Dynamic prediction of mechanized shield tunneling performance

Wang, Ruohan; Li, Dianqing; Chen, Elton J.; Liu, Yong

doi:10.1016/j.autcon.2021.103958

Cited by 51 publications

(17 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recurrent neural network (RNN) (Jin et al, 2018) is a special network structure in neural network for processing time series data, which is widely used in sequence related fields such as language model (Zhao and Dong, 2018), part-of-speech tagging (Si et al, 2018). The long-short term memory (LSTM) network (Vlachas et al, 2018;Liu et al, 2019;Zhang et al, 2020b;Wang et al, 2021) improves the unit structure of the traditional recurrent neural network and effectively solves the long-term dependency problems such as gradient disappearance in the RNN network. However, LSTM network is a one-way sequence structure, which cannot use the subsequent time information to calculate the output, and the data processing process is cumbersome.…”

Section: Introductionmentioning

confidence: 99%

Prediction Method of TBM Tunneling Parameters Based on PSO-Bi-LSTM Model

Zhang

Zhu

et al. 2022

Front. Earth Sci.

View full text Add to dashboard Cite

With the wide application of full-face rock tunnel boring machine (TBM) in tunnel construction, the self-adaptive adjustment of TBM tunneling parameters is of great significance for the safety and efficiency of TBM tunnelling. Aiming at the shortcomings of the current TBM data mining capability and optimization methods of tunneling parameters, this paper proposes a prediction method of TBM tunneling parameters based on particle swarm optimization-bi-directional long short-term memory (PSO-Bi-LSTM) model, which selects the complete tunneling cycle data to predict the TBM tunneling parameters, and uses a number of numerical methods such as binary state discriminant function and 3σ criterion to preprocess the operation data of TBM3 bid section of Songhua River water conveyance project. By comparing with the Bi-LSTM model and evaluating the prediction effect under different surrounding rock levels, the applicability and prediction performance of the model to different strata are verified. The results show that the prediction accuracy of the model is proportional to the surrounding rock grade. Compared with the Bi-LSTM, the overall prediction effect of the proposed PSO-Bi-LSTM model is better, which can assist the intelligent construction of TBM with similar geological conditions.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction Method of TBM Tunneling Parameters Based on PSO-Bi-LSTM Model

Zhang

Zhu

et al. 2022

Front. Earth Sci.

View full text Add to dashboard Cite

show abstract

“…Input values are measured in the DRM that determines the categorization of the data into branches. Decision trees help to discreet and continue the variables of data sets following the DRM [7]. A subset of data is the significant attribution of the decision trees.…”

Section: Implication Of Data Mining In Decision Trees and Decision Rulesmentioning

confidence: 99%

The Application of Decision Tree Method for Data Mining

G¹,

Ravishankar²

2022

TTIDMKD

View full text Add to dashboard Cite

The study showcases the impact of data mining considering the application of decision trees that helps to develop data from large number of datasets. Linear data set has been generated through the model of decision trees. Secondary data collection method has been selected in this study with inductive approach. Cross sectional research design has been used in this study to derived insight of the subject of the study. Themes are developed using the peer reviewed journals published after 2019 and secondary collected data has been interpreted in this study to meet the goal of the subject. Implications of data mining including decision trees and decision rules in different field have been discussed over here with positive perspective approach. Customization of data mining in creativity also has been focused here. Different types of decision trees methods have been depicted here with comparison vision. Advantages and disadvantages of decision trees also have been discussed over here to evaluate the actual impact of application of decision trees in data mining. Interpretation of results also has been depicted here to analyse the consequences of decision trees in data mining to conclude the study with informative justification.

show abstract

“…Regarding load prediction, several models [14][15][16] combining geological, operating, and structural parameters have been proposed to estimate the thrust and torque. In addition, a multi-step prediction method, 17 dynamic regulation model, 18 and hybrid deep neural network 19 have also been used for cutterhead torque prediction. Due to the considerable separation between the vertical sliding surface of the soil and the actual curved sliding surface, the traditional theory of arching in soils 20 underestimates the vertical load.…”

Section: Introductionmentioning

confidence: 99%

Load characterization of the main bearing of a large tunnel boring machine based on dynamic characteristic parameters

Sun

Liu

Cheng

et al. 2023

Advances in Mechanical Engineering

View full text Add to dashboard Cite

This research was carried out to solve the problem of the reasonable characterization of the working load of the main bearing of a large tunnel boring machine (TBM) under complex engineering geological conditions and equipment working statuses. A typical telescopic swing main drive system is considered, and a characterization approach based on acquired dynamic characteristic parameters is proposed. First, the axial load Fa, radial load Fr, overturning moment Mk, and torque T are considered as the load indexes of the main bearing. The main drive load model is then developed, and the load indexes are expressed by exploring the relationships between each load index and dynamic characteristic parameters such as the pressures, displacements of the hydraulic cylinders, and torques of the driving motors. Finally, the load indexes are characterized based on a subsea tunnel shield project, representative engineering geologies, and characteristic load inputs. The results indicate that by taking into account the variable attitude of the main drive system, each load index can be expressed as functions of the pressures, displacements of the hydraulic cylinders, and torques of the driving motors. According to the variation of the dynamic characteristic parameters, the load condition of the main bearing during application is accurately characterized. Different geologies are found to correspond to different load levels; the load under the dolomitic limestone and filling karst cave strata is found to be almost 1.5–1.9 times greater than that under diabase, while the torque is almost five times greater. The proposed load characterization approach provides an accurate load input conforming to engineering practice for the design and selection of the main bearing.

show abstract

Dynamic prediction of mechanized shield tunneling performance

Cited by 51 publications

References 37 publications

Prediction Method of TBM Tunneling Parameters Based on PSO-Bi-LSTM Model

Prediction Method of TBM Tunneling Parameters Based on PSO-Bi-LSTM Model

The Application of Decision Tree Method for Data Mining

Load characterization of the main bearing of a large tunnel boring machine based on dynamic characteristic parameters

Contact Info

Product

Resources

About