The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON SYSTEMS
Transactions ID Number: 52-300
Full Name: George Tsekouras
Position: Lecturer
Age: ON
Sex: Male
Address: Terma Hatzikyriakou, Piraeus
Country: GREECE
Tel: 2104581361
Tel prefix:
Fax:
E-mail address: tsekouras_george_j@yahoo.gr
Other E-mails:
Title of the Paper: An Optimized Neural Network for Predicting Settlements During Tunneling Excavation
Authors as they appear in the Paper: George J. Tsekouras ; John Koukoulis ; Nikos E. Mastorakis
Email addresses of all the authors: tsekouras_george_j@yahoo.gr, cv94068@hotmail.com, mastor@wseas.org
Number of paper pages: 17
Abstract: Several empirical and analytical relations exist between different tunnel characteristics and surface and subsurface deformation, while numerical analyses (mainly using finite difference programs) have also been applied with satisfactory results. In the last years, the solution of some soil mechanics problems has been derived using the approach of the applied computational intelligence methods, especially the artificial neural networks (ANN). The objective of this paper is to describe an optimized artificial neural network (ANN) method in order to estimate the settlements of roof, face and walls during tunneling excavation. The ANN method uses as input variables the overload factor, the placement of the temporary lining ring behind the face, the thickness of the shotcrete, the modulus of elasticity for the surrounding rock-mass and the position where the measurement of the settlement takes place referring to the tunnel axis. The respective settlements are being cal!
culated using different ANN¢s. For each ANN an optimization process is conducted regarding the values of crucial parameters such as the number of neurons, the time parameter and the initial value of the learning rate, etc. using the respective values of a pre-chosen evaluation set. The success of each ANN in predicting the respective settlements is measured by the correlation index between the experimental and predicted values for the evaluation set. Finally, the ANN with the closest to 1 correlation index is specified. A sensitivity analysis for different parameters (the input variables, the population of input vectors, etc.) is also presented showing the reliability of the proposed method.
Keywords: Artificial neural network, Tunneling excavation, Face settlement, Roof settlement, Walls settlement, Parameters optimization, Settlement prediction
EXTENSION of the file: .pdf
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 79.107.58.124