<<
[an error occurred while processing this directive] | [an error occurred while processing this directive] >>
Study of Standardized Pre-processing Method of TBM Tunnelling Data
(1. State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin, China Institute of Water Resources and Hydropower Research, Beijing 100048; 2. School of Civil Engineering and Transportation, Hohai University, Nanjing 210098; 3. School of Civil Engineering, Beijing Jiaotong University, Beijing 100044)
Abstract During TBM construction process, massive data are collected through information technology, and the stan? dardized pre-processing of TBM data is a precondition for multiple studies. Thus, a standardized pre-processing method of TBM tunnelling data is put forward. Based on the big data generated in TBM tunnelling and the TBM rock-breaking characteristics, the basic tunnelling parameters (e.g. cutterhead rotation speed, advancing speed, cutterhead thrust, and cutterhead torque) are selected to analyze the data characteristics during TBM tunnelling. The judgement methods of starting points of idle stage, ascent stage, stable stage and descent stage during the process of cyclic tunnelling are proposed, and furthermore the standard deviation judgement method, mean value judgement method and histogram judgement method are put forward for the starting point of the stable stage, so as to meet the requirements for segmentation of real-time and non-real-time data. Finally, the standardized pre-processing of TBM data is conducted for two TBM tunnel projects to realize the standardization of big data during construction.The result shows that the proposed standardized pre-processing method can realize the effective segmentation of cyclic tunnelling data. The research results can be applied to the standardized data processing of many TBM tunnel projects to effectively create the database for machine learning.
Abstract：
During TBM construction process, massive data are collected through information technology, and the stan? dardized pre-processing of TBM data is a precondition for multiple studies. Thus, a standardized pre-processing method of TBM tunnelling data is put forward. Based on the big data generated in TBM tunnelling and the TBM rock-breaking characteristics, the basic tunnelling parameters (e.g. cutterhead rotation speed, advancing speed, cutterhead thrust, and cutterhead torque) are selected to analyze the data characteristics during TBM tunnelling. The judgement methods of starting points of idle stage, ascent stage, stable stage and descent stage during the process of cyclic tunnelling are proposed, and furthermore the standard deviation judgement method, mean value judgement method and histogram judgement method are put forward for the starting point of the stable stage, so as to meet the requirements for segmentation of real-time and non-real-time data. Finally, the standardized pre-processing of TBM data is conducted for two TBM tunnel projects to realize the standardization of big data during construction.The result shows that the proposed standardized pre-processing method can realize the effective segmentation of cyclic tunnelling data. The research results can be applied to the standardized data processing of many TBM tunnel projects to effectively create the database for machine learning.
Wang Shuangjing1,
2 Wang Yujie1 Li Xu3 Liu Lipeng1 Yin Tao1,
2
.Study of Standardized Pre-processing Method of TBM Tunnelling Data[J] MODERN TUNNELLING TECHNOLOGY, 2022,V59(2): 38-44
2022, Vol. 59 