Data-driven multi-step robust prediction of TBM attitude using a hybrid deep learning approach

被引:37
作者
Wang, Kunyu [1 ]
Wu, Xianguo [1 ]
Zhang, Limao [1 ]
Song, Xieqing [1 ]
机构
[1] Huazhong Univ Sci & Technol, Sch Civil & Hydraul Engn, 1037 Luoyu Rd, Wuhan 430074, Hubei, Peoples R China
基金
中国国家自然科学基金;
关键词
TBM attitude; Multi-step prediction; C-GRU; Sensitivity analysis; RECURRENT NEURAL-NETWORKS; BEHAVIOR; MACHINE; MODEL;
D O I
10.1016/j.aei.2022.101854
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
A robust multi-step TBM attitude prediction approach named convolutional gated-recurrent-unit neural network (C-GRU) is proposed in this research and the random balance design Fourier amplitude sensitivity test method is used for sensitivity analysis to reveal the interaction between input and output of the C-GRU model. A tunnel construction project in Singapore is taken as an example to prove the robustness and effectiveness of the pro-posed approach. Results indicate that the length of the output sequence of the model can maintain high robustness and accuracy within 21 steps. In the 21-step prediction, the highest R2 can reach 0.9652 while the mean R2 is 0.9004 even though some attitude parameter is with large fluctuations. Each step in the 21-step prediction can maintain a stable accuracy. The data of the past 11 time steps of the TBM attitude parameters are the most sensitive. The proposed method has higher accuracy and robustness than state-of-art time-series based methods.
引用
收藏
页数:18
相关论文
共 47 条
[1]   A comprehensive comparative analysis of machine learning models for predicting heating and cooling loads [J].
Abdelkader, Eslam Mohammed ;
Al-Sakkaf, Abobakr ;
Ahmed, Reem .
DECISION SCIENCE LETTERS, 2020, 9 (03) :409-420
[2]   Bayesian prediction of TBM penetration rate in rock mass [J].
Adoko, Amoussou Coffi ;
Gokceoglu, Candan ;
Yagiz, Saffet .
ENGINEERING GEOLOGY, 2017, 226 :245-256
[3]   A performance comparison of sensitivity analysis methods for building energy models [J].
Anh-Tuan Nguyen ;
Reiter, Sigrid .
BUILDING SIMULATION, 2015, 8 (06) :651-664
[4]   Modelling the Torque with Artificial Neural Networks on a Tunnel Boring Machine [J].
Cachim, Paulo ;
Bezuijen, Adam .
KSCE JOURNAL OF CIVIL ENGINEERING, 2019, 23 (10) :4529-4537
[5]  
Cho KYHY, 2014, Arxiv, DOI arXiv:1406.1078
[6]   Recurrent neural networks and proper orthogonal decomposition with interval data for real-time predictions of mechanised tunnelling processes [J].
Freitag, S. ;
Cao, B. T. ;
Ninic, J. ;
Meschke, G. .
COMPUTERS & STRUCTURES, 2018, 207 :258-273
[7]   Spatio-temporal feature fusion for real-time prediction of TBM operating parameters: A deep learning approach [J].
Fu, Xianlei ;
Zhang, Limao .
AUTOMATION IN CONSTRUCTION, 2021, 132
[8]   A direct random sampling method for the Fourier amplitude sensitivity test of nonuniformly distributed uncertainty inputs and its application in C/C nozzles [J].
Gao, Bo ;
Yang, Qiang ;
Peng, Zujun ;
Xie, Weihua ;
Jin, Hua ;
Meng, Songhe .
AEROSPACE SCIENCE AND TECHNOLOGY, 2020, 100
[9]   A novel MP-LSTM method for ship trajectory prediction based on AIS data [J].
Gao, Da-wei ;
Zhu, Yong-sheng ;
Zhang, Jin-fen ;
He, Yan-kang ;
Yan, Ke ;
Yan, Bo-ran .
OCEAN ENGINEERING, 2021, 228
[10]   Real-Time Dynamic Earth-Pressure Regulation Model for Shield Tunneling by Integrating GRU Deep Learning Method With GA Optimization [J].
Gao, Min-Yu ;
Zhang, Ning ;
Shen, Shui-Long ;
Zhou, Annan .
IEEE ACCESS, 2020, 8 :64310-64323