【24h】

MULTI-AGENT BASED FRAMEWORK FOR LARGE SCALE VISUAL PROGRAM REUSE

机译:大型视觉程序重用的基于多代理的框架

获取原文
获取原文并翻译 | 示例

摘要

Today's application engineers are committed to the reuse of programs for performance and economic reasons. Moreover, they increasingly have to complement application programs with less information and in shorter time. The reuse of already implemented programs is therefore fundamental. We have implemented a process automation specific framework that supports reuse of our domain specific visual language. The visual Function Block Language is used for power plant and paper machine controls. The reuse framework discussed in this paper relies in identification and usage of templates, which are used for generating actual application software instances. The framework automates data mining with software agents collecting metadata. The metadata is send ahead to the receiver agent that stores the data into the central database. Another agent analyzes stored data and performs template matching. Again another agent is called for more detailed template match comparison. Although the database is centralized, the agents can be distributed and run in intranet. The framework implementation is pure j ava based and runs on JADE-FIPA agent platform.
机译:出于性能和经济原因,当今的应用工程师致力于程​​序的重用。而且,它们越来越需要用更少的信息和更短的时间来补充应用程序。因此,已执行程序的重用是基础。我们已经实现了特定于流程自动化的框架,该框架支持重用我们特定领域的可视语言。可视功能块语言用于发电厂和造纸机控制。本文讨论的重用框架依赖于模板的标识和使用,这些模板用于生成实际的应用程序软件实例。该框架使用收集元数据的软件代理自动进行数据挖掘。元数据被提前发送到将数据存储到中央数据库的接收器代理。另一个代理分析存储的数据并执行模板匹配。再次要求另一个代理进行更详细的模板匹配比较。尽管数据库是集中式的,但是代理可以在Intranet中分发和运行。该框架实现是基于纯Java的,并且在JADE-FIPA代理平台上运行。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号