Multidisciplinary Design Optimization of a Hatchback Structure 2012-01-0780
Lightweight automobile has an important role in saving the energy, improving the fuel economy and reducing the exhaust emission. However, reducing the mass of the automobile need to meet the structural and NVH (Noise, Vibration and Harshness) performance requirements. With the rapid development of Computer Aided Engineering (CAE) technology, more and more people tend to research the complex engineering application problem by computer simulation. An important challenge in today's simulation is the Multidisciplinary Design Optimization (MDO) of an automobile, including mass, stiffness and modal etc.
This paper presents a MDO study in a minicar hatchback. The aim of the study is to minimize the mass of the hatchback while meeting the following requirements: (1) Structural performance. the bending stiffness is higher than the original data and the sagging residual deformation is less than the original data. (2) NVH performance. the lowest natural vibrational frequency is higher than the original data.
The sample points are obtained by the Design of Experiment (DOE) with optimal Latin Hypercube and the approximation models of mass, modal, bending stiffness, sagging are established with polynomial response surface method. The thicknesses of the major components in the hatchback are selected as design variables. The approximation models are optimized by sequential quadratic programming method.
A good agreement between the predictive values of the approximation and the results of finite element simulation with the error less than 5% is demonstrated. The predictive values of the approximation meet the engineering requirement. Consequently, the optimized results have higher accuracy. In the process of the optimization design, the weight of the hatchback reduced by 6.5% and the new hatchback meets all the prescribed requirements about structural and NVH performance.