Bi-system Co-evolutionary Algorithm And Key Techniques Of The CAD System For Satellite Module Layout

The layout design of satellite module plays an important role in satellite schematic design based on satellite public platform. In general, the layout design of satellite module means that a group of the given apparatus and equipment (objects for short) are rationally located in the limited space inside (or outside) the module. Several design objectives need to be optimized possibly, and constraint conditions of space and performance (such as interference, mass distribution characteristic, thermal analysis and electromagnetic compatibility) need to be satisfied. This study has an important impact upon reducing the design period, saving the cost, improving the performance of the satellite, etc. In mathematics, the layout design of satellite module belongs to combinatorial optimization and NP-hard problem. In engineering, it belongs to the complex engineering system. The primary difficulties for solving this problem are not only to solve the combinatorial explosion, but also to achieve the practicability of engineering.The engineering background involved the project that the study and development of simulation platform for satellite layout optimization, which has been supported by the National Nature Science Foundation of China. This dissertation presents an efficient and practical layout optimization algorithm for satellite module and several key development techniques of the CAD system. The main contributions are as follows:(1) The background involved the layout design of spacecraft (such as satellite). Based on the idea of collaborative design, this dissertation presents a heuristic algorithm for the layout design of complex engineering system, i.e., the bi-system co-evolutionary algorithm with variable-grain model (BCCEA for short). According to the cooperative co-evolutionary algorithm (CCEA for short), the strategies of variable-grain model and bi-system complement, initially the original system P (Grade 1) is divided into several parallel sub-systems based on physical (structural or discipline) units. Then the original system P is duplicated to two independent system A and B (Grade 2) respectively. Both system A and B include the abovementioned several parallel sub-systems (Grade 3). The bi-system co-evolutionary framework is constructed by the two systems with complementary strategies, which are different in the number of design variables and the complexity of fitness function. This dissertation adopts the variable-grain model of design variants. A and B systems share the design variants of original problem. Both system A, B and subsystems of B co-solve problem using evolutionary algorithm. System A and B (actually, they are subsystems of system A and B) collaborate with each other through the individual migration between the two systems. This study aims at increasing the diversity of population, and reducing the complexity of computation. The proposed algorithm improves the precision, efficiency and success rate of computation for the optimization and simulation of engineering system.(2) This dissertation presents several key development techniques in the layout design process of satellite system platform. These key techniques include the overall design of CAD platform, the three descriptive models of object and layout space, the techniques of self-orientation assembling, the two coarse-fine interference collision detection in Pro/E and so on. From the viewpoint of engineering, this dissertation combined the proposed algorithm framework with several popular algorithms libraries (such as GAlib genetic algorithm library, RAPID three-dimensional collision detection package) and the commercial engineering software (such as Pro/Engineer). The layout optimization software has been developed for three-dimensional complicated model entity. This software has a steady kernel of algorithm library, and has the abilities of three-dimensional interference detection and self-orientation assembling. This software is the important component of the abovementioned system platform. These techniques of secondary development and the layout software have been validated by the CAD system of layout optimization, and the study and application of CAD system secondary development will benefit from them.The satellite module in this dissertation is different from the previous study. It adds the antennae and solar arrays. The states of launching and flying for the antennae and solar arrays have been analyzed, where the form of apparatus and equipment is three-dimensional complicated model entity. The results of numerical experiments for the satellite module showed that the proposed method was feasible and effective for the layout design of satellite module. The system platform for satellite layout optimization design is integrated the proposed method, engineering software, algorithm library and so on. The system platform has been checked and accepted by user, and has been applied in the layout design and simulation of a certain satellite.In theory, this dissertation presents a bi-system co-evolutionary algorithm with variable-grain model. The research will advance the further study of co-evolutionary algorithm.In practice, this dissertation studies the application of the proposed algorithm in the layout design of the satellite. Additionally, several key development techniques are also presented. The developed system platform can be applied to the layout design of the other spacecrafts and satellites.