Abstract:

Reasonable and effective visualization monitoring techniques and tools can help operators understand the alarm information and make timely responses. Aiming at the shortcomings and deficiencies of existing visualization techniques, such as the insufficient resources utilization, unclear alarm priority and incomplete alarm root-cause analysis, four novel visualization tools are proposed, namely information fusion based interpretative structural modelling (static and dynamic), hierarchical high density alarm plot, hierarchical priority color plot and performance level trend plot, to achieve the goals of process hierarchical model construction, alarm root-cause analysis, nuisance alarms identification, alarm prioritization and alarm system performance periodic assessment separately. As a case example, Tennessee Eastman process is utilized to show the practicality and effectiveness of the proposed visualization techniques and tools. They can not only provide the overview of original alarm information but also quickly identify the alarm root-cause, key alarms, nuisance alarms and alarm system performance level. The efficient monitoring has been realized and alarm floods has been resolved to a certain extent.

Key words: alarm floods, visualization monitoring, root-cause analysis, alarm prioritization, process industry

摘要：

合理有效的可视化监控技术及工具有助于操作员及时理解报警信息并采取响应措施。针对现有可视化技术存在的缺点及不足, 如资源利用不充分、报警等级划分不明确、报警根源分析不彻底等, 构建了4种新型可视化工具:基于信息融合的解释结构模型(静态和动态)、层次高密度报警图、层次优先级色彩图、性能水平趋势图, 分别实现了过程递阶模型建立、报警根源分析、滋扰报警识别、报警优先级划分、报警系统性能常规评估等目的。以TE仿真模型为例, 阐明了上述可视化技术及工具的实用性和有效性, 不仅可以展示报警全貌原始信息, 还可快速识别报警根源、关键报警、滋扰报警以及报警系统性能水平, 实现了高效监控,从一定程度上解决了报警泛滥问题。

关键词: 报警泛滥, 可视化监控, 根源分析, 报警优先级划分, 过程工业