Recurrent neural networks and proper orthogonal decomposition with interval data for real-time predictions of mechanised tunnelling processes

Freitag, Steffen; Cao, Ba Trung; Ninić, Jelena; Meschke, Günther

doi:10.1016/j.compstruc.2017.03.020

Cited by 77 publications

(32 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this paper, the hybrid surrogate modeling for interval data [4] has been extended to process fuzzy data by means of a -representation. The time-dependent behaviour of several selected points in future steps is predicted by Recurrent Neural Networks (RNNs), whereas order reduction techniques (Proper Orthogonal Decomposition and Non-Negative Matrix Factorization) are utilised to approximate the complete surface field based on the RNN predictions.…”

Section: Discussionmentioning

confidence: 99%

“…More details about the explanation and implementation of the method is given in [12] and [4]. The GPOD approach is applied to reconstruct missing elements of a given vector S ⇤ .…”

Section: Gpod and Gappy Nnmf For Fuzzy Datamentioning

confidence: 99%

“…To obtained results with an optimization based interval analysis takes several hours, which is not practical for real-time applications in mechanized tunneling. To enable real-time capabilities, an extension of the hybrid surrogate model is presented in [4] to directly map interval input-output data, which are expressed by a midpoint-radius representation. The key idea of the strategy is to also replace the complete time-consuming optimization loop by surrogate models and not only the deterministic FE simulation model.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Real-Time Fuzzy Analysis of Machine Driven Tunneling

Cao¹,

Freitag²,

Meschke³

2017

Proceedings of the 2nd International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECO

View full text Add to dashboard Cite

Abstract. Reliability assessment in mechanized tunneling requires to take into account limited information describing the local geology and the corresponding geotechnical parameters. The geotechnical data are often quite limited and generally not available in the form of precise models and parameter values. In this case, epistemic uncertainty should be considered within the reliability assessment. The concept of fuzzy numbers is applied to predict tunneling induced settlements in real-time. An advanced numerical simulation is utilized as forward model for the settlement prediction. However, to achieve real-time capabilities, surrogate models are required. Deterministic surrogate models can be used together with an ↵-cut optimization approach to compute fuzzy data. In this paper, a surrogate modeling strategy is introduced to directly process fuzzy input-output data. Within this approach, the time-consuming optimization procedure is replaced by a surrogate model to obtain the fuzzy settlement field prediction in realtime. The significant reduction in computation time maintaining similar prediction performance leads to potential applications in steering of mechanized tunneling processes.

show abstract

Section: Discussionmentioning

confidence: 99%

“…More details about the explanation and implementation of the method is given in [12] and [4]. The GPOD approach is applied to reconstruct missing elements of a given vector S ⇤ .…”

Section: Gpod and Gappy Nnmf For Fuzzy Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Fuzzy Analysis of Machine Driven Tunneling

Cao¹,

Freitag²,

Meschke³

2017

Proceedings of the 2nd International Conference on Uncertainty Quantification in Computational Sciences and Engineering (UNCECO

View full text Add to dashboard Cite

show abstract

“…Despite the numerical ly very efficient RNNGPOD surrogate model, this takes several minutes. More advantageous is direct calculation of the entire interval settlement field through the integra tion of the interval analysis in the hybrid RNNGPOD surrogate model [24]. With this procedure, the computing time is reduced from the original 90 min to 4 s, i.e.…”

Section: Simulation-based Surrogate Modelsmentioning

confidence: 99%

Building Information Modelling in mechanised shield tunnelling – A practitioner's outlook to the near future

Stascheit¹,

Ninić

Meschke

et al. 2018

Geomechanics and Tunnelling

View full text Add to dashboard Cite

The contribution takes a glance at the application of BIM technologies in the design and construction phases of shield tunnelling projects. The intention is to show how Building Information Modelling can be translated into actual benefit, not only in the operation phase but also in the design and construction phases of bored tunnels. Emphasising the integrative character of referencing data uniformly in space and time, examples are given of seamless communication between 3D geometrical modelling, efficient numerical simulation and model adaptation based on measured data acquired during the boring process. The article covers the complete range from predesign through structural analysis and detailed design until the actual excavation process including its interactions with the environment. Special emphasis is given to data management, which is the key to transforming a mere 3D visualisation into a Building Information Model. The article therefore presents a concept for database‐supported, web‐based integration of software modules for geometrical modelling in various levels of detail, efficient numerical simulation tools that are based upon this representation, and process controlling that manages all data acquired during the construction process in a spatially and temporally coordinated reference system.

show abstract

“…Mechanised tunneling is an efficient tunneling technology for the construction fnew underground infrastructures, in particular in urban environments [1].…”

Section: Introductionmentioning

confidence: 99%

Evaluation Method for Tunneling Stability of TBM Cutterhead

Dalian¹,

Urumchi²

2019

WJET

View full text Add to dashboard Cite

In the process of tunneling of tunnel boring machine (TBM), different geological conditions often correspond to different working conditions, and the randomness of geological conditions also causes the order of occurrence of each working condition to be different. Under the conversion of different working conditions, this makes the vibration of different types of cutterheads different. How to choose the appropriate type of cutterhead according to different geological conditions is very important for saving engineering cost and increasing cutterhead life. In view of the above situation, this paper proposes a stability evaluation method during the TBM tunneling process to select the appropriate cutterhead type. Firstly, the corresponding relationship between geology and working conditions is established according to different geological conditions, and the input loads corresponding to geological conditions are obtained. Then, it is substituted into the dynamic model of the cutterhead system, the vibration response boundaries of each degree of freedom are obtained by solving. And the average value of the maximum boundary amplitude of each degree of freedom is taken to represent the extreme vibration of the cutterhead under the corresponding working conditions. Finally, by comparing the fluctuation of the ultimate vibration amplitude of each type of cutterhead in the process of working condition conversion, the results are as follows: when the transition between homogeneous strata and composite strata is normal and there is no large turning and deviation correction, the vibration response of the two-part cutterhead is the smallest, and the two-part cutterhead is the best choice. Otherwise, the five-part cutterhead is the best choice, while the stability of the integrated cutterhead is the worst.

show abstract

Recurrent neural networks and proper orthogonal decomposition with interval data for real-time predictions of mechanised tunnelling processes

Cited by 77 publications

References 43 publications

Real-Time Fuzzy Analysis of Machine Driven Tunneling

Real-Time Fuzzy Analysis of Machine Driven Tunneling

Building Information Modelling in mechanised shield tunnelling – A practitioner's outlook to the near future

Evaluation Method for Tunneling Stability of TBM Cutterhead

Contact Info

Product

Resources

About