Automatization, artificial intelligence and digital strategies in the maintenance of railway TBM tunnels

 

 

 

 


Abstract
The aspects related to the serviceability and long-term of TBM tunnels have assumed increasing relevance in the design, maintenance and management of the infrastructures. In particular, the behaviour of the tunnel structures is closely related to several factors involving the design and construction phase, inspections, regulations and management strategies. To guarantee a high level of knowledge of the tunnel through time and improve durability, it is fundamental to carry out new strategies able to digitalize, automatize and processed the aforementioned factors. The existing methods to perform diagnosis and detection of anomalies are generally costly and time-consuming. ETS introduced a new method for the diagnostic of existing tunnels through an innovative multi-dimensional survey system (ARCHITA), and a new approach for the Management and Identification of the Risk for Existing Tunnels (MIRET). Laser scanning and high-definition photo are obtained with minimal impact on the serviceability of the line for a detailed preliminary assessment of the tunnel condition. The results, in terms of defects analysis on the structures, are digitalized and manipulated in different IT environments. The aforementioned industrial and digital approaches allow the development of deep learning-based methods for the segmentation of defects with Artificial Intelligence algorithms. The paper explains the framework of digital strategies, automatization and artificial intelligence for TBM tunnelling, from the catalogue of defects to the management of the data, focusing on these elements can help the strategic management of a case study.

Read the whole document