• 主题
高级搜索  >
历史搜索 清空历史
首页> 美国政府科技报告 >Hierarchical causal modeling for large industrial plants supervision
【24h】

Hierarchical causal modeling for large industrial plants supervision

机译：大型工业厂房监管的分层因果模型

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A supervision system has to analyse the process current state and the way it will evolve after a modification of the inputs or disturbance. It is proposed to base this analysis on a hierarchy of models, witch differ by the number of involved variables and the abstraction level used to describe their temporal evolution. In a first step, special attention is paid to causal models building, from the most abstract one. Once the hierarchy of models has been build, the most detailed model parameters are estimated. Several models of different abstraction levels can be used for on line prediction. These methods have been applied to a nuclear reprocessing plant. The abstraction level could be chosen on line by the operator. Moreover when an abnormal process behaviour is detected a more detailed model is automatically triggered in order to focus the operator attention on the suspected subsystem. (authors). 11 refs., 11 figs. (Atomindex citation 28:034516)

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