
Wei Lide, eta.l Application of Genetic Neural Network Method to Forecasting Horizontal Deformation of Support Structure at Foundation Pit [J ]. Journal of Engineering Geology, 2003(3): 297 -301.
Zhou Kaili, Kang Yaohong.Neural Networks Model and Program Design by MATLAB Simulate [M ]. Beijing: Tsinghua University Press, 2005.
[5] C.G. Chua , Anthony T. C. G oh . Estimating wall deflections in deep excavations using Bayesian neural networks [ J ]. Tunnelling and Underground Space Technology, 2005(20): 400 -409.
[6] Hashash, Y.M. A.&A. J.W hittle. Ground Movement Prediction for Deep Excavations in soft clay [ J ]. Journal of Geotechnical Engineering,1996,122(6):474 - 486.
Hua Ruiping, eta.l Application of Neural Network to Forecasting Deformation of Bracing of Deep Excavation Pit [J ]. Journal of PLA University of Science and Technology, 2002(5): 67 -70.
W u Xianguo, eta.l Deformation Predication of Bracing Of Deep Excavation Pit Based on Fuzzy Neural Network [J]. Optmiization of Capital Construction, 2005(5): 83 - 87.
Ge Changfeng, eta.l A Study on Application of Artificial Neural Network in Prediction of Ground Surface Settlement around Deep Foundation Pit [J ]. Journal of Disaster Prevention and Mitigation Engineering, 2008(4): 519 -523.
Zheng Meitian, eta.l Safety Monitoring Technology and Information - based Construction of Building Foundation Pit [ J ]. Engineering Construction, 2007 (5): 41 -46.
Liu Zhaowei, Zhao Yuncheng. Construction Monitoring and Information Feedback Technique of Urban Underground Works [M ]. Beijing: Science and Technology Press, 2006.
2010, Vol. 27 
版权所有 © 2014 《铁道工程学报》编辑部