Surrogate modeling for mechanized tunneling simulations with uncertain data
نویسندگان
چکیده
Computational reliability assessment in mechanized tunneling requires realistic numerical models accounting for the uncertainty of geotechnical and tunneling process parameters. For real-time prognoses, surrogate models are in general inevitable to substitute expensive complex numerical simulations. In this paper, different strategies for surrogate modeling in mechanized tunneling simulations are presented. The focus is on surrogate models for reliability analyses with polymorphic uncertain data described by e.g. stochastic numbers, p-boxes, fuzzy stochastic numbers, fuzzy numbers as well as intervals. Beside surrogate models based on Proper Orthogonal Decomposition Method (POD) and Artificial Neural Networks (ANN), a hybrid surrogate model, i.e. a combination of POD and ANN, is presented. Thereby, uncertain time varying surface settlements of several monitoring points are predicted by recurrent neural networks. Based on that, spatially varying surface settlements are evaluated using POD. Utilizing this surrogate model helps to reduce the computation time, when limit states at multiple surface positions are required.
منابع مشابه
Global sensitivity analysis for subsoil parameter estimation in mechanized tunneling
The present paper validates two alternative global sensitivity analysis methods, namely variance-based and elementary effect, for the purpose of detecting key subsoil parameters that influence the output of mechanized tunnel finite element simulation. In the elementary effect method, a strategy for considering the dependencies, that result from a set of constraints between different parameters,...
متن کاملOptimal Shaping of Non-Conventional Permanent Magnet Geometries for Synchronous Motors via Surrogate Modeling and Multi-Objective Optimization Approach
A methodology is proposed for optimal shaping of permanent magnets with non-conventional and complex geometries, used in synchronous motors. The algorithm includes artificial neural network-based surrogate model and multi-objective search based optimization method that will lead to Pareto front solutions. An interior permanent magnet topology with crescent-shaped magnets is also introduced as t...
متن کاملThree Dimensional Re-meshing for Real Time Modeling of Advancing Process in Mechanized Tunneling
The simulation of the advancing process for arbitrary alignments during shield tunneling requires a continuous adaption of the finite element mesh in the vicinity of the tunnel face in conjunction with a steering algorithm for the Tunnel Boring Machine (TBM) advance. Moreover, the finite element mesh should match the actual motion path of the shield machine resulting from the FE-analysis in eac...
متن کاملPerformance Study and Analysis of Heterojunction Gate All Around Nanowire Tunneling Field Effect Transistor
In this paper, we have presented a heterojunction gate all around nanowiretunneling field effect transistor (GAA NW TFET) and have explained its characteristicsin details. The proposed device has been structured using Germanium for source regionand Silicon for channel and drain regions. Kane's band-to-band tunneling model hasbeen used to account for the amount of band-to...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2014