Based On Control Theory, The Aerodynamic Optimization Design Study

Recently, a great progress has been made in aerodynamic optimization design technologies based on computational fluid dynamics. The control theory approach advocated by Antony Jameson is the typical methodology among them. Its major advantage is that the computation of the sensitivity derivatives of cost function with respect to design variables is nearly independent of the number of design variables, which greatly saves the computational cost. Therefore this approach has become the highlight of the CFD field, and has received much attention. Based on this optimization design theory, this thesis deeply studies some kinds of aerodynamic design problem in inviscid and viscous flow, such as aerodynamic inverse design and drag reduction problems. At the same time, the aerodynamic design optimization programs have been developed successfully. It includes the following researches.By programming an interactive interface or utilizing other visualized software, we gather the characteristic points information on fuselage surface and distribute the grid point numbers between them. A practical surface grid generation approach for complex fuselage shape has been developed. This method is flexible and practical, and assures better grid quality.Based on the transfinite interpolation theory, an improved algebraic grid generation method has been developed by introducing the technologies, such as grid orthogonality control, surface normal vector control and weighted average smoothing, etc. This improved method overcomes some old shortages and greatly enhances the ability of computational grid generation. It is much applicable to the need of rapid grid generation in optimization design cycles. We have especially researched single block grid generation of the complex shape like wing-body figuration with improved four-boundary interpolation method and the above technologies. The developed methods were tested in practical uses and proved high efficiency and practicability.Collar grid has been used specially in chimera grid computational technique. We generated its grid surface on the fuselage or missile body according to the geometry projection relation between aerodynamic components and the bilinear interpolation approach. Finally, we successfully developed a new algebra grid generation technique in virtue of the improved four-boundary interpolation.In this thesis, we put emphasis on the researches of aerodynamic inverse design and drag reduction questions for airfoil and wing using Euler equations and control theory proposed by Jameson. According to different design cases. The corresponding adjoint equations and boundary conditions are derived in detail. The numerical algorithm ofsolving the adjoint equations for different design cases have been developed by using finite volume methodology which is usually used to solve the flow governed equation. It includes the some important aspects, such as flux formulation, wall and far-field boundary treatment methodology, dissipative term formulation, etc. After the solution of the adjoint equations is obtained, the derivatives of the cost function with respect to all the design variables can be evaluated with the same operation. This can yields a significant saving over the other gradient-based techniques when there are many design variables. The effective optimization design programs for different cases are developed by integrating the following several aspects which involves the flow analysis, adjoint equation solution, gradient solution, optimal arithmetic and grid generation. Some practical design tests for airfoil and wing show that the continuous adjoint approach is very effective and useful method for aerodynamic optimization design.At the same time, we have done the research of aerodynamic optimum design for airfoils by using Navier-Stokes equations. Design cases involve the aerodynamic inverse design and drag reduction problem under fixed lift coefficient and geometry. For above design cases, we derived the corresponding adjoint equations and boundary conditions.