A real-time prediction method for tunnel boring machine cutter-head torque using bidirectional long short-term memory networks optimized by multi-algorithm

作     者：Xing Huang Quantai Zhang Quansheng Liu Xuewei Liu Bin Liu Junjie Wang Xin Yin Xing Huang;Quantai Zhang;Quansheng Liu;Xuewei Liu;Bin Liu;Junjie Wang;Xin Yin

作者机构：State Key Laboratory of Geomechanics and Geotechnical EngineeringInstitute of Rock and Soil MechanicsChinese Academy of SciencesWuhan430071China Key Laboratory of Geotechnical and Structural Engineering Safety of Hubei ProvinceSchool of Civil EngineeringWuhan UniversityWuhan430072China The 2nd Engineering Company of China Railway 12th Bureau GroupTaiyuan030032China 

出 版 物：《Journal of Rock Mechanics and Geotechnical Engineering》 (岩石力学与岩土工程学报（英文版）)

年 卷 期：2022年第14卷第3期

页      面：798-812页

核心收录：

学科分类：081406[工学-桥梁与隧道工程] 08[工学] 0814[工学-土木工程] 082301[工学-道路与铁道工程] 0823[工学-交通运输工程] 

基　　金：financially supported by the National Natural Science Foundation of China (Grant Nos. 52074258  41941018  and U21A20153) 

主　　题：Tunnel boring machine(TBM) Real-time cutter-head torque prediction Bidirectional long short-term memory (BLSTM) Bayesian optimization Multi-algorithm fusion optimization Incremental learning 

摘      要：Based on data from the Jilin Water Diversion Tunnels from the Songhua River(China),an improved and real-time prediction method optimized by multi-algorithm for tunnel boring machine(TBM)cutter-head torque is ***,a function excluding invalid and abnormal data is established to distinguish TBM operating state,and a feature selection method based on the SelectKBest algorithm is ***,ten features that are most closely related to the cutter-head torque are selected as input variables,which,in descending order of influence,include the sum of motor torque,cutter-head power,sum of motor power,sum of motor current,advance rate,cutter-head pressure,total thrust force,penetration rate,cutter-head rotational velocity,and field penetration ***,a real-time cutterhead torque prediction model’s structure is developed,based on the bidirectional long short-term memory(BLSTM)network integrating the dropout algorithm to prevent ***,an algorithm to optimize hyperparameters of model based on Bayesian and cross-validation is *** stopping and checkpoint algorithms are integrated to optimize the training ***,a BLSTMbased real-time cutter-head torque prediction model is developed,which fully utilizes the previous time-series tunneling *** mean absolute percentage error(MAPE)of the model in the verification section is 7.3%,implying that the presented model is suitable for real-time cutter-head torque ***,an incremental learning method based on the above base model is introduced to improve the adaptability of the model during the TBM *** of the prediction performance between the base and incremental learning models in the same tunneling section shows that:(1)the MAPE of the predicted results of the BLSTM-based real-time cutter-head torque prediction model remains below 10%,and both the coefficient of determination(R^(2))and correlation coefficient(r)between measured and