韧性产品交互式概念设计与知识推理方法研究  

Research on Interactive Conceptual Design and Knowledge Reasoning Method of Resilient Products

作　　者：郭鑫[1,2] 王海涵 王杰[1,2] 张凯[1,2] 赵武[1,2] GUO Xin;WANG Haihan;WANG Jie;ZHANG Kai;ZHAO Wu(School of Mechanical Engineering,Sichuan University,Chengdu 610065;The Innovation Method and Creative Design Key Laboratory,Sichuan Province,Chengdu 610065)

机构地区：[1]四川大学机械工程学院,成都610065 [2]四川大学创新方法与创新设计四川省重点实验室,成都610065

出　　处：《机械工程学报》2023年第11期74-83,共10页Journal of Mechanical Engineering

基　　金：国家重点研发计划(2018YFB1700702);国家自然科学基金(52175241);四川省科技计划资助(2022YFG0065)资助项目。

摘　　要：设计具有更好的容错性和修正空间的韧性产品,有效应对复杂环境带来的扰动,是未来智能产品创新的关键因素之一。传统概念设计研究着重强调需求转化与功构映射,忽略了环境扰动对设计目标的影响,导致概念设计过程知识推理模糊、产品韧性水平不稳定等问题。论文以设计出具有抵御扰动、并能从破坏或冲击中恢复自身功能的韧性产品为目标,提出一种韧性产品交互式概念求解过程模型。基于“用户需求”正向收敛与“复杂环境”逆向扰动,构建韧性产品“需求-扰动”下设计目标理解及功能分解、基于功构配置的知识推理扩展方法(Requirement function knowledge deployment,RFKD)以及基于多阶段、交互式遗传算法的概念方案求解机制(Multi-stage interactive genetic algorithm for resilient conceptual design,MIGA-RCD),并以核环境下某型号智能吊装平台概念设计过程为例验证了方法的可行性与有效性。相关成果在扰动-意图-方案交互驱动的韧性概念设计理论研究中实现了创新,有助于实现设计方案的智能求解与推理。Designing resilient products with better fault tolerance and correction space,and effectively responding to disturbances caused by complex environments,is one of the key factors for future smart product innovation.Traditional conceptual design research emphasizes demand transformation and function-structure mapping,ignoring the impact of environmental disturbances on design goals,leading to problems such as fuzzy knowledge reasoning in the conceptual design process and unstable product resilience levels.The manuscript aims to design resilient products that are resistant to disturbances and can recover themselves from damage or impact,an interactive concept-solving process model for resilient products is proposed.Based on the forward convergence of“customer requirement”and the reverse disturbance of“complex environment”,the design goal understanding and function decomposition under“requirement-disturbance”of resilient products are constructed,and the knowledge reasoning extension method based on requirement function knowledge deployment(RFKD)and multi-stage interactive genetic algorithm for resilient conceptual design(MIGA-RCD).Taking the conceptual design process of a certain type of intelligent hoisting platform in a nuclear environment as an example to verify the feasibility and effectiveness of the method.The results are innovative in the theoretical study of resilient conceptual design under the influence of interference-intention-solution interactions and help to achieve intelligent solution and reasoning for design schemes.

关 键 词：韧性产品 概念设计 交互式遗传算法 知识推理 

分 类 号：TG156[金属学及工艺—热处理]