基于融合多注意力机制的深度学习的盾构荷载预测方法

doi:10.12068/j.issn.1005-3026.2023.11.016

东北大学学报（自然科学版） ›› 2023, Vol. 44 ›› Issue (11): 1631-1638.DOI: 10.12068/j.issn.1005-3026.2023.11.016

基于融合多注意力机制的深度学习的盾构荷载预测方法

陈城¹，史培新¹，王占生²，贾鹏蛟¹

(1. 苏州大学轨道交通学院，江苏苏州215000； 2. 苏州轨道交通集团有限公司，江苏苏州215004)

发布日期:2023-12-05
通讯作者: 陈城
作者简介:陈城(1993-)，男，江苏盐城人，苏州大学博士研究生；史培新(1975-)，男，江苏溧阳人，苏州大学教授，博士生导师.
基金资助:
国家自然科学基金资助项目(52278405).

Shield Load Prediction Method Based on Deep Learning with Multiattention Mechanism

CHEN Cheng¹， SHI Pei-xin¹， WANG Zhan-sheng²， JIA Peng-jiao¹

1. School of Rail Transportation， Soochow University， Suzhou 215000， China; 2. Suzhou Rail Transit Group Co.， Ltd.， Suzhou 215004， China.

Published:2023-12-05
Contact: SHI Pei-xin
About author:-
Supported by:
-

摘要/Abstract

摘要： 盾构荷载作为盾构的主要性能指标，准确的荷载预测对于保证盾构安全高效工作和周边环境稳定具有重要意义.鉴于传统预测方法精度差的局限性，本研究以数据的高维度特征和时序特征为切入点，提出一种结合卷积神经网络、双向长短期记忆神经网络和注意力机制的混合模型(CNN-BiLSTM-Multiattention，CBM)，对盾构荷载进行精准预测.该模型不仅可以提取数据的高维度特征和时序特征，还能突出高维度特征的重要性和关键时间节点信息.通过实验证明了相较于4种现有的模型，本文所提出的模型在3种评价指标上均优于其他模型，对推力和扭矩预测的准确率达到94.2%和96.2%.

关键词: 深度学习；注意力机制；时序特征；高维度；荷载预测

Abstract: Shield load is the main performance indicator of the shield， accurate load prediction is significant to ensure the safety and efficiency of the shield and the stability of the surrounding environment. Recognizing the limitations of the traditional prediction methods， this paper proposes a hybrid model(CBM)， combining convolutional neural network (CNN)， bi-directional long short-term memory (BiLSTM) and attention mechanism， to predict the shield load accurately based on the high-dimensional feature and time series characteristic of the data. The proposed model not only can extract the high-dimensional features and time series characteristics of the data， but also can highlight the importance of high-dimensional features and important time node information. The experiment results show that compared with the existing models， the proposed model achieves a higher prediction performance， the prediction accuracy of the thrust and torque is 94.2% and 96.2%.

Key words: deep learning; attention mechanism; time series feature; high-dimension; load prediction

中图分类号:

TU472

陈城，史培新，王占生，贾鹏蛟. 基于融合多注意力机制的深度学习的盾构荷载预测方法[J]. 东北大学学报（自然科学版）, 2023, 44(11): 1631-1638.

CHEN Cheng， SHI Pei-xin， WANG Zhan-sheng， JIA Peng-jiao. Shield Load Prediction Method Based on Deep Learning with Multiattention Mechanism[J]. Journal of Northeastern University(Natural Science), 2023, 44(11): 1631-1638.

参考文献

[1]Wang L T，Yang X，Gong G F，et al.Pose and trajectory control of shield tunneling machine in complicated stratum［J］. Automation in Construction，2018，93:192-199.
[2]Huo J，Xu Z，Meng Z，et al.Coupled modeling and dynamic characteristics of TBM cutterhead system under uncertain factors［J］. Mechanical Systems and Signal Processing，2020，140:106664.
[3]Yang B，Chen S M，Sun S S，et al.Vibration suppression of tunnel boring machines using non-resonance approach［J］. Mechanical Systems and Signal Processing，2020，145:106969.
[4]徐冲.砂卵石地层泥水平衡盾构隧道掘进参数控制［J］.铁道建筑，2019，59(2):94-97.(Xu Chong.Control of tunnel driving parameters of slurry balance shield in sandy gravel stratum［J］. Railway Engineering，2019，59(2):94-97.)
[5]李杰，张斌，付柯，等.基于现场掘进数据的复合地层盾构掘进性能预测方法研究［J］.现代隧道技术，2019，56(4):97-104.(Li Jie，Zhang Bin，Fu Ke，et al.Site data-based prediction of shield driving performance in compound strata［J］. Modern Tunnelling Technology，2019，56(4):97-104.)
[6]Farrokh E.A study of various models used in the estimation of advance rates for hard rock TBMs［J］. Tunnelling and Underground Space Technology，2020，97:103219.
[7]Zhao P X，Zhuo R S，Li S G，et al.Analysis of advancing speed effect in gas safety extraction channels and pressure-relief gas extraction［J］. Fuel，2020，265:116825.
[8]Hu X Y，Zhang Z X.Physical model test on the behavior of large slurry shield-driving tunnel in soft clay［J］. Journal of Shanghai Jiaotong University(Science)，2013，18:693-698.
[9]Zheng G，Lu P，Diao Y.Advance speed-based parametric study of greenfield deformation induced by EPBM tunneling in soft ground［J］. Computers and Geotechnics，2015，65:220-232.
[10]王新刚，韩凯忠，王超，等.基于迁移学习的轴承剩余使用寿命预测方法［J］.东北大学学报(自然科学版)，2021，42(5):665-672.(Wang Xin-gang，Han Kai-zhong，Wang Chao，et al.Bearing remaining useful life prediction method based on transfer learning ［J］. Journal of Northeastern University(Natural Science)，2021，42(5):665-672.)
[11]申彦斌，张小丽，夏勇，等.Bi-LSTM神经网络用于轴承剩余使用寿命预测研究［J］.振动工程学报，2021，34(2):411-420.(Shen Yan-bin，Zhang Xiao-li，Xia Yong，et al.Bi-LSTM neural network for remaining useful life prediction of bearings［J］. Journal of Vibration Engineering，2021，34(2):411-420.)
[12]Armaghani D J，Koopialipoor M，Marto A，et al.Application of several optimization techniques for estimating TBM advance rate in granitic rocks［J］. Journal of Rock Mechanics and Geotechnical Engineering，2019，11(4):779-789.
[13]Salimi A，Rostami J I，Moormann C.Application of rock mass classification systems for performance estimation of rock TBMs using regression tree and artificial intelligence algorithms［J］. Tunnelling and Underground Space Technology，2019，92:103046.
[14]张钦礼，刘奇，赵建文.全尾砂絮凝沉降参数预测模型研究［J］.东北大学学报(自然科学版)，2016，37(6):875-879.(Zhang Qin-li，Liu Qi，Zhao Jian-wen.Study on the parameters prediction model of flocculating sedimentation of crude tailings ［J］. Journal of Northeastern University(Natural Science)，2016，37(6):875-879.)
[15]王述红，朱宝强.山岭隧道洞口段地表沉降时序预测研究［J］.岩土工程学报，2021，43(5):813-821.(Wang Shu-hong，Zhu Bao-qiang.Time series prediction for ground settlement in portal section of mountain tunnels［J］.Chinese Journal of Geotechnical Engineering，2021，43(5):813-821.)
[16]Lin S S，Zhang N，Zhou A N，et al.Time-series prediction of shield movement performance during tunneling based on hybrid model［J］. Tunnelling and Underground Space Technology，2022，119:104245.

基于融合多注意力机制的深度学习的盾构荷载预测方法

Shield Load Prediction Method Based on Deep Learning with Multiattention Mechanism

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价