A Certain Type Of Guided Rockets, Multidisciplinary Design Optimization Study

Guided rocket design involves many different disciplines, including aerodynamics, trajectory, and propulsion, etc. With the increasing products requirements, engineers recognize that these disciplines should be treated as a whole system instead of separate ones with the goal to obtain a global optimal solution based on considering the coupled effects among disciplines. This makes such complex system design become a large-scale optimization problem which may not be solved by traditional methods. Multidisciplinary Design Optimization (MDO) is a new promising method to solve such complex engineering optimization problems.This paper focuses on MDO methods and its application to guided rocket design. The major work of this paper includes:In the first part, after a brief overview of the development of MDO, related technologies to MDO especially the Concurrent Subspace Design Optimization (CSSO) are discussed, which include the categories of MDO methods, CSSO and its procedure, Genetic Algorithm (GA) and some related theory and Radial Basis Function (RBF) net-work and some other approximation technology such as Variable Complexity Models, Kriging Method and so on.In the second part, the MDO model of guided rocket is discussed and analyzed from the point of system. The aim of this part is to determine the optimization procedure, objective function, design variables and parameters, constraints and so on. After brief discussion, it is pointed out that the CSSO will be adopted, and GA is used to optimize the objective function while RBF network is used to create approximate model. In the followed part, the models of aerodynamics, propulsion, trajectory and mass are created from the disciplinary view.In the third part, the results and discussions are provided. In this part, figures of some design parameters are listed. From the results, the objective function of guided rocket improves greatly which proves the efficiency and feasibility of CSSO.