Study And Application Of Theories Of Optimization In The Integrated Design Of Flying Vehicles

With the development of various flying vehicles.the designers learn a lot about their . design. And system design technology of them gains more and more attention. To enhance the level of flying vehicle design.in this dissertation.with the help of theories of optimization.the optimal design of flying vehicles is studied systematically. The objects of this research include the Strategic Ballistic Missile.the Intercept Missile.the Air-launched Vehicle.the Reusable Launch Vehicle.the Aeroassisted Orbit Transfer Vehicle and Space Operation Vehicle. Some meaningful results are gotten and shown as following:The static optimization theory is studied according to the practice of the design of flying vehicles. A general principle of the choice of the optimization method is given. Some existing algorithms are improved and some new algorithms are presented. On this basis.an optimization algorithm library named OPTLIB is formed.The study of dynamic optimization theory is followed. To solve the problem of the coupling of the static and the dynamic parameters of the flying vehicle design.some auxiliary variables are introduced to deal with them together. Then the calculus of variation is adopted and augment form of Pontryagin's Maximum Principle is derived. As a result.it shows that the static or dynamic subsystem optimum alone is not equal to the system optimum.Three initial value problems met in the optimal design of flying vehicle are studied and conclusions are derived that: orthogonal test method can be adopted to decide the initial value of static optimization problem.some mathematical techniques can be used to deal with the costate variables of Maximum Principle and decide the initial value of the costate variable.the indirect method can be used to get the analytical solution under ideal case to guide the choice of the initial control curve in the direct method. With some numerical examples of trajectory optimization.it shows that all these methods are useful not only in accelerating the convergence but also in converging to the global optimum.The neural network is adopted to study the burning rate identification of the solid fuel ramjet. To accelerating the convergence.the Trust Region Method is used to improve the standard BP algorithm. The pre- and the post- process of the application of the BP algorithm are discussed. A general process of the identification using neural network is given. The identification result of neural network is compared with that of the Least Square Method.which shows the advantage of the former.The integrated optimal design of the system/SRM of the solid strategic missile is studied. The optimization model is setup with the minimum takeoff mass. On this basis.a large-scale software IDAMS is made.which can be used to optimize the system.ballistic and SRM parameter simultaneously. The model and the software are testified with certain missile and the sensitivity analysis of the result is made.which shows that this design can improve the system performance of the missile.The system optimization of the Intercept Missile is studied. The optimization model issetup and the corresponding software is made. The system.ballistic.aerodynamic and SRM parameter can be optimized simultaneously. These also improve the system performance of the missile. The initial scheme of this missile is also evaluated.The concept of air-launch technology is studied. On the technical basis of our country.a system scheme of three-stage solid rocket is present. The key technology and the vista of the air-launch technology are analyzed.The state-of-the-art of the Multidisciplinary Design Optimization (MDO) is summarized. The characteristic and the content of it are introduced. And the step of applying it in the design of the flying vehicle is studied. The vista of its application in the design of flying vehicle is also shown.A widely adopted MDO method named Response Surface Method (RSM) is studied. After comparison of existing RSM.a new RSM based on Orthogonal Polynomials is present. This method is...