An Enhanced Input Uncertainty Representation Method for Response Surface Models in Automotive Weight Reduction Applications

Vehicle weight reduction has become one of the viable solutions to ever-growing energy and environmental crisis. In vehicle design, response surface model (RSM) is commonly used as a surrogate of the high fidelity Finite Element (FE) model to reduce the computational time and improve the efficiency of design process. However, RSM introduces additional sources of uncertainty, such as model bias, which largely affects the reliability and robustness of the prediction results. The bias of RSM need to be addressed before the model is ready for extrapolation and design optimization. For the purpose of constructing and correcting the bias in RSMs, scheduling Design of Experiments (DOEs) must be conducted properly. This paper develops a method to arrange DOEs in order to build RSMs with high quality, considering the influence of input uncertainty. The proposed method focuses on assigning samples corresponding to the obtained Probability Density Function (PDF) of represented input, and it shows its advantage in the process of model bias correction. A real-world vehicle weight reduction design example is used to illustrate the validity of the proposed method.