Prediction of Geological Parameters during Tunneling by Time Series Analysis on In Situ Data.

Shanglin Liu,Jie Cai,Kaihong Yang,Siyang Zhou,Qian Zhang,Thippa Reddy G

doi:10.1155/2021/3904273

Abstract

A tunnel boring machine (TBM) is a type of heavy load equipment that is widely used in underground tunnel construction. The geological conditions in the tunneling process are decisive factors that directly affect the control of construction equipment. Because TBM tunneling always takes place underground, the acquisition of geological information has become a key issue in this field. This study focused on the internal relationships between the sequential nature of tunnel in situ data and the continuous interaction between equipment and geology and introduced the long short-term memory (LSTM) time series neural network method for processing in situ data. A method for predicting the geological parameters in advance based on TBM real-time state monitoring data is proposed. The proposed method was applied to a tunnel project in China, and the R2 of the prediction results for five geological parameters are all higher than 0.98. The performance of the LSTM was compared with that of an artificial neural network (ANN). The prediction accuracy of the LSTM was significantly higher compared with that of the ANN, and the generalization and robustness of LSTM are also better than those of ANN, which indicates that the proposed LSTM method could extract the sequence properties of the in situ data. The rule of equipment-geology interaction was reflected by increasing the memory structure of the model through the introduction of the “gate” concept, and the accurate prediction of geological parameters during tunneling was realized. Additionally, the influence of time window and distance of prediction on the model is discussed. The proposed method provides a new approach toward obtaining geological information during TBM construction and also provides a certain reference for the effective analysis of the in situ data with sequence properties.

Highlights

With the rapid development of sensing technology, variety of large engineering equipment are using a large number of sensors to monitor hundreds of equipment operating parameters in real time during service. e effective analysis and modeling of in situ data are helpful in realizing the intelligent perception and prediction of the service environment and the equipment’s working state.e tunnel boring machine (TBM) is a heavy-duty equipment with high construction efficiency and safety and is widely used in modern tunnel construction
As a crucial parameter in the long short-term memory (LSTM), the time window was set to 5, which was used as the default value in subsequent analysis, and the reason will be explained in the analysis part later. e data set was divided into the training set and test set by order according to the ratio of 7 : 3, and these data sets were used to train and test the LSTM model established under the above-mentioned parameter settings. e test set data were not considered in the training process and were instead used to individually test the prediction effect
By considering the sequence characteristic of the TBM in situ data, the method based on the LSTM neural network was selected to analyze the time series of various parameters recorded during TBM construction, and the geological parameters in front of the tunneling excavation were subsequently predicted. e proposed method was applied to the Tianjin Metro Line 9, which is an urban subway project constructed using an earth pressure balance (EPB) TBM in China. e results reveal that the real-time geological parameter prediction method based on the LSTM could realize the real-time and accurate prediction of five geological parameters, namely, the bulk density, cohesion, static earth pressure coefficient, internal friction angle, and elastic modulus

Summary

Introduction

With the rapid development of sensing technology, variety of large engineering equipment are using a large number of sensors to monitor hundreds of equipment operating parameters in real time during service. e effective analysis and modeling of in situ data are helpful in realizing the intelligent perception and prediction of the service environment and the equipment’s working state.e tunnel boring machine (TBM) is a heavy-duty equipment with high construction efficiency and safety and is widely used in modern tunnel construction. Various other geological detection methods for TBM construction have been proposed, for example, the methods using seismic waves [1,2,3]; the bore-tunneling electrical ahead monitoring. TBM tunneling equipment generally carries a large number of sensors, and during the tunneling process, hundreds of equipment airborne parameters, such as the advance rate (AR), revolution per minute (RPM), cutterhead torque (T), and total thrust (F), can be monitored in real time. Since TBM tunneling is a process of continuous interaction between the equipment and the surrounding geology, the changes in the geology during tunneling are reflected in the changes of the airborne monitoring parameters [10,11,12,13]

Methods

Results

Conclusion