数据与计算发展前沿 ›› 2020, Vol. 2 ›› Issue (5): 110-121.

doi: 10.11871/jfdc.issn.2096-742X.2020.05.011

• 技术与应用 • 上一篇    

一种工业报警相关性数据挖掘算法

杨润佳1,*(),刘泽三2()   

  1. 1.国家能源集团,秦皇岛发电有限责任公司,发展策划处,河北 秦皇岛 066000
    2.国网信息通信产业集团信通研究院,北京 100000
  • 收稿日期:2020-07-23 出版日期:2020-10-20 发布日期:2020-10-30
  • 通讯作者: 杨润佳
  • 作者简介:杨润佳,国家能源集团秦皇岛发电有限责任公司,双学士,发展策划处高级主管。主要工作经历及方向为电力信息化建设及电力营销。
    本文中负责收集某电厂报警数据,并完成数据整理;同时,针对电厂报警相关性的应用特点而对凝聚层次聚类算法进行改进,并设计了相关性度量方法。
    Yang Runjia, a double bachelor degree holder, is the senior director of the Development Planning Department at CHN ENERGY Qinhuangdao Power Generation Co., LTD. His main work experience and directions are for electric power enterprise information construction and electric power marketing.
    In this paper, he is responsible for preparing enough alarm data, and improving the traditional agglomerative hierarchical clustering algorithm for alarm correlation analysis in electric power generation plant.
    E-mail: 13811330386@163.com|刘泽三,国网信息通信产业集团信通研究院,研发工程师,硕士,主要研究方向为SOA、云计算、大数据、电力信息系统。
    本文中负责报警在时间维度上的矢量相关性度量方法的设计工作。
    Liu Zesan, Postgraduates, is an R & D Engineer of State Grid Information & Telecommunication Group Co., Beijing, China. His recent research interest areas are SOA, cloud computing, big data, and power information systems.
    In this paper, he is responsible for designing a vector correlation concept and using conditional probability to make the alarm correlation among different tags comparable.
    E-mail: liuzesan@sgitg.sgcc.com.cn

An Data Mining Algorithm for Analyzing Industrial Alarm Data Correlation

Yang Runjia1,*(),Liu Zesan2()   

  1. 1. Development planning division, CHN ENERGY Qinhuangdao power generation co. LTD, Qinhuangdao, Hebei 066000, China
    2. Research Institute of Information & Telecommunication, State Grid Information & Telecommunication Group Co. Ltd., Beijing 100000, China;
  • Received:2020-07-23 Online:2020-10-20 Published:2020-10-30
  • Contact: Yang Runjia

摘要:

【目的】 报警在工业系统安全管理中具有重要作用。报警之间有些是孤立的,有些相互之间存在着某种关联,报警间相关性的研究是报警预测和防范的重要手段。 【方法】 本文针对某电厂大量时序报警数据,提出了一种基于权重的多态时序通用型相似度算法,应用该相似度算法能提高传统凝聚层次聚类算法在报警数据上聚类的精度;在报警聚类的基础上,以时间维度作为矢量相关,阐述了报警之间相互影响的方向性,提出以条件概率形式量化分析报警矢量相关性,使得不同点位之间报警相关性具有可比性;在此基础上,设计了针对大批量报警数据的挖掘报警矢量相关性的挖掘算法。 【结果】 以某电厂报警数据为实验数据样本测试,该算法能有效地挖掘出报警间的关联关系,可预测相关报警,指导生产。 【结论】 该算法打破了传统报警相关性研究方法的部分局限,在挖掘序列报警和非序列报警、规整报警和非规整报警方面都具有同等效果。为了能更好地展示矢量相关性,本文设计了一种二维矩阵的展示方式。

关键词: 报警, 聚类, 相似度算法, 矢量相关

Abstract:

[Objective] Alarms are important in industrial safety management. The technique to capture the correlation information of alarm variables, especially from historical alarm data, is very beneficial for risk prediction and prevention. But the task is very difficult because some alarms are independent while others are coupled mutually. [Methods] In this paper, a general weight-based multi-state sequential algorithm for correlation analysis is applied for alarm data mining to improve the validity and accuracy of alarm clustering when combined with the traditional agglomerative hierarchical clustering algorithm. To measure the directional relations between alarm variables, this paper proposes a vector correlation concept and use conditional probability modeling to make the alarm correlation among different tags comparable. [Results] The proposed data mining algorithm is shown to be able to find out the vector correlation of alarm variables effectively and correctly when applied in the analysis of power plant alarm data. [Conclusions] Some limitations of the traditional research on alarm variables correlation analysis are solved by the proposed method, which works well on all sequential, non-sequential, regular and irregular alarms. Furthermore, a two-dimensional matrix is used to visualize the vector correlation of alarm variables intuitively and visually.

Key words: alarm, clustering, similarity algorithm, vector correlation