Multi-objective Optimization Method for Large-diameter Shield Construction Parameters Based on NGBoost-NSGA-Ⅲ: Jiangnan Avenue Renovation and Upgrading Project in Hangzhou Metro Line 6
1 School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
2 CCCC Wuhan Zhixing International Engineering Consulting Co., Ltd., Wuhan 430040, China
3 College of Civil and Transportation Engineering, Shenzhen University, Shenzhen 518060, China
Show less
文章历史+
Received
Published
Revised
2025-07-02
2026-07-18
2026-01-07
Online Date
2026-08-04
PDF (10120K)
摘要
为解决复杂地层条件下大直径盾构施工中掘进效率、能耗与地表沉降的多目标协同优化难题,提出一种基于数据增广的多目标优化框架。该框架创新性融合自然梯度提升(natural gradient boosting,NGBoost)概率预测模型与第三代非支配排序遗传算法(non-dominated sorting genetic algorithm Ⅲ,NSGA-Ⅲ),并引入逼近理想解排序方法(technique for order preference by similarity to an ideal solution,TOPSIS)从Pareto最优解集中筛选最佳盾构操作参数组合。以杭州市某实际盾构工程为案例,针对不同埋深、土体力学特性场景,开展盾构操作参数优化研究。实验结果表明:所提方法对掘进速度、比能及地表沉降的预测精度较高:预测结果与实际结果的均方根误差(root mean square error,RMSE)分别为3.434 8 mm/min、9.166 2 MJ/m3、5.821 0 mm,决定系数(coefficient of determination,R2)分别达到0.775 9、0.844 4、0.902 6;参数优化效果显著,相较优化前,掘进速度提升19.81%,比能降低62.09%,地表沉降降低21.53%。所提出的多目标优化框架可有效平衡大直径盾构施工中的效率、能耗与安全指标,显著提升复杂地层条件下盾构施工的效率与安全水平,并降低施工能耗,为类似工程的参数优化提供可靠技术参考。
Abstract
Large-diameter shield tunneling under complex stratum conditions faces the challenge of multi-objective collaborative optimization of tunneling efficiency, energy consumption, and surface settlement. A multi-objective optimization framework based on data augmentation was proposed. The framework innovatively integrated the natural gradient boosting (NGBoost) probabilistic prediction model, the non-dominated sorting genetic algorithm Ⅲ (NSGA-Ⅲ), and the technique for order preference by similarity to an ideal solution (TOPSIS) method. The TOPSIS method was used to select the optimal combination of shield operating parameters from the Pareto optimal solution set. An actual shield tunneling project in Hangzhou was taken as a case study. Research on the optimization of shield operating parameters was carried out under scenarios with different burial depths and soil mechanical properties. Experimental results demonstrate that high prediction accuracy for tunneling speed, specific energy consumption, and surface settlement is achieved by the proposed method: the root mean square errors (RMSE) between predicted and actual results are 3.434 8 mm/min, 9.166 2 MJ/m3, and 5.821 0 mm, respectively, while the coefficient of determination (R2) reaching 0.775 9, 0.844 4, and 0.902 6. The parameter optimization effect is significant: compared with the pre-optimization state, the tunneling speed is increased by 19.81%, the specific energy consumption is reduced by 62.09%, and the surface settlement is decreased by 21.53%. In conclusion, the proposed multi-objective optimization framework can effectively balance efficiency, energy consumption, and safety indicators in large-diameter shield tunneling. Under complex stratum conditions, it significantly improves construction efficiency and safety level, and reduces construction energy consumption. This framework provides a reliable technical reference for parameter optimization in similar engineering projects.
CHEN Yang-yang, LIU Wen, WANG Zhi-hua, CHEN Jian, AI De-mi, ZHU Hong-ping, DU Yan-liang.
Multi-objective Optimization Method for Large-diameter Shield Construction Parameters Based on NGBoost-NSGA-Ⅲ: Jiangnan Avenue Renovation and Upgrading Project in Hangzhou Metro Line 6[J].
Science Technology and Engineering, 2026, 26(20): 8782-8790 DOI:10.12404/j.issn.1671-1815.2504935
构建一个集成自然梯度提升(natural gradient boosting,NGBoost)概率预测模型、NSGA-Ⅲ优化算法与逼近理想解排序方法(technique for order pre-ference by similarity to an ideal solution,TOPSIS)的“预测-优化-决策”闭环多目标优化架构,实现了盾构掘进过程中控制参数的动态调整与优化,如图1所示。具体而言,NGBoost模型基于历史与实时更新数据,进行掘进参数与能耗的概率预测,生成具有不确定性度量的初始解集;NSGA-Ⅲ利用非支配排序与参考点机制,在高维目标空间中实现解集的多样性保持与全局优化;TOPSIS方法则在帕累托最优解集上,基于接近理想解原则,辅助选取最优控制参数方案。最终形成集预测精度、优化效率与决策科学性于一体的盾构参数智能调控框架。该框架可在每一环掘进完成后,利用新采集的数据不断更新模型,实时生成下一环的最优控制参数,实现盾构施工的闭环调控与高效运行。
使用决定系数(coefficient of determination,R2)与均方根误差(root mean square errors,RMSE)评估回归模型性能。R2衡量模型对数据变异性的解释程度,值范围为0~1,越接近1表示模型能够更好地解释数据的变动。RMSE则通过计算实际值与预测值之间差异的平方根的均值衡量模型预测误差的大小。RMSE越小,表示预测误差越小,模型越精确。
QianQihu, HuXiaoqiang, LiShuchen, et al. Recent advances in key technologies of shield tunnel engineering in China[J]. Tunnel Construction, 2024, 44(5): 897-926.
[3]
El-HassanH, El-MaaddawyT, Al-SallaminA, et al. Performance evaluation and microstructural characterization of GFRP bars in seawater-contaminated concrete[J]. Construction and Building Materials, 2017, 147: 66-78.
QinYinping, LiZhanguo, LiuYongsheng, et al. Key technologies and development trend of intelligent large-diameter shield tunnel construction in China[J]. Tunnel Construction, 2024, 44(S1): 25-35.
WangYang, CaiQi, WuYuqi, et al. Research and application of intelligent construction platform for super-large diameter shield tunneling[J]. Science Technology and Engineering, 2025, 25(17): 7351-7364.
[10]
LiuW L, ChenY, LiuT X, et al. Shield tunneling efficiency and stability enhancement based on interpretable machine learning and multi-objective optimization[J]. Underground Space, 2025, 22: 320-336.
[11]
ChenY Y, LiuW, AiD M, et al. Real-time model updating for prediction and assessment of under-construction shield tunnel induced ground settlement in complex strata[J]. Journal of Computing in Civil Engineering, 2025, 39(2): 1-25.
RenXudong, ZhangFengkai, DingWantao, et al. Prediction of ground settlement and optimization of tunneling parameters of slurry shield based on neural network[J]. Science Technology and Engineering, 2025, 25(5): 2090-2099.
LiQiang, GanPenglu. On cutter wearing control technology of the shield passing through mixed strata[J]. Modern Tunnelling Technology, 2020, 57(1): 168-174.
ZhangMengxi, FanJie, WuHuiming, et al. Optimization of main controI parameters for ground settIement during slurry balance shieId tunnelling based on AG-PSO and its applicaüon[J]. Modern Tunnelling Technology, 2024, 61(S1): 223-233.
ChenLibo, ZhangMingshu, ChenHaiyong, et al. Analysis and optimal control of shield tunnel construction parameters using categorical boosting-nondominated sorting genetic algorithm-Ⅲ[J]. Tunnel Construction, 2024, 44(8): 1587-1598.
WuXianguo, FengZongbao, LiuJun, et al. Multi-objective optimization of surface settlement safety controlduring shield construction based on RF-NSGA-Ⅱ[J]. China Safety Science Journal, 2022, 32(8): 45-51.
ChenYang, LiuWen, LiuWenli, et al. Multi-objective optimization based on the stability of shield tunnel excavation surface[J]. Science Technology and Engineering, 2024, 24(18): 7882-7888.
WangJinfeng. Prediction and optimization control of deformation of existing tunnel induced by undercrossing of shield tunnel based on particle swarm optimization-random forest-nondominated sorting genetic algorithm Ⅱ[J]. Tunnel Construction, 2023, 43(5): 745-753.
[26]
TonyD, AnandA, DaisyY D, et al. NGBoost: natural gradient boosting for probabilistic prediction[C]// Proceedings of the 37th International Conference on Machine Learning. Vienna: International Machine Learning Society, 2019: 2690-2700.
GongLei, QinYongjun, TianYuan, et al. Optimization and performance study of desert sand concrete mix proportion based on response surface and NSGA-Ⅱ algorithm[J]. Science Technology and Engineering, 2025, 25(13): 5571-5578.
[29]
MaH P, ZhangY J, SunS Y, et al. A comprehensive survey on NSGA-Ⅱ for multi-objective optimization and applications[J]. Artificial Intelligence Review, 2023, 56(12): 15217-15270.
ChenRenpeng, WangZhiteng, WuHuaina, et al. Risk assessment for shield tunneling beneath buildings based on interval improved TOPSIS method and FAHP method[J]. Journal of Shanghai Jiao Tong University, 2022, 56(12): 1710-1719.
ShiWei, WuJian. A multi-attribute evaluation method for low-altitude security system solutions based on DEMATEL and TOPSIS[J]. Science Technology and Engineering, 2025, 25(14): 6109-6117.
KongXiangxun. Analysis of bias-load effect and allocation method of posture-adjustment thrust for EPB shield in soil-rock composite strata[D]. Harbin: Harbin Institute of Technology, 2022.
WangXiang, LiuJun, CaoYuan, et al. Multi-objective optimization of shield attitude based on LightGBM-NSGA-Ⅲ[J]. Railway Standard Design, 2024, 69(7): 130-137.
WuXianguo, LiuJun, CaoYuan, et al. Shield attitude prediction and optimization based on CatBoost-NSGA-Ⅲ algorithm[J]. China Safety Science Journal, 2024, 34(8): 69-77.