【作者】 倪晓宇；

【导师】 易红； 汤文成；

【作者基本信息】 东南大学 ， 机械制造及其自动化， 2005， 博士

【摘要】 基于CAE技术的发展趋势以及协同技术的运用,本文从协同CAE系统的构建及系统实现的相关技术方面进行了研究,主要内容及成果包括:(1)协同策略是实现CAE协同的关键。本文对CAE技术特点与协同要求进行分析,深入研究了协同优化理论,提出了CAE协同过程中的三种协同策略,分别定义为多任务间协同策略、单任务前处理协同策略、以及CAE与CAD设计流程的协同策略。(2)提出了层次化、模块化的协同CAE系统总体结构,以及适应多用户多任务协同分析的系统组织模式。在协同策略和系统总体结构研究的基础上,进一步给出了协同CAE系统的体系结构,包括系统的平台框架结构和系统功能结构。又根据CAE分析特点和网格计算技术的优势,将网格计算技术应用于协同CAE系统网络环境中,构建了协同CAE系统的物理结构。(3)对协同CAE分析方案的规划,主要包括建模、分析方法、优化技术以及计算结果的评价和反馈等内容进行了研究。为提高分析效率和保证计算结果的正确性和可靠性,研究了机床结构有限元前处理和后处理过程的专家系统,在分析过程中使得分析方案规划过程与计算结果的分析与评价过程具有一定的智能性。(4)对协同CAE系统过程进行了微观和宏观层面上的分析和描述,创建了协同CAE过程的树型结构模型,构成系统的宏观层和微观层。利用IDEF0对系统微观层的活动进行抽象建模,而在系统宏观层,则采用了分级规划的思想对协同分析任务进行领域级和零件级的划分,构造了协作任务模型和多用户多任务的协作任务运行结构模型,并根据CAE分析特点,提出了“定时轮询进程”的任务动态调度算法。在此基础上,提出了基于PDM的协同CAE系统的数据管理框架,并对其实现方法进行了描述,为协同CAE系统的过程管理提供了方法与技术上的支持。(5)对协同CAE过程分析中的冲突进行了首次研究,提出了用协调分解规划法来消解因“局优”组合不等于“全优”而引起的产品层冲突,并提出了用模糊评判法来消解产品部件层上的冲突。冲突消解的实现为协同CAE分析过程的顺利进行提供了有力的保证。(6)网格计算技术的应用,给协同CAE系统创造了超强的计算能力。本文首次将网格计算技术运用到协同CAE系统的网络环境的构建中,创造性地构建了网格系统的中间件,并对网格节点资源的静动态性能进行了数学描述。在此基础上,又通过模糊算法实现网格节点资源的调度。(7)根据协同产品开发环境中的协同CAE系统结构和对系统理论的研究,在先进的计算机技术和硬件条件的基础上,成功地实现了机床组件协同CAE分析系统。在文章最后详细介绍了该系统的实现方法及应用实例。更多还原

【Abstract】 Based on the trend of CAE technology development and the application of collaboration technology, some research on the system of collaborative CAE and the realization for the system was done, including as following:(1) Collaborative strategy is the key to realize the collaborative CAE. Some analyses were done on the character of CAE technology and the demand of collaboration, and some research on the collaboration optimization theory was done deeply, then three strategies were put forward for CAE collaboration, which are multi-task collaborative strategy, single-task pre-process collaborative strategy and strategy of collaboration between CAE and CAD design flow.(2) A hierarchical, modularized structure of collaborative CAE system was built. And organization pattern were brought forward to suit to the multi-user and multi-task analysis. Based on the collaborative strategies and gross structure, the architecture of collaborative CAE was built, which includes platform frame and functional structure. According to the character of CAE analysis and the advantage of gridding compute technology, the gridding compute technology was adopted into collaborative CAE. Based on that, the physical structure was created.(3) Some research on analysis scheme planning of collaborative CAE was done, which includes modeling, analysis methods, optimization technologies, evaluation and feedback of results. To improve the validity and reliability of result, some study about expert system for pre-process and post-process was done. Expert system ensures the processes, including analysis scheme planning and evaluation of results, to have intelligence to some extent.(4) The tree model of collaborative CAE process was built, which includes micro-layer and macro-layer. The activities in micro-layer were abstracted and the process models were built using IDEF0 language. In macro-layer, the task was decomposed by hierarchical planning method from analysis fields and product’s work-piece grade. Then collaborative task model, multi-task and multi-user operational structural model were created. Based on that, the method named timing polling process was put forward for task schedule. According to the character of collaborative CAE, PDM-based database management framework was brought forward and the methods of realization were described for information transferring and controlling in collaborative CAE process. The methods were adopted to implement the process management for collaborative CAE system.(5) The conflict in collaborative CAE was analyzed for the first time. The coordination decomposition planning method was proposed to resolve the conflict in product layer, caused by the variational inequality between combination of local optimum and global optimum. And the fuzzy appraisement method was proposed to resolve the conflict in part layer. The realization of resolution for conflict ensured the analysis process of collaborative CAE to run successfully.(6) The application of grid computing technology can create super-powerful compute ability for collaborative CAE system. In this paper, some research on application of grid computing technology in collaborative CAE system was done for the first time. The middle ware of grid system was built creatively. The static and dynamic performance of the grid node resource was described in mathematics. Based on this, fuzzy method was adopted to solve the problem of grid node resource schedule.(7) According to the research on the collaborative CAE architecture and system theory, on the basis of advanced computer technology and hardware conditions, the system of collaborative CAE was built successfully. At last, the methods of system realization and an example of application were introduced in detail.更多还原