关键词: 数字孪生, 模块化设计, 模块粒度优化, 内聚耦合, Floyd算法, 微耕机变速箱

Abstract: The real-time product modular design and module granularity optimization based on digital twin can strengthen their abilities of quick response to the market.A module granularity optimization analysis model was presented through acquiring and managing data timely with the digital twin technology.Four kinds of relevance relationship were provided to establish the direct relevance relationship matrix.The fully relevance relationship matrix was gained by adding the indirect relevance relationship measured with Floyd algorithm to the direct one.The module granularity analysis process based on digital twin was established and the granularity analysis indexes of component correlation strength,correlation density,module cohesion and coupling degree were proposed to provide guidance for the division and optimization of module granularity.An adaptive crossover and mutation genetic algorithm with elite-preserving was designed to optimize the sorting of full relevance relationship matrix.The feasibility of the method was proved by the application in a gearbox of micro-cultivator of an agricultural machinery enterprise,which laid a foundation for the further research on evolution of master structure of modular product family.

Key words: digital twin, moduler design, module granularity optimization, cohesion and coupling, Floyd algorithm, micro-cultivator gearbox