Research Of Optimization Design For Automobile Aerodynamic Shape Based On Adjoint Method

The mileages of highway and vehicle volume in China are both at the first place in the world while the independent research and development ability such as aerodynamic shape optimization is not fit for this condition. Aerodynamic shape optimization can reduce the aerodynamic drag through which the aerodynamic performance and fuel economy are both improved. With the lack of oil resources, the study of optimization technique has very important practical and theoretical value.Adjoint Method which is widely used in the field of optimal design of aircraft shape is applied to this study to solve the problem of blindness and low efficiency exists in the automobile aerodynamic shape optimization.First, the theoretical basis for automobile computational fluid dynamics is studied, including the generation of aerodynamic drag, the relationship between drag and car shape, the discrete method and calculation method of the numerical simulation and so on. Then the numerical calculation accuracy is studied. Ahmed model—a simple model that has similar appearance with the shape of car is applied in this study and calculation domain which has the same size with wind tunnel test is established. Boundary conditions are consistent with wind tunnel test. There are two strategies of mesh dividing—tetrahedron & triangular prism grids and hexahedron grids, and two types of turbulence model—RNG k-ε model and Realizable k-ε model. Based on the four schemes of mesh strategies and turbulence model, numerical calculation for Ahmed model is carried out. The accuracy, convergence and the consequence of wake flow field in different schemes are compared with each other through which the scheme consists of hexahedron grids and Realizable k-ε model is the most accurate and reasonable.After that, the scheme consists of hexahedron grids and Realizable k-ε model is applied to the shape optimization of Ahmed model. Adjoint Method which has small amount of calculation and wide applicable scope is applied to aerodynamic shape optimization in FLUENT through which distribution of shape sensitivity data can be obtained, and then modify the critical areas and the whole model. As a result, the aerodynamic drag coefficient is reduced by 16.66%.Finally, Adjoint Method is used for aerodynamic shape optimization of a streamlined car. The numerical calculation of a real car is carried out and aerodynamic performance is analyzed. Wind tunnel test is carried out to verify the reliability of the calculated result. Then calculation based on Adjoint Method and shape optimization is done based on this. As a result, aerodynamic drag coefficient is reduced by 2.54%.Adjoint Method is applied to the automobile industry, and a practical car model is verified which provides theory support and effective means for aerodynamic shape optimization of automobile.