【作者】 唐文献；

【导师】 方明伦；

【作者基本信息】 上海大学 ， 机械设计及理论， 2003， 博士

【摘要】 以数字(Digital)、可视(Visual)、网络(Web)、虚拟(Virtual)、协同(Collaborative)、集成(Integration)、智能(Intelligence)、绿色(Green)和安全(Security)为特征的制造业信息化进程的不断推进，知识已成为企业赢得竞争的重要手段和获取利润的关键资源。21世纪，企业竞争的焦点在于产品的创新开发和对市场的快速响应。成熟的商品化CAD系统不能支持设计师在产品开发早期的协同性创造思维。由此，作者提出了知识驱动协同创新(Knowledge Driven Collaborative Creation,KDCC)的产品开发模式，构建知识驱动协同创新的产品开发支持系统基本框架，研究以过程链为手段的方法体系；以广义知识模型为主体的模型体系；以层次化多级模糊综合评价和灰色关联分析为基础的分析体系；以知识建模、协同创新、仿真分析、虚拟优化、系统评估、群组决策为内涵的技术体系。 对比分析传统的和并行的产品开发方法，融合知识工程技术于产品创新开发领域，以知识驱动几何(Knowledge Driven Geometry,KDG)的CAD技术为基础，提出知识驱动协同创新的产品开发模式。从“发散—收敛”的协同性思维模式、内涵、原理、方法及支撑技术等方面探讨其深层的产品开发理念；构造由分布式广义知识库支持的知识驱动协同创新的产品开发体系，论述其体系特征及实现的关键技术；引入六层结构圆饼图描述知识驱动协同创新的运行和管理模型。在知识驱动协同创新的产品开发模式下进行的产品创新开发，知识是产品概念创新、构型创新的驱动力，知识的获取、积累、共享和重用是基于产品创新的行为和需要，创新的产品及其所具有的竞争力是源于知识、基于协同的，是不同的设计师和协同工作组大量互动作用的结果。 知识驱动协同创新的产品开发模式的实现不仅与技术有关，而且涉及流程和管理。与此相适应，建立具有统一知识管理的过程链是基础，可靠的产品描述是保证。为此，提出广义知识及其管理的概念，创建基于广义知识管理(General Knowledge Management,GKM)的产品协同创新开发过程链。把知识驱动协同创新的产品开发模型分为产品描述模型、开发资源模型和开发过程模型三种，采用面向对象技术，以STEP／EXPRESS和KBE／Intent(KBE,Knowledge-Based Engineering)语言为建模工具，探讨支持产品协同创新开发的过程链建模技术。在哲学思维上，把本体论(Ontology)和技术创新的自组织进化(Self-organized Evolution)观有机结合起来，研究基于本体知识的协同创新自组织进化方法论，从本体知识驱动、自组织进化设计的层面打开创新设计的黑箱，深化、丰富和发展现有的创新设计理论。为增强知识驱动协同创新的实施效果，从知识获取、表达和使用等方面论述有效支持协同创新开发过程链实现的策略。 决策与评价是设计工作中密切相关的两项工作，提出支持产品协同创新开发的六元决策目标及对应的决策模型。建立产品协同创新开发评价的指标体系及其评价模型，系统地论述层次化多级模糊综合评价和基于灰色关联分析的决策分析方法。将创新设计与虚拟技术相结合，提出基于仿真分析、虚拟优化过程及灵敏度分析的上海大学博士学位论文用纸智能再设计策略:运用虚拟样机技术，阐述产品协同创新的仿真方法，并对仿真结果的静态数据进行秩和验证和统计特性分析，对动态数据进行时序分析验证和马尔可夫预测研究:采用优化设计方法，对协同创新开发的产品进行虚拟参数优化并对优化结果进行灵敏度分析。这样，能够将分散的单元技术，如CAD、CAE和仿真分析(Simulation AnalysiS)等，进行集成应用，提供一个更全面的了解产品整体性能的方法，为产品的协同创新开发及再设计提供分析与仿真方法的支持。 在基于知识的工程(KBE)环境下一UGll/KF(Knowledge Fusion)，研究支持知识驱动协同创新的产品开发原型系统(KBE Bui lder)的构建方法，提出系统实现的关键技术的解决途径。以缝纫机和断路器两类产品的协同创新开发为例进行原型系统应用验证及应用效果分析。 应用案例表明:知识驱动协同创新的产品开发模式，采用协同性思维模式，吸纳了单元技术创新的成果，创造了在知识工程环境下，集成多种先进单元技术的群体协同创新的新模式，是产品创新开发领域新方向的有益探索;基于广义知识管理的产.另;协同创新开发过程链的引入，有利于加快产品创新开发的知识搜寻，优化知识驱动协同创新的产品开发过程;分析方法和仿真方法提供了对产品开发过程链的有效支持;基于本体知识的协同创新自组织进化方法，实现了产品创新开发的知识驱动自动化(Knowledge Driven Automation，KDA)，是现有创新设计理论内涵、原理，，方法的深化和拓展。KBE Bui lder具有可操作和实用价值，它能够有效地把企业i三‘有的知识资源应用于产品创新开发领域，提高产品创新度，创造有竞争力的产品;另一方面，应用KBE Bui lder的好处是易于产品概念优化，有利于产品及过程知识的积累、共享和重用，它是制造业信息化的典型示范。更多还原

【Abstract】 With the constant advance of manufacturing informationization, which is characterized with digital, visual, web, virtual, collaborative, integration, intelligence, green, security, knowledge has become the important means to win the competition and the key resources to obtain profits for enterprises. The competition among enterprises focuses on the creation development of products and the rapid response to the market in the 21st century. The commercial CAD system can’t support collaborative creation thinking of designer in the early days of product development. Therefore we propose the new pattern of product development of knowledge driven collaborative creation (KDCC), and construct the basic framework of the system, which supports the product development of KDCC. At the same time, we study the method system taking process chain as means; the model system taking the general knowledge model as main-body; the analysis system based on layering fuzzy synthetic evaluation and grey relative analysis; the technical system taking the knowledge modeling, collaborative creation, simulation analysis, virtual optimization, system evaluation, group decision as connotation.By comparing traditional methods of product development with concurrent engineering, we fuse the knowledge engineering technology to the field of product creative development and creatively present the product development pattern of KDCC, which is based on the CAD technology of knowledge driven geometry. In respect of the collaborative thinking pattern of "divergence-convergence", connotation, principle, method and supported techniques, we discuss the deep hierarchic development ideas of KDCC for creative product. The architecture of product development of KDCC is build up, which collaborative creation is supported by distributed general knowledge base, and discuss the system character and key technology to realize. In order to descript the running and management mode of KDCC, the pie with six hierarchies is lead into. In this pattern to develop creative product, the knowledge is driving force for product conceptual creation and structure creation. It is based on behavior and demand of product creation to capture, store, share and reuse knowledge. Creative product and its competition force are rooted in knowledge and based on collaboration, which are outcome that different designer and term work are mutual consultation.Implementation of KDCC is referential to technology, and come down to flow and management. In conformity with this, it is the basis, which builds the process chain with unified knowledge management, and reliable product description is guarantee. For this reason, we put forward to concept of general knowledge management (GKM) and build the process chain of product collaborative creation development (PCCD) based on GKM. Under the pattern of KDCC, the product development model is divided into three types, which are product description model, development resource model and development process model. By adopting object-oriented technology and taking STEP/EXPRESS and KBE/Intent language as modeling toolkit, modeling methods are studied, which supports KDCC. We introduce theory of ontology and self- organized evolution to study methods of self-organized evolution based on ontology knowledge for collaborative creation, and open the "black box" of creative design through the ontological knowledge drive and self-organized evolution design. It deepens, enriches and expands existent theory ofcreative design. In order to enhance the effects of KDCC in practice, strategies are discussed in matters of capturing, expressing and using knowledge, which effectually supports to realize process chain of PCCD.Decision and appraisal are related closely during the course of design. The six-element decision aim and its model are proposed which supports PCCD. Index system to evaluate PCCD and appraisal model are established. We discuss systematically layering fuzzy synthetic evaluation based on AHP and decision analyzing method based on grey relative analysi更多还原