ECCOMAS 2024

Two-Objective Design Exploration of Body Proportions and Aerodynamics of an Automobile

  • NAKASHIMA, Takuji (Hiroshima University)
  • MIZOGUCHI, Ryota (Hiroshima University)
  • HIRAOKA, Takenori (Hiroshima University)
  • KOIZUMI, Iwao (Hiroshima University)
  • NOUZAWA, Takahide (Hiroshima University)
  • SEO, Kohei (Mazda Motor Corporation)
  • SHIMIZU, Keigo (Hiroshima University)
  • NAKAMURA, Yusuke (Hiroshima University)
  • OYAMA, Akira (JAXA)
  • TSUBOKURA, Makoto (Kobe University)

Please login to view abstract download link

To achieve sustainable development, aerodynamic drag reduction is strongly required for automobiles. On the other hand, the design of the vehicle body influences the customer's purchasing decision. Both of these performances depend on the external body shape, and design support technology for external body shape considering them is required. In this study, we demonstrate a two-objective design exploration for vehicle body proportions and aerodynamic performance, as a feasibility study of a design support technology that considers both design and aerodynamic performance. The original body shape for the design exploration is a 3D model of a product SUV vehicle for rendering, and the morphing parameters defined in our previous study[1] are used to generate the vehicle body shape. Then, the values of the morphing parameters for reproducing the macroscopic shape of SUV vehicles on the market are obtained, and their principal component scores are adopted as the design variables. One objective function is the score of first principal component of the proportions obtained from a group of production automobiles. Because these scores had a correlation with the impression that humans received from exterior of the vehicle body, we considered them as a surrogate index for the design characteristics. The other objective function has been defined as aerodynamic drag, and a time-series simulation of turbulent flow field around the vehicle based on Building Cube Method (BCM) and Immersed Boundary Method (IBM) was conducted to evaluate the objective function. A multi-objective Efficient Global Optimization (EGO) based on the Kriging model was applied in the two-objective design exploration. As a result of the design exploration, we confirmed that the Pareto front of the two objective functions progresses with the sample additions. As a group of Pareto solutions, we obtained a group of vehicle bodies with the lowest aerodynamic drag for each vehicle body proportion characteristic.