As a tunnel boring machine (TBM) manufacturer, the serious challenge of designing the machine is the most crucial part. However, as a TBM operator, the most crucial challenge is selecting a proper TBM based on the ground conditions. The disc cutter is an effective factor involved in the functionality of TBMs. Proper disc cutter design can greatly reduce the concerns of manufacturers and operators. An appropriate design can be attained through a precise evaluation of the disc cutter life (DCL) under varying operating conditions. In this work, the Xtreme Gradient Boosting (XGBoost) method was used to evaluate the TBM‒DCL. 200 datasets, including nine effective parameters were utilized in the model (80% for training and 20% for testing). The performance ability of the XGBoost method was increased through tuning its hyper‒parameters using several meta‒heuristic optimization algorithms. All the hybrid models showed potential ability in the TBM‒DCL prediction. However, the XGBoost‒ particle swarm optimization model was the most robust one. Sensitivity analysis revealed that the cutter rotation speed had the greatest impact on the model's output. This work's significance is that it can address many of the manufacturer and operator concerns about TBM design and use in different ground conditions.
Shtwai Alsubai, Abdullah Alqahtani, Abed Alanazi — Department of Computer Science, College of Computer Engineering and Sciences in Al-Kharj, Prince Sattam bin Abdulaziz University, P.O. Box 151, Al-Kharj 11942, Saudi Arabia
Taoufik Saidani — Center for Scientific Research and Entrepreneurship, Northern Border University, Arar 73213, Saudi Arabia
Arsalan Mahmoodzadeh — Center of Research and Strategic Studies, Lebanese French University, Erbil, Iraq
Mohamed Ghalla — Faculty of Engineering & Digital Technologies, University of Bradford, Bradford BD71DP, UK
PDF Viewer
Preview is limited to the first 3 pages. Sign in to access the full PDF.
