Real-time modeling prediction for excavation behavior

Two real-time modeling prediction (RMP) schemes are presented in this paper for analyzing the behavior of deep excavations during construction. The first RMP scheme is developed from the traditional AR(p) model. The second is based on the simplified Elman-style recurrent neural networks. An on-line learning algorithm is introduced to describe the dynamic behavior of deep excavations. As a case study, in-situ measurements of an excavation were recorded and the measured data were used to verify the reliability of the two schemes. They proved to be both effective and convenient for predicting the behavior of deep excavations during construction. It is shown through the case study that the RMP scheme based on the neural network is more accurate than that based on the traditional AR(p) model.

[1]  Alain Thoraval,et al.  The effect of joint constitutive laws on the modelling of an underground excavation and comparison with in situ measurements , 1997 .

[2]  Stefan C. Kremer,et al.  On the computational power of Elman-style recurrent networks , 1995, IEEE Trans. Neural Networks.

[3]  Shuai Li,et al.  Shallow cylindrical tuned liquid damper for vibration control of high‐rise structures , 2002 .

[4]  Qiusheng Li,et al.  The stochastic finite segment in the analysis of the shear-lag effect on box-girders , 2001 .

[5]  H. Imai,et al.  The effects of a shaft excavation experiment on the hydrology of the Tono research field, Japan , 1995 .

[6]  Dikai Liu,et al.  Evaluation of structural dynamic responses by stochastic finite element method , 1999 .

[7]  The quintic finite element and finite strip with generalized degrees of freedom in structural analysis , 2001 .

[8]  John W. Wilson,et al.  Deformation Mechanism of the Underground Excavations at the WIPP Site , 1999 .

[9]  Dikai Liu,et al.  Modelling of structural response and optimization of structural control system using neural network and genetic algorithm , 2000 .

[10]  Dikai Liu,et al.  A modified finite segment method for thin-walled single-cell box girders with shear lag , 2001 .

[11]  Tsuyoshi Ishida,et al.  Strain monitoring of borehole diameter changes in heterogeneous jointed wall rock with chamber excav , 2000 .

[12]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[13]  Dikai Liu,et al.  Using neural networks to model and predict amplitude dependent damping in buildings , 1999 .

[14]  Dikai Liu,et al.  Damping in buildings : its neural network model and AR model , 2000 .

[15]  Mingju Zhang,et al.  Ground movement analysis of soil nailing construction by three-dimensional (3-D) finite element modeling (FEM) , 1999 .

[16]  Qiusheng Li,et al.  Finite Segment Method for Shear Lag Analysis of Cable-Stayed Bridges , 2002 .

[17]  Qiusheng Li,et al.  Analysis of multi-braced earth retaining structures , 2003 .

[18]  Pierre Roussel-Ragot,et al.  Training recurrent neural networks: why and how? An illustration in dynamical process modeling , 1994, IEEE Trans. Neural Networks.

[19]  A new function transform technique for the solution of FEM dynamic equations , 1999 .

[20]  Hojjat Adeli,et al.  Neural Networks in Civil Engineering: 1989–2000 , 2001 .

[21]  Duc Truong Pham,et al.  Training Elman and Jordan networks for system identification using genetic algorithms , 1999, Artif. Intell. Eng..

[22]  Chang-Yu Ou,et al.  Three-Dimensional Finite Element Analysis of Deep Excavations , 1996 .

[23]  Guido Bugmann,et al.  NEURAL NETWORK DESIGN FOR ENGINEERING APPLICATIONS , 2001 .

[24]  Anthony T. C. Goh,et al.  Empirical design in geotechnics using neural networks , 1995 .