The Study Of Several Shape Optimization Problems In Engineering Design

In engineering design, the optimization of the shape is to be the lightest quality, smallest, largest and most profitable capacity and so on. In solving complex engineering optimization problems, the shape optimization method can find the optimal design from many designs. This will not only improve the design quality and design efficiency, but also achieve significant economic and social benefits.In this thesis, several optimizations in engineering design are optimized as follows:1) In this thesis, the optimization methods of the components related to engine are studied; The force of the components are analyzed; the optimization's methods of the components with symmetry are researched; analytical sensitivity analysis optimization methods are in-depth studied, the successful integration of the methods of Symmetric Optimal Design and the analytical sensitivity analysis optimization together. With the given optimization method, the symmetric shape of the component has been optimized design. The maximum stress value obtained is compared the results of the existing maximum value, the maximum stress value decreased by 2.4%, the shape of this optimization method is proved that it's correct.2) In this thesis, a predicted effective constrained optimization and analytical sensitivity analysis optimization methods of the components related to engine are researched, the integration of the predicted effective constrained optimization and analytical sensitivity analysis optimization methods together, a predicted effective sensitivity analysis optimization methods is proposed. With this method, the design of the hole on the components is optimized. The maximum stress value decreased by 6.4%, compared with the existing maximum value, the shape of this optimization method is proved that it's correct.3) In this thesis, the optimization methods of the components related to engine and a predicted effective constrained optimization is researched. Sensitivity analysis optimization method of the components is researched. The integration of the symmetry optimization's methods, the predicted effective constrained optimization and analytical sensitivity analysis optimization methods together, a predicted effective constrained symmetry sensitivity analysis optimization methods is proposed. With this method, the design of the hole on the components is optimized. The maximum stress value decreased by 8.8%, compared with the existing maximum value, the shape of this optimization method is proved that it's correct.4) A robotic fish driven by magnetic is researched, the fish head is embedded the NdFeB permanent as magnet drive, the principle is that alternating magnetic field can be converted to machine swing the fish head of mechanical energy from changing the driving frequency of alternating magnetic field, driving the NdFeB magnetic drive embedded in the robotic fish head. Then they drive copper films caudal fin as volatility and produced thrust coupled with the liquid to achieve the above bionic swimming.The micro-robot fish's caudal fin is the use of variable cross-section film like cantilever beam which is fixed at the micro-robot fish's head. The shape of Micro-Robotic Fish tail is optimized, the equal area conditions are proposed, and the discrete variable method is used to optimize the shape of the caudal fin. This optimization of the best fish tail shape of the machine, the machine can significantly improve the fish propulsion performance, theoretical analysis and experimental prove that the proposed discrete optimization method of the tail fin shape is effective, and it's laid the foundation for in-depth study of fish and bionic structural shape optimization research.5) This retreat optimization methods are carried out. When the function itself is not a derivative case, an accurate retreat optimization methods is given. It means that the search range is determined every time, accurate method has the small search range than the retreat method. Convergence accuracy is certain, using accurate retreat optimization methods' objective function to get optimal solution, has high precision and speed than the retreat method.