首页> 外文会议>Proceedings of the and Steel Technology Conference >Techniques for the Successful Commissioning and Tuning of Level 2 Models
【24h】

Techniques for the Successful Commissioning and Tuning of Level 2 Models

机译:2级模型成功调试和调整的技术

获取原文

摘要

A successful Level 2 model commissioning is determined by the performance improvement in the measured deliverables and the time required to reach this performance, while incurring minimal impact on productivity. Commissioning a new model to these standards is no longer considered exceptional or out of the ordinary but is the norm expected by today's customer. This paper discusses techniques used to successfully commission and tune Level 2 models and control functions. These techniques include first principle-based simulation, either on a discrete piece basis for the models or on a real-time basis for the in-bar control functions, the replaying of captured real-world I/O in a software mill simulator, a realistic and thorough factory acceptance test, and the real-time shadowing of live inputs with the full automation system. Once the models are commissioned, analytical, and statistical methods using large sets of historical data are used to optimize performance and to develop and test new models and control strategies. Finally, concepts are presented detailing how software design facilitates the implementation of these techniques in a manner that both minimizes maintenance and maximizes portability to future computing platforms. Examples are provided from the recent commissioning of new Roughing Mill, Finish Mill, Shape, and Coiling Temperature models at the ISG-Mittal Cleveland 84" Hot Strip Mill.
机译:成功的2级模型调试由测量可交付成果的性能改进以及达到这种性能所需的时间,同时产生对生产率的最小影响。将新模式调试到这些标准不再被认为是异议的特殊或出于普通的,而是当今客户预期的常态。本文讨论了用于成功委托和调整2级模型和控制功能的技术。这些技术包括基于主要的基础模拟,用于模型的离散碎片基础,或者在栏中控制功能的实时基础上,在软件磨机模拟器中重放捕获的现实世界I / O,a现实和彻底的工厂验收测试,以及具有完整自动化系统的实时输入的实时阴影。一旦模型被委托,使用大集合数据的分析和统计方法都用于优化性能并开发和测试新模型和控制策略。最后,提出了概念,详细介绍了软件设计如何以最大限度地减少维护和最大化对未来计算平台的可移植性的方式实现这些技术的实现。在ISG-Mittal克利夫兰84“热带磨机中最近的新粗轧机,饰面轧机,形状和卷积温度模型的近期调试提供了示例。

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