Model validation and calibration via back analysis for mechanized tunnel simulations – The Western Scheldt tunnel case. (September 2015)

• Record Type: Journal Article
• Authors: Zhao, Chenyang; Lavasan, Arash Alimardani; Barciaga, Thomas; Zarev, Veselin; Datcheva, Maria; Schanz, Tom
• Is Part Of: Computers and geotechnics. Volume 69(2015)
• Page Start: 601
• Page End: 614
• Publication Date: 2015-09
• Subjects: Mechanized tunneling -- Numerical simulation -- Sensitivity analysis -- Meta-modeling -- Back analysis -- Optimization; Engineering geology -- Data processing -- Periodicals; Soil mechanics -- Data processing -- Periodicals; Rock mechanics -- Data processing -- Periodicals; 624.1510285
• DOI: 10.1016/j.compgeo.2015.07.003
• Languages: English
• ISSNs: 0266-352X

Abstract

Abstract: In this research, Finite Element (FE) method is applied to simulate the shield supported mechanized excavation of Western Scheldt tunnel in the Netherlands. Both 2D and 3D numerical models are created to predict the system behavior. Sensitivity analysis and parameter identification techniques are utilized to calibrate and validate the model based on field measurement. The mechanical behavior of the soil is modeled by an advanced elasto-plastic model, namely Hardening Soil model correlating small strain stiffness (HSS). Global sensitivity analysis is carried out in this paper to evaluate the relative sensitivity of model response to each input parameter. Thereafter, a parameter identification technique (back analysis) is employed to find the optimized values of the selected parameters. To accomplish this, the computationally expensive FE-model is replaced by a meta-model in order to reduce the calculation time and effort. Moreover, a soft soil constitutive model based on the modified Cam-clay model deals with primary compression of fine grained soils, is assigned to the clay layer to further improve the numerical prediction of system behavior. Due to the importance of model subsystems, such as face pressure and volume loss, the sensitivity of model response to subsystems has been evaluated. The results show that optimized parameters obtained via back analysis make the numerical simulation capable to well predict the ground settlement.