Search

This review discusses the application scenarios of the machine learning-supported performance prediction and the optimization efficiency of tunnel boring machines (TBMs). The rock mass quality ratings, which are based on the Chinese code for geological survey, were used to provide “labels” suitable for supervised learning. As a result, the generation of machine prediction for rock mass grades reasonably agreed with the ground truth documented in geological maps. In contrast, the main operational parameters, i.e., thrust and torque, can be reasonably predicted based on historical data. Consequently, 18 collapse sections of the Yinsong project have been successfully predicted by several researchers. Preliminary studies on the selection of the optimal penetration rate and cost were conducted. This review also presents a summary of the main achievements in response to the initiatives of the Lotus Pool Contest in China. For the first time, large and well-documented TBM performance data has been shared for joint scientific research. Moreover, the review discusses the technical problems that require further study and the perspectives in the future development of intelligent TBM construction based on big data and machine learning.

This review summarizes the research outcomes and findings documented in 45 journal papers using a shared tunnel boring machine (TBM) dataset for performance prediction and boring efficiency optimization using machine learning methods. The big dataset was collected during the Yinsong water diversion project construction in China, covering the tunnel excavation of a 20 km-section with 199 items of monitoring metrics taken with an interval of one second. The research papers were the result of a call for contributions during a TBM machine learning contest in 2019 and covered a variety of topics related to the intelligent construction of TBM. This review comprises two parts. Part I is concerned with the data processing, feature extraction, and machine learning methods applied by the contributors. The review finds that the data-driven and knowledge-driven approaches in extracting important features applied by various authors are diversified, requiring further studies to achieve commonly accepted criteria. The techniques for cleaning and amending the raw data adopted by the contributors were summarized, indicating some highlights such as the importance of sufficiently high frequency of data acquisition (higher than 1 second), classification and standardization for the data preprocessing process, and the appropriate selections of features in a boring cycle. The review finds that both supervised and unsupervised machine learning methods have been utilized by various researchers. The ensemble and deep learning methods have found wide applications. Part I highlights the important features of the individual methods applied by the contributors, including the structures of the algorithm, selection of hyperparameters, and model validation approaches.

With the development of underground rail transit, the concept of intelligent tunnel construction has been proposed and promoted. High-quality networking during tunnel construction is a prerequisite for this, making it highly urgent to establish networking during tunnel construction. When studying intra-tunnel networking, it is necessary to consider the propagation characteristics of radio waves in the tunnel. In this paper, according to the actual needs of engineering in tunnel construction and the characteristics of tunnel scenes, an improved ray tracing method is proposed, which considers the type and installation position of antennas, transceiver frequency band and power in channel modeling, and proposes a field strength calculation method under different coordinate systems according to the characteristics of straight and curved segments during tunnel construction. In addition, the propagation characteristics of radio waves in dynamic tunnel construction scenarios are quantitatively analyzed. In this paper, by establishing antenna diagram, two-dimensional and three-dimensional models of tunnels, the computer simulation method is applied to compare with the improved algorithm, and the results have good consistency, in addition, the improved algorithm does not require a lot of modeling work in the early stage, and has high applicability and portability. Not only that, this paper also makes actual measurements of the subway under construction in Zhengzhou, China in different scenarios, and verifies the effectiveness of the method.

This paper addresses the significance of preprocessing big data collected during a tunnel boring machine (TBM) excavation before it is used for machine learning on various TBM performance predictions. The research work is based on two water diversion tunneling projects that cover 29.52 km and 17 051 boring cycles. It has been found that the penetration rate calculated from the raw measured penetration distances exhibits more random behavior owing to their percussive and vibratory behavior of the cutterhead. A moving average method to process the negative instantaneous velocities and a noise reduction filter to deal with signals with abnormal frequencies have been recommended. An index called the drilling efficiency index is introduced to assess the relationships between the mechanical parameters in a boring cycle, whose linear regression coefficient R2 is taken for a preliminary investigation of possible problems requiring preprocessing. The research work defines the irrelevant data whose errors are caused by human or mechanical mistakes, and therefore should be cleaned or amended. These irrelevant data can be divided into five categories: (1) premature cycles, (2) sensor defects, (3) mechanical defects, (4) human interruption, and (5) missing files. A program TBM-Processing has been coded for the recognition and classification of these categories. PDF books generated by the program have been uploaded at GitHub to encourage discussions, collaboration, and upgrading of the data processing work with our peers.