【作者】 许红静；

【导师】 郭伟；

【作者基本信息】 天津大学 ， 机械制造及其自动化， 2007， 博士

【摘要】 数字化设计技术的发展,促进了从传统的“基于经验和试制试验驱动设计”的开发手段向“基于组织知识和分析驱动设计”的研发体系转变,集成化、智能化、参数化是数字样机技术发展的重要趋势。复杂产品多学科分析、协同优化、生命周期多阶段性能分析,以及设计过程中基于知识组织、管理及重用等问题,对设计理论方法、集成分析建模方法提出了新的要求。本文针对上述问题,开展了研究与探索,研究内容和成果主要体现在以下几方面。1.提出了基于数字样机的复杂产品集成分析方法体系,对集成分析方法基本概念、技术框架、信息模型和形式化表示进行了多角度研究,研究了以数字样机建模方法、构件化分析、多学科协同优化和多阶段性能分析为核心的综合分析方法,建立了实现多层次多学科多阶段的分析过程的信息、知识以及过程表示的基础理论。2.研究分析了本体论关于知识组织与管理的特点,通过对本体类、属性和关系等要素定义,构建了集成分析方法本体模型框架,提出了基于本体的数字样机建模方法,为集成分析的实现奠定了知识表示、组织与重用等实现方法的基础。3.提出了以参数驱动为特征、可重用的构件化分析方法。以分析元作为分析知识、模型和信息等相关内容的基本组织方式,以分析树作为分析元层次结构及相互关系的组织方式,对构件化分析的基本机制、实现过程、关键技术和任务规划相关内容进行研究,形成了复杂产品分析过程参数化驱动模型和过程规划的理论方法。4.研究并提出了构件化分析与多学科协同优化建模的映射方法。该方法以分析元作为优化的基本单元,复杂产品多学科多层次分析元的性能目标作为优化约束,有效集成和利用各领域成熟的分析软件,建立分析元优化、领域优化和产品优化的多层次优化模型,实现各领域、各层次对信息和资源的共享,为协调处理各种耦合和冲突,实现多学科的协同优化设计提供了模型映射方法。5.研究面向产品生命周期多阶段的性能分析问题,并建立构件化分析与多阶段分析建模的映射方法。分析产品生命周期多阶段数字样机演化规律,将其分解为分析元的构成要素变化,建立描述产品全生命周期各类信息的产品模型组织方法。通过分析元的演化和重构,实现产品生命周期多阶段性能的分析与评估。本文着重研究水电机组主轴设计数字样机向加工数字样机的建模与实现过程,二者分析结果的差异性证明了多阶段性能分析的必要性。在以上基本理论的研究基础上,本文以水电机组作为复杂产品的具体结合对象,构建其典型产品的分析树和分析元,实现分析任务规划、构件化分析、多学科协同优化和多阶段性能分析等方法在水电机组设计开发中的应用,充分验证集成分析建模方法在复杂产品设计开发过程中的实用性和技术可行性,有效地提高了企业设计开发能力,显著地降低了成本,缩短了设计周期。更多还原

【Abstract】 The development of digital design technology brings about the transformation of research system from conventional“experience and trial-production based design”to“structure knowledge and analysis based design”, with integration, intelligentization and parameterization being the current trend of digital prototyping technology development. New requirements have been put forth on design method and integration analysis prototyping method by complex products multi-disciplinary study, cooperation optimization, product lifecycle performance analysis, as well as problems concerning knowledge structure, management and reutilization. This paper presents related research and exploration on such issues. The main research content and achievements are as follows.1. A digital prototyping-based complex product integrated analysis system is put forward. Multi-dimensional research has been conducted on basic concept, technical frame, information model and formalization; integrated analyses have been constructed based on digital prototyping modeling, component analysis, multi- disciplinary cooperation optimization, and multi-phase performance analysis; basic describing theories of information, knowledge and process have been established to accomplish multi-level multi- disciplinary multi-phase analysis process.2. Research on ontology has been conducted regarding the knowledge structure and management features. By defining elements of ontology, property and relation, integrated analysis method ontological modeling frame and ontology-based digital prototyping modeling method have been set up. The methodological basis has been established for knowledge representation, organization and reutilization for integrated analysis.3. A reusable component analysis method featured by parameter driving has been proposed. Using analysis unit as the fundamental organization method for analysis knowledge, model as well as information and using analysis tree as sub-structure and correlation of analysis unit, component analysis has been carried out on basic mechanism, realization process, key technology and mission planning. Investigation has also been performed on parameter-driven model for complex product analysis and theoretical methodology for process layout.4. A mapping method has been studied and established for component analysis and multi-disciplinary cooperation optimization. Taking analysis unit as the basic optimization unit and performance of complex product multi-disciplinary multi-level analysis unit as the optimization constraint, this method effectively integrates and uses mature analysis software of various fields and establishes the multi-level optimization model covering analysis unit optimization, disciplinary optimization and product optimization, making it possible for share of information and resources of various subjects and levels. It also provides a modeling mapping method for coping with coupling and conflict as well as realizing optimal design for multi-disciplinary cooperation.5. Multi-phase performance analysis for product lifecycle has been addressed. A mapping method of component analysis and multi-phase analysis has also been modeled. Evolution rules of digital prototyping model for product lifecycle are decomposed into variation of each element of analysis unit. Product modeling construction method is established for describing various kinds of information during the whole product lifecycle. Performance of product lifecycle multi-phase is analyzed and evaluated by evolution and reconstruction of analysis unit .This paper focuses on modeling and analysis on the transformation from designing digital prototyping model to manufacturing digital model of the spindle of hydro-turbine generating unit. The difference in the result proves the necessity of multi- phase performance analysis.In this paper, with the hydro-turbine generating unit as an example, analysis tree and analysis unit have been constructed based on the above-mentioned basic theoretical research. Various analyses applied in the design of hydro-electrical generator, including mission layout, component analysis, multi-disciplinary cooperation optimization and multi-phase performance analysis, successfully verify practicability and technical feasibility of integrated modeling in complex product design and development process. This method can effectively enhance company design and development ability, greatly reduce production cost, and shorten design cycle.更多还原