Hybrid Artificial Neural Networks for TBM performance prediction in complex underground conditions

Abstract

Due to complex problems in under ground conditions, Tunnel Boring Machine (TBM) performance prediction is mostly affected by conditional environments of the following: geological formation, rock mass, rock property, and fractured rock. This study has presented a new approach using Hybrid Artificial Neural Networks integrated with fuzzy reasoning evaluation model to predict TBM performance in complex underground conditions. The proposed approach is essential to evaluate the TBM performance in terms of Penetration Rate (PR) and Advance Rate (AR) for the planning and management of tunneling. In addition, the proposed approach aims to predict TBM performance and utilization in complex underground conditions such as rock mass, geology, and lithography in tunnel projects. The proposed approach has tested in experiments using data series from tunnel projects in Japan and Asian countries. To assess the significance of the findings and show added valuable parameters of the proposed approach, the results are compared with conventional statistical methods in terms of penetration and advance rates. In order to evaluate the effectiveness of this approach, experimental results show that the proposed approach performs better than other current methods to deal with complex tunneling conditions.