An Artificial Neural Network Approach to Predict the Performance of Tunnel Boring Machines: A Case Study of Water Conveyance Tunnels in Iran
A. Afradi1, A. Ebrahimabadi2, A. R. Ghazikalayeh3
1Department of Mining and Geology, Qaemshahr Branch, Islamic Azad University, Qaemshahr, Iran
2Department of Petroleum, Mining and Materials Engineering, Central Tehran Branch, Islamic Azad University, Tehran, Iran
3Sadr Arian Investment Company, Tehran, Iran
Keywords: TBM performance, penetration rate, Artificial Neural Network, water conveyance tunnels
Abstract
The aim of this paper is to present an approach to predict the performance of tunnel boring machines (TBM) in Iranian water conveyance tunneling projects using an artificial neural network (ANN) approach. With this respect, a database, including field data and machine parameters, was primarily compiled from the excavation of top five Iranian water conveyance tunnels. The database was then analyzed through ANN to yield an optimum predictive model for the rate of penetration. The results show that there is a close equation between actual (measured) data and predicted data with correlation coefficient of 0.94, and the values of coefficient of determination and root mean square error obtained in this research are equal to 0.90 and 1.2, respectively.
|
