Abstract: During shield tunneling, ground deformation poses significant safety risks. The full optimization strategy ignores interactions between parameters, resulting in suboptimal performance in the pre-control of settlement in earth pressure balance shields. To address this problem, this paper proposes an integrated strategy that combines optimization and inversion, minimizing parameter interaction interference through adaptive adjustment of shield operation parameters. This mechanism performs an optimization search on the key operation parameters for settlement control, while the remaining operation parameters are predicted through inversion. Taking the Changchun Metro Line 6 project as an example, a bidirectional long short-term memory (Bi-LSTM) model enhanced by a multi-head self-attention (MHSA) mechanism is used to predict shield tunneling-induced settlement with spatiotemporal sequence dependency relationships. Particle swarm optimization and a random forest algorithm are used for optimization and inversion operations in the integrated mechanism, respectively. Subsequent ring position tests showed that the integrated mechanism-based adaptive adjustment strategy limited the average fluctuation of uncontrollable parameters to ±13.95% compared to ±34.27% for the full optimization strategy. The actual average settlement was only 3.81 mm compared to 4.72 mm for the full optimization strategy through collaborative parameter adjustment. The application validated the feasibility and applicability of the integrated mechanism, providing important references for the adaptive adjustment of shield parameters and tunnel construction automation.

盾构隧道施工地表沉降预控的土压平衡（EPB）运行参数自适应调整策略：案例研究

[1]AlMehedi MA, AmurA, MetcalfJ, et al., 2023. Predicting the performance of green stormwater infrastructure using multivariate long short-term memory (LSTM) neural network. Journal of Hydrology, 625:130076.

[2]BaoWJ, SuWZ, ZhaoX, et al., 2025. A study on short-term vegetable price prediction based on the CNN-LSTM-Attention model. Discover Food, 5(1):176.

[3]ChenHY, LiXY, FengZB, et al., 2023. Shield attitude prediction based on Bayesian-LGBM machine learning. Information Sciences, 632:105-129.

[4]ChenRP, ZouN, WuHN, et al., 2022. Review of prediction and control for surface settlement caused by shield tunneling based on machine learning. Journal of Huazhong University of Science and Technology (Natural Science Edition), 50(8):56-65 (in Chinese).

[5]ChengJCP, LuQQ, DengYC, 2016. Analytical review and evaluation of civil information modeling. Automation in Construction, 67:31-47.

[6]ChengXS, ZhangSL, ZhouXH, et al., 2022. Parameter analysis of excavation face stability of shield tunnel under high water pressure seepage. Physics and Chemistry of the Earth, Parts A/B/C, 128:103218.

[7]DengXH, WangBL, 2013. Analysis of influence of shield tunneling crossing underneath intercity railway and shield tunneling parameters optimization. Applied Mechanics and Materials, 353-356:1619-1624.

[8]DigeN, DiwekarU, 2018. Efficient sampling algorithm for large-scale optimization under uncertainty problems. Computers & Chemical Engineering, 115:431-454.

[9]DingLY, WangF, LuoHB, et al., 2013. Feedforward analysis for shield-ground system. Journal of Computing in Civil Engineering, 27(3):231-242.

[10]DongM, GuanMZ, WangKH, et al., 2025. Machine learning-based measurement and prediction of ground settlement induced by shield tunneling undercrossing existing tunnels in composite strata. Sensors, 25(5):1600.

[11]GaleraJM, PescadorS, RodríguezÁ, et al., 2007. Prediction of the ground conditions ahead the TBM face in the tunnels of guadarrama (Spain), using geophysical methods and in situ testing. ITA-AITES World Tunnel Congress–Underground Space–the 4th Dimension of Metropolises, p.203-208.

[12]GaoMT, YangWF, LiuZY, et al., 2024. LSTM goaf surface subsidence prediction method combining convolutional neural network and attention mechanism. Bulletin of Surveying and Mapping, (6):53-58 (in Chinese).

[13]GhaboussiJ, RankenRE, 1977. Interaction between two parallel tunnels. International Journal for Numerical and Analytical Methods in Geomechanics, 1(1):75-103.

[14]Hage ChehadeF, ShahrourI, 2008. Numerical analysis of the interaction between twin-tunnels: influence of the relative position and construction procedure. Tunnelling and Underground Space Technology, 23(2):210-214.

[15]HerrenknechtAG, 2024. Pioneering Underground Together. Herrenknecht AG. https://www.herrenknecht.com/en

[16]HiaS, KuswantoH, PrastyoDD, 2023. Robustness of support vector regression and random forest models: a simulation study. In: Wah YB, Berry MW, Mohamed A, et al. (Eds.), Data Science and Emerging Technologies. Springer, Singapore, p.465-479.

[17]HouGY, XuZD, LiL, et al., 2020. Shield tunneling parameter matching model and UI interface. Advances in Civil Engineering, 2020:9562828.

[18]HouYJ, ZhouMZ, ZhangDL, et al., 2022. Analysis of four shield-driven tunnels with complex spatial relations in a clay stratum. Tunnelling and Underground Space Technology, 124:104478.

[19]HuAF, WangC, XieSL, et al., 2025. Machine learning for soil parameter inversion enhanced with Bayesian optimization. Journal of Zhejiang University-SCIENCE A, 26(11):1034-1051.

[20]HuXY, ChengJG, JuJW, 2021. Influence of the cutterhead configuration and operation parameters on the face stability of EPB shield tunnels in dry granular soils. International Journal of Geomechanics, 21(5):04021050.

[21]HuangHW, ChangJQ, ZhangDM, et al., 2022. Machine learning-based automatic control of tunneling posture of shield machine. Journal of Rock Mechanics and Geotechnical Engineering, 14(4):1153-1164.

[22]HuangK, SunYW, ZhouDQ, et al., 2021. Influence of water-rich tunnel by shield tunneling on existing bridge pile foundation in layered soils. Journal of Central South University, 28(8):2574-2588.

[23]KimCY, BaeGJ, HongSW, et al., 2001. Neural network based prediction of ground surface settlements due to tunnelling. Computers and Geotechnics, 28(6-7):517-547.

[24]KongFC, LuDC, MaYD, et al., 2022. Analysis and intelligent prediction for displacement of stratum and tunnel lining by shield tunnel excavation in complex geological conditions: a case study. IEEE Transactions on Intelligent Transportation Systems, 23(11):22206-22216.

[25]LiC, LiJH, ShiZQ, et al., 2022. Prediction of surface settlement induced by large-diameter shield tunneling based on machine-learning algorithms. Geofluids, 2022:4174768.

[26]LiJH, LiPX, GuoD, et al., 2021. Advanced prediction of tunnel boring machine performance based on big data. Geoscience Frontiers, 12(1):331-338.

[27]LiJP, HuangHX, WangG, 2025. A small-sample prediction model for ground surface settlement in shield tunneling based on adjacent-ring graph convolutional networks (GCN-SSPM). Buildings, 15(19):3519.

[28]LiZC, ChenRP, MengFY, et al., 2015. Tunnel boring machine tunneling-induced ground settlements in soft clay and influence of excavation parameters. Journal of Zhejiang University (Engineering Science), 49(7):1268-1275 (in Chinese).

[29]LinCJ, YangXD, GongYJ, et al., 2021. Prediction model of in-chamber earth pressure for EPB shield based on PSO-BP and parameter sensitivity analysis of tunneling parameters. Journal of Basic Science and Engineering, 29(5):1220-1233 (in Chinese).

[30]LiuWL, DingLY, 2020. Global sensitivity analysis of influential parameters for excavation stability of metro tunnel. Automation in Construction, 113:103080.

[31]LuoZJ, LiZ, TanJZ, et al., 2020. Three-dimensional fluid–soil full coupling numerical simulation of ground settlement caused by shield tunnelling. European Journal of Environmental and Civil Engineering, 24(8):1261-1275.

[32]LvJB, LiXL, FuHL, et al., 2020. Influence of shield tunnel construction on ground surface settlement under the condition of upper-soft and lower-hard composite strata. Journal of Vibroengineering, 22(5):1126-1144.

[33]MairRJ, TaylorRN, BracegirdleA, 1993. Subsurface settlement profiles above tunnels in clays. Géotechnique, 43(2):315-320.

[34]MalekiMR, 2018. Rock joint rate (RJR): a new method for performance prediction of tunnel boring machines (TBMs) in hard rocks. Tunnelling and Underground Space Technology, 73:261-286.

[35]MOHURD (Ministry of Housing and Urban-Rural Development of the People’s Republic of China), SAMR (State Administration for Market Regulation of the People’s Republic of China), 2019. Standard for Geotechnical Testing Method, GB/T 50123-2019. National Standards of the People’s Republic of China(in Chinese).

[36]NiuYL, RenTL, ZhouQ, et al., 2023. Analysis of excavation parameters on face stability in small curvature shield tunnels. Sustainability, 15(8):6797.

[37]PourtaghiA, Lotfollahi-YaghinMA, 2012. Wavenet ability assessment in comparison to ANN for predicting the maximum surface settlement caused by tunneling. Tunnelling and Underground Space Technology, 28:257-271.

[38]RashidTA, FattahP, AwlaDK, 2018. Using accuracy measure for improving the training of LSTM with metaheuristic algorithms. Procedia Computer Science, 140:324-333.

[39]RichaT, LebdaouiS, PereiraJM, et al., 2023. A comparative study of ensemble methods for prediction of surface settlement induced by TBM tunneling. Geo-Risk 2023: Innovation in Data and Analysis Methods, p.211-219.

[40]ShenZW, YangHZ, ZhangSJ, 2021. Neural network approximation: three hidden layers are enough. Neural Networks, 141:160-173.

[41]ShiG, QinCJ, TaoJF, et al., 2021. A VMD-EWT-LSTM-based multi-step prediction approach for shield tunneling machine cutterhead torque. Knowledge-Based Systems, 228:107213.

[42]Sobol’IM, 2001. Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates. Mathematics and Computers in Simulation, 55(1-3):271-280.

[43]SongMY, YangMH, YaoGZ, et al., 2024. Artificial intelligence driven tunneling-induced surface settlement prediction. Automation in Construction, 168:105819.

[44]StandingJR, BurlandJB, 2006. Unexpected tunnelling volume losses in the Westminster area, London. Géotechnique, 56(1):11-26.

[45]SunSL, 2022. Shield tunneling parameters matching based on support vector machine and improved particle swarm optimization. Scientific Programming, 2022:6782947.

[46]SuwansawatS, EinsteinHH, 2006. Artificial neural networks for predicting the maximum surface settlement caused by EPB shield tunneling. Tunnelling and Underground Space Technology, 21(2):133-150.

[47]TaylorKE, 2001. Summarizing multiple aspects of model performance in a single diagram. Journal of Geophysical Research: Atmospheres, 106(D7):7183-7192.

[48]TsaiMY, ChoHH, 2021. Differential-evolution-based weights fine tuning mechanism for GRU to predict 5G traffic flow. International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS), p.1-2.

[49]WangL, PanQJ, WangSY, 2024. Data-driven predictions of shield attitudes using Bayesian machine learning. Computers and Geotechnics, 166:106002.

[50]WangLY, QueHB, WuF, 2025. The CNN-LSTM-Attention model for short term prediction of the polar motion. Measurement Science and Technology, 36(1):016323.

[51]WangRH, LiDQ, ChenEJ, et al., 2021. Dynamic prediction of mechanized shield tunneling performance. Automation in Construction, 132:103958.

[52]WeiHB, WeiDS, JiangBY, et al., 2023. Prediction analysis of settlement of existing road under shield tunneling based on IPSO-SVR. Journal of South China University of Technology (Natural Science Edition), 51(6):62-71 (in Chinese).

[53]WuB, XuJ, HuangW, et al., 2019. Prediction of ground settlement on shielded water-rich sand layer considering groundwater level and tunnel depth. IOP Conference Series: Earth and Environmental Science, 358(4):042045.

[54]XuYJ, ChenXS, 2023. Current situation and prospect of comprehensive development of underground space in metro area. Chinese Journal of Underground Space and Engineering, 19(1):1-12 (in Chinese).

[55]YangMH, SongMY, GuoYW, et al., 2025. Prediction of shield tunneling-induced ground settlement using LSTM architecture enhanced by multi-head self-attention mechanism. Tunnelling and Underground Space Technology, 161:106536.

[56]YeXW, ZhangXL, ZhangHQ, et al., 2023. Prediction of lining upward movement during shield tunneling using machine learning algorithms and field monitoring data. Transportation Geotechnics, 41:101002.

[57]ZhangN, ZhangN, ZhengQ, et al., 2022. Real-time prediction of shield moving trajectory during tunnelling using GRU deep neural network. Acta Geotechnica, 17(4):1167-1182.

[58]ZhangP, ChenRP, WuHN, 2019. Real-time analysis and regulation of EPB shield steering using random forest. Automation in Construction, 106:102860.

[59]ZhangP, ChenRP, WuHN, et al., 2020a. Ground settlement induced by tunneling crossing interface of water-bearing mixed ground: a lesson from Changsha, China. Tunnelling and Underground Space Technology, 96:103224.

[60]ZhangP, WuHN, ChenRP, et al., 2020b. A critical evaluation of machine learning and deep learning in shield-ground interaction prediction. Tunnelling and Underground Space Technology, 106:103593.

[61]ZhangP, YinZY, JinYF, et al., 2020c. A novel hybrid surrogate intelligent model for creep index prediction based on particle swarm optimization and random forest. Engineering Geology, 265:105328.

[62]ZhangWG, LiHR, WuCZ, et al., 2021. Soft computing approach for prediction of surface settlement induced by earth pressure balance shield tunneling. Underground Space, 6(4):353-363.

[63]ZhangYQ, CaoWG, ZhouSH, et al., 2021. Prediction of three-dimensional subface and subsurface settlement caused by shield tunnelling based on Peck formula. Journal of Railway Science and Engineering, 18(1):153-161 (in Chinese).

[64]ZhaoJ, YangWC, ZhuF, 2024. A CNN-LSTM-Attention model for near-crash event identification on mountainous roads. Applied Sciences, 14(11):4934.

[65]ZhouXZ, ZhaoC, BianXC, 2023. Prediction of maximum ground surface settlement induced by shield tunneling using XGBoost algorithm with golden-sine seagull optimization. Computers and Geotechnics, 154:105156.

[66]ZhouZ, ZhangJJ, DingHH, et al., 2023. Settlement prediction model of shield tunnel under-crossing existing tunnel based on GA-Bi-LSTM. Chinese Journal of Rock Mechanics and Engineering, 42(1):224-234 (in Chinese).