The Correction Scheme of Enhanced Response Surface Method to Support Full Vehicle MDO Design

In this paper, multi-disciplinary design optimization (MDO) analysis is applied into the virtual vehicle design process of a new vehicle (XYZ) of Guangzhou Automobile Group Co. (GAC). To conduct the MDO analysis, a meta-model is built by using the response surface methods (RSM). Due to the fast-changing nature of automotive body design, the Pareto frontier is calculated to conduct multiple optimized cases with different performance constraints to provide flexible choices to support the complicate decision-making process of body design. Later, this character of Pareto frontier is founded useful to deal with the uncertainties introduced by the errors of RSM when compared to their corresponding confirmation runs done by finite element analyses (FEA). Furthermore, an iteration process is developed to ensure the convergence of the global Pareto frontier. The advantages of the proposed methods are demonstrated in the given MDO analysis example with the involvement of highly nonlinear natures of the vehicle crash safety responses.