Abstract: Shield tunneling is the method most commonly used for underground projects. Segment typesetting, which involves determining the optimal assembly point for each segment ring and sequentially assembling them into a complete tunnel, is a critical step in shield tunneling. Currently, this typesetting process relies heavily on the operator’s experience at construction sites, which does not guarantee quality. Furthermore, research focuses mainly on the commonly used 16-point segment typesetting, largely ignoring other segment types. In addition, the reliability of these studies in site applications remains unsatisfactory. To address these issues, we propose an intelligent method for segment typesetting using an artificial neural network (ANN) and transfer learning. Due to insufficient historical data for ANN training, a dataset creation method was devised based on the Monte Carlo method and manual annotation. An ANN model was then developed to typeset 16-point segments, with its hyperparameters optimized through Bayesian optimization. Subsequently, the trained model was adapted to other segment types via transfer learning, using 10-point segments as a case study. Based on the test set established in this study, our proposed method showed superior performance compared with several commonly used machine learning methods and a representative and well-validated segment typesetting method. It was also validated using real data collected from construction sites, achieving an accuracy of 93.75% for 16-point segments and 91.43% for 10-point segments, both of which significantly surpass the results from manual typesetting on-site. The proposed method achieves accurate, rapid, and intelligent segment typesetting, which is adaptable to various segment types.

基于人工神经网络和迁移学习的盾构施工管片智能排版方法

[1]AnSH, ParkUY, KangKI, et al., 2007. Application of support vector machines in assessing conceptual cost estimates. Journal of Computing in Civil Engineering, 21(4):259-264.

[2]ChenYJ, TongZM, ZhengY, et al., 2020. Transfer learning with deep neural networks for model predictive control of HVAC and natural ventilation in smart buildings. Journal of Cleaner Production, 254:119866.

[3]CuiJ, ZhangDL, ZhuGL, et al., 2025. Spatio-temporal data fusion-based method for working cycles identification of hydraulic excavators. IEEE Transactions on Instrumentation and Measurement, 74:2510814.

[4]GuanXQ, LiangJY, QianYH, et al., 2017. A multi-view OVA model based on decision tree for multi-classification tasks. Knowledge-Based Systems, 138:208-219.

[5]HuZZ, LiB, HuYZ, 2017. Fast sign recognition with weighted hybrid K-nearest neighbors based on holistic features from local feature descriptors. Journal of Computing in Civil Engineering, 31(5):04017034.

[6]LiDY, LiuZD, XiaoP, et al., 2022. Intelligent rockburst prediction model with sample category balance using feedforward neural network and Bayesian optimization. Underground Space, 7(5):833-846.

[7]LiJB, ChenZY, LiX, et al., 2023. Feedback on a shared big dataset for intelligent TBM Part II: application and forward look. Underground Space, 11:26-45.

[8]LinW, SheilB, ZhangP, et al., 2024. Seg2Tunnel: a hierarchical point cloud dataset and benchmarks for segmentation of segmental tunnel linings. Tunnelling and Underground Space Technology, 147:105735.

[9]LiuFY, DingWQ, QiaoYF, et al., 2021. Compressive behavior of hybrid steel-polyvinyl alcohol fiber-reinforced concrete containing fly ash and slag powder: experiments and an artificial neural network model. Journal of Zhejiang University-SCIENCE A, 22(9):721-735.

[10]LiuR, HuJL, ZhangDL, et al., 2022. Genetic algorithm-based intelligent selection method of universal shield segment assembly points. Applied Sciences, 12(14):6926.

[11]LiuR, HuJL, XuZ, et al., 2024. Differential evolution-based universal segment preliminary layout method for shield tunneling. The 30th International Conference on Mechatronics and Machine Vision in Practice (M2VIP), p.1-6.

[12]LiuR, LiuYS, XieZ, et al., 2025. A relative pose measurement method for multi-shield TBM adjacent shields based on multi-sensor integration. IEEE Transactions on Instrumentation and Measurement, 74:2534115.

[13]LópezD, AlaízCM, DorronsoroJR, 2020. Modified grid searches for hyper-parameter optimization. The 15th International Conference on Hybrid Artificial Intelligence Systems, p.221-232.

[14]LuoHB, LiLH, ChenK, 2022. Parametric modeling for detailed typesetting and deviation correction in shield tunneling construction. Automation in Construction, 134:104052.

[15]NaGS, 2024. One-shot heterogeneous transfer learning from calculated crystal structures to experimentally observed materials. Computational Materials Science, 235:112791.

[16]PaldinoGM, de CaroF, de StefaniJ, et al., 2024. Transfer learning-based methodologies for dynamic thermal rating of transmission lines. Electric Power Systems Research, 229:110206.

[17]ShenX, YuanDJ, CaoLQ, et al., 2022. Experimental investigation of the dynamic sealing of shield tail grease under high water pressure. Tunnelling and Underground Space Technology, 121:104343.

[18]SuWL, LiXG, JinDL, et al., 2020. Analysis and prediction of TBM disc cutter wear when tunneling in hard rock strata: a case study of a metro tunnel excavation in Shenzhen, China. Wear, 446-447:203190.

[19]SzczotkaM, WojciechS, 2008. Application of joint coordinates and homogeneous transformations to modeling of vehicle dynamics. Nonlinear Dynamics, 52(4):377-393.

[20]TarekZ, ElsheweyAM, ShohiebSM, et al., 2023. Soil erosion status prediction using a novel random forest model optimized by random search method. Sustainability, 15(9):7114.

[21]VictoriaAH, MaragathamG, 2021. Automatic tuning of hyperparameters using Bayesian optimization. Evolving Systems, 12(1):217-223.

[22]VinodV, KleinekathöferU, ZaspelP, 2024. Optimized multifidelity machine learning for quantum chemistry. Machine Learning: Science and Technology, 5(1):015054.

[23]XieP, ChenK, ZhuYT, et al., 2023. Dynamic parametric modeling of shield tunnel: a WebGL-based framework for assisting shield segment assembly point selection. Tunnelling and Underground Space Technology, 142:105395.

[24]YanC, DuH, KangEZ, et al., 2022. AVEI-BO: an efficient Bayesian optimization using adaptively varied expected improvement. Structural and Multidisciplinary Optimization, 65(6):164.

[25]YeXW, ZhangXL, ChenYB, et al., 2024. Prediction of maximum upward displacement of shield tunnel linings during construction using particle swarm optimization-random forest algorithm. Journal of Zhejiang University-SCIENCE A, 25(1):1-17.

[26]YousefpourA, FoumaniZZ, ShishehborM, et al., 2024. GP+: a Python library for kernel-based learning via Gaussian processes. Advances in Engineering Software, 195:103686.

[27]ZhangWC, ZhangJ, YanJR, et al., 2021. Selection of the key segment position for trapezoidal tapered rings and calculation of the range of jack stroke differences with a predetermined key segment position. Advances in Civil Engineering, 2021:3606389.

[28]ZhangYK, GongGF, YangHY, et al., 2022. Towards autonomous and optimal excavation of shield machine: a deep reinforcement learning-based approach. Journal of Zhejiang University-SCIENCE A, 23(6):458-478.

[29]ZhangYK, GongGF, YangHY, et al., 2024. From tunnel boring machine to tunnel boring robot: perspectives on intelligent shield machine and its smart operation. Journal of Zhejiang University-SCIENCE A, 25(5):357-381.

[30]ZhouY, SuWJ, DingLY, et al., 2017. Predicting safety risks in deep foundation pits in subway infrastructure projects: support vector machine approach. Journal of Computing in Civil Engineering, 31(5):04017052.

[31]ZhouY, WangY, DingLY, et al., 2018. Utilizing IFC for shield segment assembly in underground tunneling. Automation in Construction, 93:178-191.

[32]ZhuX, BaoTF, YuH, et al., 2020. Utilizing building information modeling and visual programming for segment design and composition. Journal of Computing in Civil Engineering, 34(4):04020024.