Estimation of tunnelling-induced settlement by modern intelligent methods

Abstract The purpose of this study is to apply modern intelligent methods to predict subway settlement. These methods can also be used for settlement prediction of tunnel future levels. While many parameters affect settlement, some of them are applicable to empirical or analytical equations. With this in mind, the capability of Adaptive Neuro-Fuzzy Inference System (ANFIS) and Gene Expression Programming (GEP) methods are studied for settlement prediction. The intelligent methods have been studied on the basis of data obtained from 53 tunnels all over the world, which have been excavated using the NATM. These parameters were collected from previous research data, which the values of settlement (S) were obtained from numerical modeling (FLAC2D software). The values of S are predicted by using soil strength parameters (E, C and o), depth (Z) and diameter (D) of the tunnel. 40 data sets were utilized for intelligent modeling, while 13 ones were used for evaluation of its function. Two methods of ANFIS (GP and FCM) were used in this article, and two equations (with and without constant) were also rendered by GEP method. Finally, the results of these two methods were compared. The accuracy of the GEP equation with a constant number of R2 equals to 0.9559 was the best. According to the results, both intelligent methods are recommended for the prediction of subway settlement.

[1]  K. Ahangari,et al.  Subsidence estimation utilizing various approaches - A case study: Tehran No. 3 subway line , 2012 .

[2]  Chuen-Tsai Sun,et al.  Neuro-fuzzy modeling and control , 1995, Proc. IEEE.

[3]  Douglas Fisher,et al.  Machine Learning Approaches to Estimating Software Development Effort , 1995, IEEE Trans. Software Eng..

[4]  John R. Koza,et al.  Genetic programming - on the programming of computers by means of natural selection , 1993, Complex adaptive systems.

[5]  S. N. Sivanandam,et al.  Introduction to genetic algorithms , 2007 .

[6]  Hamid Nikraz,et al.  Predicting pile dynamic capacity via application of an evolutionary algorithm , 2014 .

[7]  Amir Hossein Alavi,et al.  Novel Approach to Strength Modeling of Concrete under Triaxial Compression , 2012 .

[8]  Amir Hossein Alavi,et al.  Expression Programming Techniques for Formulation of Structural Engineering Systems , 2013 .

[9]  Håkan Stille,et al.  Ground behaviour and rock engineering tools for underground excavations , 2007 .

[10]  A. Gandomi,et al.  Nonlinear Genetic-Based Models for Prediction of Flow Number of Asphalt Mixtures , 2011 .

[11]  Jingsheng Shi,et al.  MODULAR NEURAL NETWORKS FOR PREDICTING SETTLEMENTS DURING TUNNELING , 1998 .

[12]  S. Suwansawat,et al.  Artificial neural networks for predicting the maximum surface settlement caused by EPB shield tunneling , 2006 .

[13]  Bashar Tarawneh,et al.  Pipe pile setup: Database and prediction model using artificial neural network , 2013 .

[14]  Cândida Ferreira,et al.  Gene Expression Programming: A New Adaptive Algorithm for Solving Problems , 2001, Complex Syst..

[15]  Kaveh Ahangari,et al.  Predicting crest settlement in concrete face rockfill dams using adaptive neuro-fuzzy inference system and gene expression programming intelligent methods , 2013 .

[16]  Ibrahim Ocak,et al.  Control of surface settlements with umbrella arch method in second stage excavations of Istanbul Metro , 2008 .

[17]  A. R. Selby,et al.  Tunnelling in soils – ground movements, and damage to buildings in Workington, UK , 1999 .

[18]  Mohamed A. Shahin,et al.  Load–settlement modeling of axially loaded steel driven piles using CPT-based recurrent neural networks , 2014 .

[19]  O. J. Santos,et al.  Artificial neural networks analysis of São Paulo subway tunnel settlement data , 2008 .

[20]  Bahman Bohloli,et al.  Intelligent approaches for prediction of compressional, shear and Stoneley wave velocities from conventional well log data: A case study from the Sarvak carbonate reservoir in the Abadan Plain (Southwestern Iran) , 2010, Comput. Geosci..

[21]  Hossein Nezamabadi-pour,et al.  Application of the adaptive neuro-fuzzy inference system for prediction of a rock engineering classification system , 2011 .

[22]  Jyh-Shing Roger Jang,et al.  ANFIS: adaptive-network-based fuzzy inference system , 1993, IEEE Trans. Syst. Man Cybern..

[23]  Kang Tai,et al.  Mathematical modelling of burr height of the drilling process using a statistical-based multi-gene genetic programming approach , 2014 .

[24]  James C. Bezdek,et al.  Fuzzy mathematics in pattern classification , 1973 .

[25]  Stamatios V. Kartalopoulos,et al.  Understanding neural networks and fuzzy logic , 1995 .

[26]  A. Bobet,et al.  Predictions of ground deformations in shallow tunnels in clay , 2002 .

[27]  C. Kayadelen,et al.  Soil liquefaction modeling by Genetic Expression Programming and Neuro-Fuzzy , 2011, Expert Syst. Appl..

[28]  Liliana Teodorescu,et al.  High Energy Physics event selection with Gene Expression Programming , 2008, Comput. Phys. Commun..

[29]  K. M. Neaupane,et al.  Prediction of tunneling-induced ground movement with the multi-layer perceptron , 2006 .

[30]  Pan Guorong.  Applying Theory of Artificial Neural Network to Deformation Prediction of Ground Deformation During Tunnel Excavation , 2001 .

[31]  J. Hou,et al.  Prediction of surface settlements induced by shield tunneling: An ANFIS model , 2008 .