Agent-Based Precision Manufacturing Systems

基于代理的精密制造系统

• 批准号: 9908174
• 负责人: Paul Wright
• 金额: $ 1万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1999
• 资助国家: 美国
• 起止时间: 1999-10-01 至 2003-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9908174&HistoricalAwards=false
• 关键词: Agent; Based; Precision; Manufacturing; Systems

This grant provides additional funding for the development of an agent-based system for the design, process planning and manufacturing of precision milled parts. The agents that will be developed consist of software packages containing information, databases, and instructions related to specialized knowledge on important aspects of the physics of manufacturing or specialized procedures. In this work, the primary interest is on developing algorithms that embody the process physics and that will allow the minimization of residual stresses, enhancement of edge quality and improvement in tolerances in milling of precision mechanical components. The result of the agent output will be minor adjustments to specific tool paths or operating conditions that will make a considerable improvement in part quality with acceptable sacrifice in "time to first part." This research will measure this trade-off between production speed and part quality quantitatively. Experiments will be performed in which parts will be fabricated with a high edge quality and tolerance, and with concern for other part quality issues such as minimizing subsurface deformations and residual stresses. The agents will be integrated into a high-end CAD system for detailed design, macro and micro planners, a tool path planner, and manufacturability feedback.If successful, the results of this research will lead to new software capability creating a broader Internet-based Agent Based Manufacturing System. The system will specifically include agents for precision machining improvements and address issues of conflicting advice from competing agents. The goal is to show that a considerable benefit in edge quality and tolerance can be achieved with only minimal sacrifice in production time compared to "normal" process planning. Finally, a major focus will be on conflict resolution between agents and trade-offs between agent advice and cost-benefit analyses. Specific methodologies, based in part on game theory, will be developed to address these conflicts.

这笔拨款为开发基于代理的系统提供了额外的资金，用于精密铣削零件的设计、工艺规划和制造。将要开发的代理由软件包组成，其中包含与制造物理或专门程序的重要方面的专业知识相关的信息、数据库和指令。在这项工作中，主要的兴趣是开发算法，体现过程物理，将允许残余应力最小化，提高边缘质量和改进公差铣削精密机械部件。代理输出的结果将是对特定的刀具路径或操作条件进行微小的调整，这将在可接受的牺牲“第一个零件的时间”的情况下大大提高零件质量。本研究将定量测量生产速度和零件质量之间的权衡。将进行实验，其中零件将以高边缘质量和公差制造，并关注其他零件质量问题，例如最小化地下变形和残余应力。这些代理将集成到一个高端CAD系统中，用于详细设计、宏观和微观规划、刀具路径规划和可制造性反馈。如果成功，这项研究的结果将导致新的软件功能，创建一个更广泛的基于internet的基于Agent的制造系统。该系统将特别包括用于精密加工改进的代理，并解决来自竞争代理的冲突建议的问题。我们的目标是表明，与“正常”工艺计划相比，在边缘质量和公差方面可以获得相当大的好处，而在生产时间方面只需要最小的牺牲。最后，重点将放在代理之间的冲突解决以及代理建议和成本效益分析之间的权衡。具体的方法，部分基于博弈论，将开发解决这些冲突。