智能报警管理若干研究问题

Research topics of intelligent alarm management

报警是及时发现生产过程中异常情况的直接手段,因此科学、高效、准确的智能报警管理对提高生产过程安全性和可操作性至关重要,也是进一步进行优化、控制和故障诊断的基础。本文就智能报警管理这一研究领域进行综述,首先从工业需求上论述报警管理的重要意义,说明报警对于过程安全的影响,指出数量过多和难以辨别是主要难题,并将现实情况与国际标准相对比;然后概述了报警管理的概念和内涵,介绍报警系统改造的全生命周期。其次总结该领域的主要研究问题,并针对单个报警和多个报警的设计展开讨论,揭示误报与漏报、检测速度与鲁棒性的矛盾,并给出常用的解决方法。针对单个报警,通过滤波、延迟和死区,对过程时间序列进行处理,实现报警的快速和准确。针对多个报警,通过多元统计分析全面表征多个变量之间的关系,利用相关性分析和传递熵分析来识别因果报警。最后通过一个实例来说明智能报警管理的过程。

Alarming is the direct way of promptly detecting abnormal states in production processes, hence the scientific, efficient and correct intelligent alarm management is crucial for improving the safety and operability of production processes, and is also the foundation of further optimization, control and fault diagnosis. In this paper, the area of intelligent alarm management is overviewed. First the significance of the alarm management is presented in the perspective of industrial requirements followed by a brief summary of its concepts and contents. Then the major research topics in this area are summarized and the design problems concerning single alarms and multiple alarms are discussed, discovering the trade-off between false alarms and missed alarms, along with that between detection speed and robustness. In addition, some common solutions are proposed, including filtering, time delay and deadband for single alarms, stochastic analysis for multiple alarms, and identification of chattering alarms and cause-effect alarms. Finally the procedure of intelligent alarm management is illustrated via an example. Finally the procedure of intelligent alarm management is illustrated via an example.