An Expeditious High Fidelity ABAQUS-Based Surrogate Tire Model for Full Vehicle Durability Analysis in ADAMS 2011-01-0187

This paper discusses an approach to construct a high fidelity surrogate tire model using a two-phase optimization-based algorithm that draws on data generated by off-line nonlinear ABAQUS tire simulations. It subsequently describes the process of Simulink-based interfacing of the resulting surrogate model to a full ADAMS vehicle model to enable accurate and expeditious durability studies. The two-phase surrogate model construction relies on an identification method that draws on the Instantaneous Center Manifold (ICM) theory. In the proposed method, a generally forced non-autonomous nonlinear structural system is represented as a sequence of harmonically excited autonomous nonlinear systems. The close-form solution of each of these systems is produced using the ICM theory. The first phase of the surrogate model construction uses an optimal Orthogonal Matching Pursuit (OMP) algorithm to unify all ICMs used to approximate the reaction force of the tire at its spindle. In the second phase, Quadratic Programing (QP) is used to form a single, closed-form mathematical model that is subsequently used as a black box surrogate model. While the method can produce surrogate models with a desired level of accuracy, higher accuracy requires more training data. This data can be a combination of experimental data and off-line simulation results. For a tire model, the proposed algorithm is validated against ABAQUS simulation results. The validated surrogate tire model is subsequently interfaced through Simulink to ADAMS for simulation typically used in the durability analysis of a High-Mobility Multipurpose Wheeled Vehicle (HMMWV). The surrogate model turns out to be more than 1,000 times faster than the ABAQUS simulation in producing tire forces while still maintaining a good level of accuracy.