Search

Copyright of Journal of Zhejiang University: Science A is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

During the construction process of tunnels, the cutterhead of shield tunneling machines may get clogged due to clay adhesion, which may seriously affect the efficiency of the project. Therefore, finding an intelligent diagnosis method to detect the clogging status is of great importance. In this study, a deep residual network-based method for diagnosing cutterhead clogging on shield tunneling machines is proposed. First, working state data of the shield tunneling machine is screened out, and parameters reflecting the clogging state are selected for further analysis. After eliminating extreme outliers, an empirical formula is proposed to label the data. At the same time, several time-domain features of the selected excavation parameters within every five minutes are extracted. These features are then fed into the proposed model as the input data to realize clogging detection. Because the original dataset is unbalanced, the combination of f1-score and accuracy is used to evaluate the performance of the proposed model. The results show that the accuracy of the proposed algorithm reaches 95.71%, which is 1.21%, 2.84%, 9.84%, 6.04%, and 0.86% higher than the support vector machine-based, random forest-based, AdaBoost-based, extreme gradient boosting-based and deep neural network-based methods. The f1 score of the proposed model is 0.923, which is also 0.038, 0.042, 0.269, 0.169 and 0.02 higher than those compared methods. Therefore, the proposed deep residual network-based method can accurately detect cutterhead clogging conditions. [ABSTRACT FROM AUTHOR]

During tunneling process, it is of critical importance to dynamically adjust operation parameters of shield machine due to changes of geological conditions. Cutterhead torque is one of the key load parameters, and its accurate prediction could adjust operational parameters including cutterhead rotational speed and tunneling speed in advance and avoid potential cutterhead jamming. Based on operation and state data collected by the monitoring system, we propose a residual-convolutional-LSTM neural network (RCLSTMNet) for forecasting cutter head torque in shield machine. On the basis of correlation analysis, parameters closely related to cutter head torque are selected as inputs by employing cosine similarity, which significantly reduces input dimension. Convolutional-LSTM neural network is fused and constructed for extracting deep useful features, while residual network module is utilized to avoid gradient disappearing and improve regression performance. Comparisons with recent data-driven cutterhead torque prediction methods are made on the actual engineering datasets, which demonstrate the presented RCLSTMNet outperforms the other data driven models in most cases. Moreover, the predicted curves of cutterhead torque using the proposed RCLSTMNet coincide with the actual curves much better than predicted curves using the other models. Meanwhile, the highest and average accuracy of RCLSTMNnet reach 98.1% and 95.6%, respectively. [ABSTRACT FROM AUTHOR]

Aiming at the lack of accuracy and effectiveness of the current shield machine speed prediction method, the study proposes to improve the residual network and combine this improved algorithm with the surrounding rock category prediction model to construct the underground shield machine digging speed prediction model. With an average accuracy of 87.4%, an F1 value of 0.86, and an accuracy of 0.84, the study’s prediction model of surrounding rock categories was determined to be valid and superior to the other compared models. The effectiveness of the improved residual algorithm constructed by the study was verified, and it was found to have a better fit to the actual values, with a maximum deviation error value of 4.6 mm/min and a root mean square error of 1.835, which was lower than the other comparative algorithms. The empirical analysis of the underground shield machine digging speed prediction model constructed by the study revealed that the area under the line of the work characteristic curve of the subjects was 0.74, and the F1 value was 0.35, and the accuracy was as high as 84.6%, which was significantly better than that of other comparative models. The shield machine digging speed prediction model, which is based on an enhanced residual network built in the study, performs better than other comparison models, according to the results, which can serve as a theoretical guide for the digital management of coal mine output.

Shield machine may deviate from its design axis during excavation due to the uncertainty of geological environment and the complexity of operation. This study therefore introduced a framework to predict the attitude and position of shield machine by combining long short-term memory (LSTM) model with attention mechanism. The data obtained from the Wuhan Rail Transit Line 6 project were utilized to verify the feasibility of the proposed method. By adding the attention mechanism into the LSTM model, the proposed model can focus more on parameters with higher weights. Sensitivity analysis based on Pearson correlation coefficient was conducted to improve the prediction efficiency and reduce the irrelevant input parameters. Compared with LSTM model, LSTM-attention model has higher accuracy. The mean value of coefficient of determination (R2) increases from 0.625 to 0.736, and the mean value of root mean square error (RMSE) decreases from 3.31 to 2.24. The proposed LSTM-attention model can provide an effective prediction for attitude and position of shield machine in practical tunneling engineering.

Through the active control of shield thrust system oil pressures, a synchronous shield tunnelling technology combining advancement and segment fabrication was proposed. The key to this technology was to completely exploit the additional stroke of the hydraulic jacks generated by the axial insertion of a key block to assemble the segments. Taking the tunnelling project of the Shanghai Airport Railway Link Line as a demonstration project, this paper uses a large model test platform for this synchronous technology. A single-ring construction process along a straight line was simulated, in which the theoretical external loads were implemented on the shield machine. The feasibility and reliability of this synchronous technology were evaluated considering the accuracy of total thrust force control, maintenance of tunnelling speed and shield postures and segment compression. The effectiveness of the redistribution principle for the missing thrust force due to the withdrawal of cylinders located in the segment fabrication region was discussed. Furthermore, some other interesting phenomena were observed in the practical application in addition to the model test. This technology primarily realizes synchronous assembly by improving the control of the propulsion system of conventional shield machines without any transformation of the shield machine’s primary structure, segment patterns or excavation method. The proposed method has good adaptability and a low construction cost, which will be beneficial for future long-distance tunnelling projects.

This study aims to develop a rational theoretical model for cutterhead-soil interaction. The cutterhead-soil interaction mechanism is divided into two components: the cutting action of the cutter on the soil and the extrusion of the cutterhead on the soil. By enhancing the Mckyes–Ali model, we analyze and deduce the force state of the cutter during shield tunneling, obtaining a calculation method for determining the force on the cutter. Additionally, we conduct an in-depth analysis of the extrusion effect of the cutterhead on the soil during shield tunneling, utilizing the fundamental solution of the Kelvin problem. Based on these theoretical calculations, we validate the tunneling thrust and cutterhead torque of the shield using our self-developed multi-functional large-scale shield tunneling test platform. The test results demonstrate that the tunneling thrust and cutterhead torque derived from the established cutterhead-soil interaction model in this paper are relatively close to the experimental monitoring values. This provides a theoretical foundation for establishing reasonable shield tunneling loads.

Abstract Shield machines are currently the main tool for underground tunnel construction. Due to the complexity and variability of the underground construction environment, it is necessary to accurately identify the ground in real-time during the tunnel construction process to match and adjust the tunnel parameters according to the geological conditions to ensure construction safety. Compared with the traditional method of stratum identification based on staged drilling sampling, the real-time stratum identification method based on construction data has the advantages of low cost and high precision. Due to the huge amount of sensor data of the ultra-large diameter mud-water balance shield machine, in order to balance the identification time and recognition accuracy of the formation, it is necessary to screen the multivariate data features collected by hundreds of sensors. In response to this problem, this paper proposes a voting-based feature extraction method (VFS), which integrates multiple feature extraction algorithms FSM, and the frequency of each feature in all feature extraction algorithms is the basis for voting. At the same time, in order to verify the wide applicability of the method, several commonly used classification models are used to train and test the obtained effective feature data, and the model accuracy and recognition time are used as evaluation indicators, and the classification with the best combination with VFS is obtained. The experimental results of shield machine data of 6 different geological structures show that the average accuracy of 13 features obtained by VFS combined with different classification algorithms is 91%; among them, the random forest model takes less time and has the highest recognition accuracy, reaching 93%, showing best compatibility with VFS. Therefore, the VFS algorithm proposed in this paper has high reliability and wide applicability for stratum identification in the process of tunnel construction, and can be matched with a variety of classifier algorithms. By combining 13 features selected from shield machine data features with random forest, the identification of the construction stratum environment of shield tunnels can be well realized, and further theoretical guidance for underground engineering construction can be provided.

Shield machines are currently the main tool for underground tunnel construction. Due to the complexity and variability of the underground construction environment, it is necessary to accurately identify the ground in real-time during the tunnel construction process to match and adjust the tunnel parameters according to the geological conditions to ensure construction safety. Compared with the traditional method of stratum identification based on staged drilling sampling, the real-time stratum identification method based on construction data has the advantages of low cost and high precision. Due to the huge amount of sensor data of the ultra-large diameter mud-water balance shield machine, in order to balance the identification time and recognition accuracy of the formation, it is necessary to screen the multivariate data features collected by hundreds of sensors. In response to this problem, this paper proposes a voting-based feature extraction method (VFS), which integrates multiple feature extraction algorithms FSM, and the frequency of each feature in all feature extraction algorithms is the basis for voting. At the same time, in order to verify the wide applicability of the method, several commonly used classification models are used to train and test the obtained effective feature data, and the model accuracy and recognition time are used as evaluation indicators, and the classification with the best combination with VFS is obtained. The experimental results of shield machine data of 6 different geological structures show that the average accuracy of 13 features obtained by VFS combined with different classification algorithms is 91%; among them, the random forest model takes less time and has the highest recognition accuracy, reaching 93%, showing best compatibility with VFS. Therefore, the VFS algorithm proposed in this paper has high reliability and wide applicability for stratum identification in the process of tunnel construction, and can be matched with a variety of classifier algorithms. By combining 13 features selected from shield machine data features with random forest, the identification of the construction stratum environment of shield tunnels can be well realized, and further theoretical guidance for underground engineering construction can be provided. [ABSTRACT FROM AUTHOR]

This paper proposes a modified process analysis method that combines with the input–output method for carbon emissions assessment in slurry shield tunnel construction. The method was applied to analyze the carbon emissions generated during the construction procedures of a slurry shield tunnel. The results indicate that the carbon emissions from building materials account for the majority of the total emissions, while those from the shield machine and construction procedure are relatively small. In addition, BP and CNN-LSTM neural network models were established to validate the accuracy of the calculation results with model error of 0.1031. Finally, recommendations for reducing carbon emissions in the construction course of slurry shield tunnels are provided. [ABSTRACT FROM AUTHOR]

Aiming at solving the problems of huge cost, long cycle and high risky during manual cutter changing process for shield machine, a cutter changing robot was designed to adapt to its narrow and complex space, the workspace and motion path of cutter changing robot arm were studied. Firstly, by analyzing the internal space and cutter layout characteristic, the configuration of the robot arm was proposed, and the dimension of each rod was designed using multi-objective optimization method. The kinematic model of the robot arm was established by using the D-H model method to analyze its reachable working space. While the inverse kinematic model of the robot arm was solved by the Newton-Raphson iterative method to complete the trajectory planning of the arm for all cutter locations, where the relationship between the target position and the joint variables was established. The motion stability of the robot was verified by virtual simulation and test. The results show that the working space of the designed robot arm meets the cutter changing requirements that all the cutter position can be reached. Moreover, the power output of the drive is smooth, which can transport the cutter weighing up to 250 kg. [ABSTRACT FROM AUTHOR]

Taking a running tunnel of Shanghai Railway Airport connecting line as a demonstration application project, the synchronous technology combining shield tunneling with segment assembling is proposed based on the active control on the oil pressure of the shield thrust system, which can solve the problem of the long construction period produced by a single shield machine employed in a long-distance shield tunnel project. The principle is to make full use of the extra stroke of propulsion cylinders generated by the axial insertion of the key block to assemble segments, and the theoretical operation time of a single ring can be reduced by 31.6%. Then, a large model test platform for this synchronization technology is established to verify its feasibility and reliability, and the redistribution method of the missing thrust force is introduced during the synchronous process, which is verified by the model test. The test results show that the actuators of the shield machine respond quickly, and the errors of the cylinder pressures and total thrust force of the propulsion system are controlled within ±2%. The attitude deviation of the shield machine is controlled within ±6 mm, and the error range of the driving speed is -2 to 4 mm/min. The segments are safe with the designed overburden thickness of 33 m, and the compressive safety coefficient reaches 1.68.

Achieving multivariate long-distance forecasting of shield machine tunneling parameters remains a challenge due to the huge number of tunneling parameters and the complexity of the variation pattern. To solve this problem, a long-distance forecasting method called Adaptive Variational Mode Decomposition and Multi-Stage Stabilized Transformer-based (AVMD-MST) for multiple tunneling parameters is proposed. It uses adaptive VMD and normalization to stabilize the pre-processed tunneling parameters, and the stabilized transformer is designed to establish relationships between historical and future data. The results on two different projects show that the MAPE of the proposed method decreases on average by 12.31%–36.8% for the 180th step predictions compared to the state-of-the-art algorithms. Therefore, the idea of stabilization-prediction-inverse stabilization can achieve high precision multi-variable long-distance forecasting. In the future, geological information overcasting will be carried out on the basis of the multivariate long-distance forecasting model, which can help the driver to determine the tunneling strategy. • A long-distance forecasting method for multiple tunneling parameters is presented. • Adaptive VMD method is presented to determine optimal number of decomposition modes. • An inverse stabilization attention calculation is proposed to retain information. • AVMD-MST could output all tunneling parameters for a specified step ahead at once. • Forecasting errors of AVMD-MST are much smaller than those of existing methods. [ABSTRACT FROM AUTHOR]

To mitigate the potential hazards of shield tunneling misalignment (STM) caused by tunneling posture deviation, a method for optimizing operational parameters tailored to tunneling posture adjustment is developed. This paper presents a generative adversarial network (GAN) framework that incorporate a conditional generative adversarial network (CGAN) and two distinct discriminators (WGAN and Path GAN) to enhance the performance of the multi operational parameter generator. Based on engineering data, comparative experiments are designed to investigate the impact of the feature extraction methods, discriminators, and training strategies on the generation performance. Research has shown that an optimal generator scheme, comprising independent convolutional neural networks (CNNs), a summation feature fusion strategy, and a shared decoder, achieves remarkable performance with an MAE of 0.009, RMSE of 0.012, and average error scope of 0.073. Applications of the model confirm its ability to provide optimization suggestions for shield tunneling posture adjustments in engineering scenarios. • A new method for adjusting the operational parameters of the shield is proposed. • The method has been validated through the data collected from the railway. • Obtained a multi-parameter generator with high accuracy. • The influence of different factors on the performance of the generator are analyzed. [ABSTRACT FROM AUTHOR]

Shield thrust is a critical operational parameter during shield driving, which is of vital significance for adjusting operational parameters and ensuring efficient and safe propulsion of shield tunneling machine. In this paper, a novel hybrid prediction model (CLM) combining attention mechanism, convolutional neural networks (CNN) and Bi-directional long short-term memory (BiLSTM) network is proposed for shield thrust prediction. Correlation analysis based on Maximal Information Coefficient (MIC) between the thrust and other parameters is first conducted to select optimal parameters and reduce input dimension. An attention mechanism is introduced into CNN to distinguish the importance of different features, with the convolution layer and pooling layer further extracting dimension features of the data. Then, a BiLSTM neural network integrating first attention layer is employed to extract time-varying characteristics of the data, with a second attention layer added to capture important time information. Field data during shield cutting bridge piles are investigated to support and validate the effectiveness and superiority of the proposed method. Results show that the proposed CLM model are general enough to avoid overfitting problems and have good performance at prediction. The predicted value match reasonably well the monitoring data, with coefficient of determination ( $\text{R}^{2}$ ) equaling to 0.85, root mean square error (RMSE) equaling to 0.05, mean absolute error (MAE) equaling to 0.02. The CLM model in this paper can accurately predict the thrust even under complicated construction conditions, which provides reference for similar industrial application.

Copyright of Journal of Shanghai Jiao Tong University (1006-2467) is the property of Journal of Shanghai Jiao Tong University Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Zhejiang University: Science A is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to solve the problems such as the kinematics modeling of the shield machine propulsion system and the difficulty of forward solution， the simplified equivalent model method is adopted to construct a simplified model of the shield machine propulsion system， and the parallel shield propulsion system of n-SPS is simplified to 4-SPS equivalent parallel propulsion mechanism and construct its kinematics model. According to the established equivalent mechanism kinematics model， a neural network-Newton hybrid algorithm is used to solve the shield tunneling pose， and the solution domain search function of the neural network is used to predict the initial value and substitute the prediction result into the Newton iteration method for calculation and solution. The pros and cons of the single neural network predicted value and the calculated results of the BP neural network-Newton iterative hybrid algorithm are analyzed and compared. The study found that the establishment of the equivalent model of the shield machine propulsion system can greatly simplify the complexity of kinematics modeling， and the use of hybrid algorithms to solve the kinematics positive solution has higher accuracy， which can improve the control accuracy of the shield machine excavation attitude， thereby improving the tunneling tunnel quality.

The use of subway tunnel engineering technology has become more professional and refined with the growth of society and the advancement of science and technology. The initial construction process of a subway tunnel shield is the most critical part of the entire engineering system. Shield launching period construction is the most prone to accidents in the shield construction process, directly related to the smooth through the shield tunnel. The line 1 of Ho Chi Minh (HCM) Metro is the first subway line, the full length of 19.7 km, the underground road length of 2.6 km from km 0 + 615 to km 2 + 360, from Ben Thanh market, and then through the Sai Gon river and 14 station (including 3 underground stations and 11 elevated stations), reach Suoi Tien park and is located in Long Binh area station, underground building blocks including Ben Thanh market station to Opera House station interval, Opera House station, Opera House station to Ba Son station interval. This paper selects Shield launching period of Opera House station to Ba Son shaft interval as an example, analyze the key construction technology, construction control parameters and launching considerations of shield machine.

It is easy to cause safety accidents due to the low precision and poor effect of earth pressure balance (EPB) control during shield machine tunneling process at present. So a data-driven optimization control method for earth pressure balance in sealed cabin of shield machine is proposed. Firstly, the earth pressure prediction model of four pressure monitoring points in the sealed cabin is established by using least squares support vector machine (LSSVM) method, and the penalty coefficient C and kernel parameters σ are optimized by particle swarm optimization (PSO). Then, an optimization function is established with the minimum sum of multi-point earth pressure prediction errors as the target. The optimal solution is obtained by using the fruit fly optimization algorithm (FOA) to solve the optimal screw conveyor speed, so as to realize the balance control of the earth pressure in sealed cabin. Finally, the simulation experiments are carried out based on field construction data. The results show that the method has great performance such as higher accuracy of calculation and better control effect, which can control the excavation face of shield machine more steady. [ABSTRACT FROM AUTHOR]

Shield machine is a complex large-scale tunneling equipment with multiple systems and driving sources. In order to improve the accuracy of fault diagnosis for shield machine, a method based on the combination of reverse feature elimination (RFE) and extreme learning machine (ELM) is proposed. For the characteristics of shield machine operation data with many dimensions and large quantity, the RFE method is introduced to reduce the dimension of the data, eliminate the redundant dimension and remove the correlation between features. To improve the accuracy and efficiency of fault diagnosis, the ELM neural network classifier model is built based on the extremely learning mechanism for fault diagnosis of shield machine. The simulation results based on the field construction data show that this method improves the accuracy of fault diagnosis of shield machine significantly and has good engineering application value. [ABSTRACT FROM AUTHOR]

The geological condition is complicated and changeable, and difficult to predict accurately, so the shield tunneling construction will face great risks, even lead to serious safety accidents. Real-time and accurate geological identification and prediction is an important guarantee for the safe construction of shield machine, which is also an important problem of shield technology. Based on the real-time tunneling control parameters of shield machine construction, an unsupervised data tree K-Means cluster method is proposed in this paper. Firstly, the correlation between shield tunneling parameters and geological types is analyzed to determine the main tunneling parameters which are greatly affected by geological changes. Secondly, the unsupervised data tree is used to optimize the K value of K-Means cluster algorithm, and then a geological recognition cluster algorithm based on main tunneling parameters is constructed. Finally, a simulation experiment is carried out based on the field construction data, and the accuracy of geological identification is 100%. The results show that the method can accurately identify geological categories and provide decision support for effective parameter regulation of shield machine. [ABSTRACT FROM AUTHOR]

Cutter-changing robot is a new product to automatically replace the cutter in shield machine, whose performance is affected by multiple criteria with diversity, multiplicity and correlation, which is a complex multiple criteria decision-making problem with limited and fuzzy available information. This paper proposes a comprehensive performance evaluation model including the performance indicator hierarchy, weight determination, measurement and assessment methods of indicators and sensitivity analysis. Fuzzy Analytical Hierarchy Process (FAHP) is applied to identify the construction requirements from external shield machine and environment and design characteristics in the whole cutter-changing process, weight for 3 first-level indicators and 14 s-level indicators are determined. The objective evaluation indexes measured by instrumentation and subjective evaluation indexes quantified by fuzzy logic are normalized within the range of 0–1. The scores and proportion of assessment grade for each indicator at each level are obtained through Fuzzy Comprehensive Evaluation (FCE). Furthermore, the sensitivity of weight and evaluation data on performance evaluation result are deeply analyzed. The applicability and validity of the proposed approach were demonstrated by a case study of a cutter-changing robot made by China Railway Construction Heavy Industry Co., Ltd. According to the result, operational capacity is the most important criteria in the first level, and cutter grasping ability is the most important indicator in the second level. Path planning ability is the weakest indicators. Cutter grasping ability and time of replacement are the most critical affecting factors. The proposed method not only helps the designers capture the weak indicators accurately and improve the design quality, but also provides engineers with a holistic perspective to the cutter-changing robot before actual application. [ABSTRACT FROM AUTHOR]

Copyright of China Mechanical Engineering is the property of Editorial Board of China Mechanical Engineering and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to study multiple problems concerning quickly tunneling in long distance tunnel, we design and develop a shield machine based on drilling, tunneling, protecting and transporting. By study the shield technique method, working sequences, guiding system composition and working principle, direction fixation, forehand detection and new support and protection of coal mining and tunneling in detail, we analyze the difference regarding the usage effectiveness between traditional tunnel tunneling and shield tunneling. Shield tunneling shows the advantage of quickly tunneling, effectiveness and efficiency, safety and cost saving points of view.

With development of railway transportation, shield tunneling technology has been widely used. As a key component of shield tunneling, the working efficiency and failure form of hob were directly related to the geological conditions and had an important influence on construction schedule and safety. Based on a subway tunneling project, its geology related was analyzed firstly, then the failure types and corresponding causes of the disc cutters were also explained. Finally, a brief discussion was given about the material properties of disc cutters. The results showed that the failure forms were mainly partial grinding, edge curl, fracture of disc cutter ring, cutter ring off, fracture of cutter shaft. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Central South University is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Earth pressure balance (EPB) shield has been widely used in underground construction. The excavation face stability is crucial to avoid the accidents caused by EPB shield tunneling, so that it is very important to propose an effective control method for the earth pressure balance in sealed cabin. Considering the problem that stable automatic control of the earth pressure in shield's sealed cabin is difficult, an optimal control method of the earth pressure is proposed based on action-dependent heuristic dynamic programming (ADHDP), which can realize online autonomous learning and adaptive control in tunneling process. According to Bellman's principle of optimality, the cost function with respect to the sealed cabin's earth pressure is given. In addition, the action network and critic network of ADHDP controller are constructed. The critic network approximates the cost function and feeds error back into the action network. With the goal of minimizing the cost function, the action network utilizes the critic network's error to optimize screw conveyor speed. The simulation results show that the earth pressure controller based on ADHDP can realize the earth pressure balance control, and the control process is steadier. Moreover, ADHDP controller has good dynamic performance and anti-interference ability. • Action-dependent heuristic dynamic programming (ADHDP) is used to control earth pressure balance in sealed cabin. • Based on Bellman's principle of optimality, ADHDP controller for earth pressure balance is designed. • ADHDP controller can realize earth pressure balance control, and the control process is steadier. • The proposed method has shorter adjusting time and strong anti-interference ability. [ABSTRACT FROM AUTHOR]

In order to reveal the nonlinear dynamics behavior of the three stage planetary transmission system of the shield driven cutter head,a pure torsional coupled nonlinear dynamics model is established for the number of planets,the gap between the teeth and the dynamic loads. The relative displacement between the various parts of the meshing point is derived,the system differential equation is established,the coordinate transformation of differential equations,and the non dimensional treatment is carried out. Then based on the dimensionless differential equations of four orders variable step Runge Kutta method. The transmission mechanism of the phase diagram,Poincare diagram are got,by changing the amplitude of excitation and the meshing stiffness,the influence of parameter variation on the nonlinear characteristics of the system is analyzed. The results show that with the increase of the amplitude of excitation,the system enters the two periodic state of the system by the steady state of the single cycle,and then enters the chaotic motion through the multi period motion. With the increase of the meshing stiffness,the system is in a state of multi periodic motion,and the system is stable in a single period. With increasing the meshing stiffness of ks3,reducing the excitation amplitude,the stability of the system can be improved.

Earth pressure in sealed chamber is affected by multisystem and multifield coupling during shield tunneling process, so it is difficult to establish a mechanism earth pressure control model. Therefore, a data-driven modeling method of earth pressure in sealed chamber is proposed, which is based on parallel least squares support vector machine optimized by parallel cooperative particle swarm (parallel cooperative particle swarm optimization-partial least squares support vector machine). The vectors are parallel studied according to different hierarchies firstly, then the initial classifiers are updated by using cross-feedback method to retrain the vectors, and finally the vectors are merged to obtain the support vectors. The parameters of least squares support vector machine are optimized by the parallel cooperative particle swarm optimization, so as to predict quickly for large-scale data. Finally, the simulation experiment is carried out based on in-site measured data, and the results show that the method has high computing efficiency and prediction accuracy. The method has directive significance for engineering application. [ABSTRACT FROM AUTHOR]

In order to reduce the construction risk caused by human operation error and improve the geological adaptive ability of the shield machine, an autonomous intelligent control method is proposed for shield machine within the framework of interaction–judgment–decision based on Deep Deterministic Policy Gradient (DDPG) deep reinforcement learning in this study. Due to the strong nonlinear relationship between the shield machine's tunneling parameters, this research builds a deep reinforcement learning environment using mechanism model of sealed cabin pressure. DDPG agent model of the shield machine is established to replace the shield machine to interact and train with the geological environment. By minimizing the difference between the target pressure setting value and the sealed cabin pressure value, the dynamic balance between the sealed cabin pressure and the pressure on the excavation surface is realized, and the best strategy is obtained. Through real-time interaction with the geological environment, the method in this paper can dynamically adjust the tunneling parameters, accurately control the sealed cabin pressure, and has a strong geological adaptive ability. By realizing the intelligent decision-making of the tunneling parameters, it greatly improves the independent decision-making ability of the shield machine system, reduces the inaccuracy of human operation, and provides an effective guarantee for the efficient and safe operation of the shield machine. This study applies deep reinforcement learning technology to the control field of earth pressure balance shield machine, promotes AI technology, and provides a new idea for the development of AI construction technology in engineering field. [ABSTRACT FROM AUTHOR]

• A novel approach for adjusting the shield's attitude and position during tunneling was proposed. • A hybrid deep learning prediction framework including comprehensive feature evaluation method, EEMD, CA-CNN and GRU-TA is established. • The test is carried out with the field data collected from the railway in Shanghai, China. • The proposed model is significantly superior to other models in prediction accuracy and prediction efficiency. • The contributions of CA and TA in the model are revealed. In constructing long-distance shield tunnels, it is a difficult challenge to maintain the tunneling trajectory consistent with the design tunnel axis. The accurate prediction of the attitude and position during tunneling can reap the advantage of optimizing the tunneling operation parameters in advance leading to the best tunneling trajectory. This study investigates a framework based on a hybrid deep learning model for attitude and position prediction of the shield machine. This prediction framework contains comprehensive feature evaluation method, ensemble empirical mode decomposition (EEMD), convolutional neural network (CNN), and gate recurrent unit (GRU). The introduction of channel attention and temporal attention in CNN and GRU further strengthens the spatial and temporal feature extraction ability of the model. The performance of the prediction framework is verified through a case study with data collected from the Shanghai urban railway tunnel section. Results reveal that the proposed model with dual attention significantly outperforms other models in prediction accuracy and speed. The bias of feature data can be alleviated by introducing channel attention, and using temporal attention can capture long-distance temporal feature data. The model can support shield construction safety by adjusting the operation parameters, and an application example is used to demonstrate the feasibility of the proposed approach. [ABSTRACT FROM AUTHOR]

The dynamics equation of the four gears parallel drive system of shield machine cutter is established.The 3D model of the four gears parallel drive system is established by using the UG software.The dynamic process of the four gears parallel drive system is simulated by using the ADAMS software.Then the influence of meshing damping,meshing stiffness,backlash and distribution of driving gear on the load sharing are also analyzed.The research shows that the larger the difference between meshing stiffness and meshing damping is,the more uneven the load sharing is,when the driving load changes,backlash has a larger effect on the load balance,the different distributions of driving gear have rather a great impact on the stability of driven system.The results can provide theoretical basis for the design and manufacture of the four gears parallel drive system.

The maximum ground surface settlement prediction is a complex problem as the settlement depends on plenty of intrinsic and extrinsic factors. To obtain the approximate range of the settlement, a hybrid prediction dataset including the geological and construction parameters is built using spatial and temporal series according to the sampling methods. The settlement prediction task is transformed into a multi-modal and multi-variate series prediction task. Hence, a spatial-temporal fusion network (STF-Network) is proposed. The spatial-temporal fusion mechanism is firstly designed to establish the spatial-temporal fusion map, which makes spatial and temporal series interact earlier. Then, the 3D residual unit structure is designed to capture the features of temporal series and spatial-temporal fusion map, and two fully-connected layers are established to capture the spatial structural information. Finally, the final output is merged by the three components. The experimental results for STF-Network demonstrate the superiority over state-of-the-art methods. • Establishing a hybrid prediction dataset using spatial and temporal series • Designing a spatial-temporal fusion network (STF-Network) for settlement prediction • Employing STFM make spatial and temporal series interact earlier • Designing 3D-ResUnit structure based on the periodicity of the temporal series data [ABSTRACT FROM AUTHOR]

[Display omitted] • A novel geological information prediction method for shield machine. • Important basis for this model to extract important features. • Correlation of geological information of adjacent working face. • Indicators of EMSACNN are significantly better than those of existing models. Due to the closed working environment of shield machines, the construction personnel cannot observe the construction geological environment, which seriously restricts the safety and efficiency of the tunneling process. In this study, we present an enhanced multi-head self-attention convolution neural network (EMSACNN) with two-stage feature extraction for geological condition prediction of shield machine. Firstly, we select 30 important parameters according to statistical analysis method and the working principle of the shield machine. Then, we delete the non-working sample data, and combine 10 consecutive data as the input of the model. Thereafter, to deeply mine and extract essential and relevant features, we build a novel model combined with the particularity of the geological type recognition task, in which an enhanced multi-head self-attention block is utilized as the first feature extractor to fully extract the correlation of geological information of adjacent working face of tunnel, and two-dimensional CNN (2dCNN) is utilized as the second feature extractor. The performance and superiority of proposed EMSACNN are verified by the actual data collected by the shield machine used in the construction of a double-track tunnel in Guangzhou, China. The results show that EMSACNN achieves at least 96% accuracy on the test sets of the two tunnels, and all the evaluation indicators of EMSACNN are much better than those of classical AI model and the model that use only the second-stage feature extractor. Therefore, the proposed EMSACNN achieves high accuracy and strong generalization for geological information prediction of shield machine, which is of great guiding significance to engineering practice. [ABSTRACT FROM AUTHOR]

In order to avoid safety accidents caused by the unbalance of earth pressure in the chamber during the tunneling process of shield tunneling machine, it is very crucial to make a scientific and accurate dynamic prediction of earth pressure change. Therefore, a hybrid deep learning model is built by using discrete wavelet transform (DWT), one-dimensional convolution neural network (1DCNN) and long-short term memory (LSTM), implementing the scientific prediction of multipoint earth pressure variation trend in this paper. The DWT is introduced to reduce noise in training data, reduce training cost, and improve training accuracy; the 1DCNN is employed to extract features quickly and reduce the training time of LSTM; in order to further improve the prediction performance of the model, the Ranger optimizer is used to optimize the LSTM, which not only improves the prediction accuracy, but also makes the training process of the prediction model more stable. Finally, the validity of the model is verified based on the actual construction data. The results show that the overall and local earth pressure variation trends in the sealed chamber are clearly visible, which can provide decision-making basis for the shield machine to realize automatic, intelligent and safe excavation construction. • The hybrid deep learning is used to construct the prediction of earth pressure. • Range optimizer is used in LSTM to further improve the prediction performance. • The model in this paper has strong practicability and high prediction accuracy. • The prediction method of earth pressure variation trend is proposed. [ABSTRACT FROM AUTHOR]

This paper discusses multi-point earth pressure balance (EPB) optimal control for EPB shield machine with five pressure sensors on its chamber clapboard. The predictive models taking into account advance speed, rotating speed of the cutter head, screw conveyor speed, and the earth pressure measurements are established for five shield chamber pressure points with the adaptive neuro-fuzzy inference system by minimizing the deviation between the corresponding point's predicted pressures and the settings. Then, the ant colony system algorithm is employed to get the optimal screw conveyor speed to control the EPB during the tunnelling process. Simulation results show that the optimal control method gives better performance with small tracking error and fast tracking speed. [ABSTRACT FROM AUTHOR]

To ensure security during the excavation process of an earth pressure balance shield, this paper presents an optimal control method that accounts for the tunnel face's stability. The tunnel face is controlled by an optimal screw conveyor speed derived from the particle swarm optimization algorithm for a designed stable normal vector angle range on the distribution surface of the chamber pressure field. These normal vector angles can be computed online by measuring the changes to the earth pressure in the shield's chamber using a BP neural network model of the chamber pressure field distribution. An experimental example that uses excavation data from an actual EPB shield is given to illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]