Hybrid surrogate modelling for mechanised tunnelling simulations with uncertain data

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

doi:10.1504/ijrs.2015.072717

Cited by 24 publications

(22 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The GPOD and NNMF approaches are then applied to predict the midpoints and radiuses of the whole interval settlement field of step 23, which is depicted by its lower and upper bounds in Figure 2 (right). The relative errors in percentage of prediction performance from the proposed method comparing to the optimisation based reference solution in [6] are 6.2% and 8.9% for the lower and upper bounds, respectively.…”

Section: Application Example In Mechanised Tunnellingmentioning

confidence: 96%

“…The results, which are computed with the presented midpoint-radius representation, are compared in terms of prediction performance and computation time with the reference solution obtained with PSO in [6]. Figure 2 (left) shows the simulation model with dimensions of 48m, 170m and 56m (in x,y,z directions).…”

Section: Application Example In Mechanised Tunnellingmentioning

confidence: 99%

“…Interval analyses can be performed with this deterministic surrogate model together with an optimisation approach, e.g. Particle Swarm Optimisation (PSO), see [6]. However, this requires several hours on a standard notebook or parallel runs on a computing cluster to reduce the computational time, which limits its practical application for real-time predictions at construction sites.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Surrogate Modelling for Real‐time Predictions of Mechanised Tunnelling Processes with Interval Data

Freitag¹,

Cao²,

Meschke³

2016

Proc Appl Math and Mech

Self Cite

View full text Add to dashboard Cite

Finite element simulations are commonly used to investigate the soil-structure interactions of mechanised tunnelling processes, such as to provide predictions on the expected surface settlement field. For real-time predictions during the tunnel construction, the finite element model is substituted by a hybrid surrogate model combining Artificial Neural Network and Proper Orthogonal Decomposition approaches. The surrogate model is employed to predict time-variant interval surface settlement fields for selected scenarios of the tunnelling process steering parameters in real-time considering uncertain geotechnical parameters as intervals. For this purpose, the surrogate model in [1], which is based on a Recurrent Neural Network and Gappy Proper Orthogonal Decomposition technique, is split into surrogate models for midpoint and radius computations of the interval data. The online computation time of the new surrogate modelling approach is only a few seconds, which enables tu apply it for real-time predictions and to support the Tunnel Boring Machine steering.

show abstract

Section: Application Example In Mechanised Tunnellingmentioning

confidence: 96%

Section: Application Example In Mechanised Tunnellingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Surrogate Modelling for Real‐time Predictions of Mechanised Tunnelling Processes with Interval Data

Freitag¹,

Cao²,

Meschke³

2016

Proc Appl Math and Mech

Self Cite

View full text Add to dashboard Cite

show abstract

“…for the prediction of the deformations induced by geotechnical interventions [15] or for the prediction of tunnelling-induced settlements [18,19,20,21,22]. Hybrid surrogate modelling approaches in mechanized tunnelling combining POD and ANNs are presented in [23] and [24].…”

Section: Introductionmentioning

confidence: 99%

A hybrid finite element and surrogate modelling approach for simulation and monitoring supported TBM steering

Ninić

Freitag

Meschke

2017

Tunnelling and Underground Space Technology

Self Cite

View full text Add to dashboard Cite

A note on versions:The version presented here may differ from the published version or from the version of record. If you wish to cite this item you are advised to consult the publisher's version. Please see the repository url above for details on accessing the published version and note that access may require a subscription. AbstractThe paper proposes a novel computational method for real-time simulation and monitoring-based predictions during the construction of machine-driven tunnels to support decisions concerning the steering of tunnel boring machines (TBMs). The proposed technique combines the capacity of a process-oriented 3D simulation model for mechanized tunnelling to accurately describe the complex geological and mechanical interactions of the tunnelling process with the computational efficiency of surrogate (or meta) models based on artificial neural networks. The process-oriented 3D simulation model with updated model parameters based on acquired monitoring data during the advancement process is used in combination with surrogate models to determine optimal tunnel machine-related parameters such that tunnelling-induced settlements are kept below a tolerated level within the forthcoming process steps. The performance of the proposed strategy is applied to the Wehrhahn-line metro project in Düsseldorf, Germany and compared with a recently developed approach for real-time steering of TBMs, in which only surrogate models are used.

show abstract

“…In [3], a hybrid surrogate model based on a combination of Recurrent Neural Networks (RNN) and Proper Orthogonal Decomposition (POD) is proposed for deterministic input-output mapping with high-dimensional outputs. The hybrid surrogate model has been utilized together with Particle Swarm Optimization to perform optimization based interval analyses for computing the time variant interval settlement field with more than 100 surface points.…”

Section: Introductionmentioning

confidence: 99%

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

Self Cite

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

Hybrid surrogate modelling for mechanised tunnelling simulations with uncertain data

Cited by 24 publications

References 30 publications

Surrogate Modelling for Real‐time Predictions of Mechanised Tunnelling Processes with Interval Data

Surrogate Modelling for Real‐time Predictions of Mechanised Tunnelling Processes with Interval Data

A hybrid finite element and surrogate modelling approach for simulation and monitoring supported TBM steering

Real-Time Fuzzy Analysis of Machine Driven Tunneling

Contact Info

Product

Resources

About