Research And Application On Design Of Multi-performance Of Complex Product Based On Fuzzy Coupling Degree

Modern complex product commonly includes multidisciplinary knowledge, with performace changing complicatedly under massive coupling factors in many physical fields. The coupling factors result in much iteration during design. Analysis of coupling degree and research of multi-performance design become hot issue in the design of engineering domain.This dissertation presents a theory and method of fuzzy coupling degree, mainly including analysis of coupling factors, method of obtaining fuzzy coupling degree, sorting algorithm of fuzzy coupling degree, method of solving design parameters. On the basis of the theoretical research,a multi-performance design system is developed, and the system reaches a good effect in multi-performance design.The dissertation is organized as follows:In the first chapter, the dissertation introduces some key technologies related to multi-disciplinary and multi-performance design, and reviews the history and present status . It also presents the research purpose and content of multi-performance design.In the second chapter, the dissertation analyzes the causes of design iteration. And the causes are come down to two aspects, including fuzzy coupling degree correlated to design parameters and fuzzy coupling degree correlated to objective factors. Incidence matrix is used to analyze the relationship between performances and parameters. And the matrix is solved by the technology of QFD. Mapping rules between incidence matrix and fuzzy coupling degree correlated to design parameters are set up.Then the global fuzzy coupling degree is obtained combined with fuzzy coupling degree correlated to objective factors which is solved according to design resource, design experience and so on. At last, the coupling relationship is classified according to mapping result.In the third chapter, Genetic Algorithm is applied to sorting multi-performance fuzzy coupling degree. Due to the sorting result can not satisfy the design need, extension method ofλ-Cut Set is researched and used to get multi-performance clustering with different coupling degree, the clustering results provide much information to designer. Decoupling method is researched based on input and output information during the design. The combination of sorting, clustering and decoupling decreases design iteration and increases design efficiency.In the forth chapter, the dissertation researches the application of parallel collaborative design technology. It mainly includes analysis of collaborative design model, the transformation from fuzzy coupling degree to collaborative design model and method of establishing design goals based on satisfactory degree conversion theory. And then the design parameters are solved under collaborative mechanism.In the fifth chapter, the platform of multi-performance design is built. It also introduces main function modules and development technology and skills. The system is successfully applied in multi-performance design in chassis and compressor.In the last chapter, the dissertation summarizes the contents and the achievements of this paper, and gives the way for further research.