过程报警事件分组的Petri-FCM方法  

Petri-FCM approach to process alarm event grouping

作　　者：高丽霄[1] 李宏光[1] 

机构地区：[1]北京化工大学信息科学与技术学院,北京100029

出　　处：《计算机工程与设计》2013年第3期1018-1022,共5页Computer Engineering and Design

摘　　要：过程报警事件分组是智能报警管理中的一项核心内容,为了提高过程报警管理的质量,提出了一种过程报警事件分组的Petri-FCM方法。利用先验知识和历史数据建立过程对象报警事件演变的Petri网模型,根据此Petri网模型确立过程报警事件分组聚类中心并对数据进行同步处理,采用模糊C均值聚类的方法对过程报警事件进行分组。将该方法应用于TE过程,得到了满意的分组结果,表明了其有效性。Process alarm event grouping is recoginized as a core task of intelligent alarm management systems. A called Petri- / FCM approach to process alarm event grouping is explicitly introduced oriented to improve performance of process alarm manage- ment. Therein, a process alarm event evolutionary Petri-net model is initially established based on both process prior knowledge and historical data, which could be used to figure out the centers of process alarm events as well as to synchronize the candidate data. Subsequently, fuzzy C-means clustering methods are employed to group the process alarm events. The proposed approach is applied to TE process, leading to satisfactory results and showing benefits of the contribution.

关 键 词：过程报警事件分组 PETRI网 模糊C均值聚类 TE过程 聚类中心 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]