Abstract
High-fidelity simulation of underbody blast loading is essential for evaluating vehicle structural response and occupant injury risk under extreme loading conditions. In practice, the predictive accuracy of blast simulations based on the structured arbitrary Lagrangian–Eulerian (S-ALE) method is highly sensitive to several flow-field control parameters, and model calibration still relies heavily on empirical trial-and-error procedures. This limits the robustness and repeatability of simulation-based analysis. To address this issue, this study proposes a calibration and validation framework for a full-vehicle underbody blast finite element model based on an improved Theil inequality coefficient (TIC). A full-vehicle blast test was conducted to obtain structural deformation and occupant lower tibia force responses under a 500 g TNT charge buried directly beneath the front passenger seat. Based on the test configuration, an S-ALE coupled finite element model was established, and six key parameters affecting the blast flow field were selected as calibration variables, namely the number of coupling points, leakage penalty coefficient, minimum volume fraction, S-ALE mesh size, soil density, and computational time step. The improved TIC metric was employed to quantify the correlation between simulated and experimental tibia force histories, and a surrogate-assisted optimization strategy was adopted to identify the optimal parameter combination. The calibrated model shows substantially improved agreement with the test results. The peak-force errors of the left and right lower tibia responses were reduced from 11.98% and 6.53%–2.35% and 3.34%, respectively, corresponding to a marked enhancement in prediction accuracy. The proposed framework provides an effective approach for simulation–experiment correlation and parameter calibration of full-vehicle blast models, and offers a practical basis for structural response assessment and subsequent protective design under underbody blast loading.
Keywords
Get full access to this article
View all access options for this article.
