首页 | 官方网站   微博 | 高级检索  
     

基于FTP-图的时间序列研究及在设备监控中应用
引用本文:闫伟,童祯恭,谢宗法. 基于FTP-图的时间序列研究及在设备监控中应用[J]. 化工自动化及仪表, 2007, 34(2): 55-58,72
作者姓名:闫伟  童祯恭  谢宗法
作者单位:山东大学,能源与动力工程学院,济南,250061;华东交通大学,土木建筑学院,南昌,330013;山东大学,机械工程学院,济南,250061
基金项目:国家高技术研究发展计划(863计划)
摘    要:采用FTP-图(频繁时序图)的时间序列方法对流程企业中的实际运行数据进行数据挖掘,首先给出基于拓扑关系的时间序列的定义,分析FTP-图连接、候选TP-图剪枝和采用哈希树的候选TP-图计数方法,然后采用模糊理论找出偏离常规运行状态的参数点并模糊化,采用时间窗对参数离散处理,得到时间序列数据库.根据Apriori算法原理,提出基于FTP-图的时间序列分析算法并编程实现,起到对设备故障预警和生产优化的决策作用.

关 键 词:数据挖掘  时间序列分析  FTP-图  Apriori算法  流程企业
文章编号:1000-3932(2007)02-0055-04
修稿时间:2007-02-14

Study and Application of Time Sequential Pattern Based on FTP-Graph to Equipment's Fault Diagnosis
YAN Wei,TONG Zhen-gong,XIE Zong-fa. Study and Application of Time Sequential Pattern Based on FTP-Graph to Equipment's Fault Diagnosis[J]. Control and Instruments In Chemical Industry, 2007, 34(2): 55-58,72
Authors:YAN Wei  TONG Zhen-gong  XIE Zong-fa
Abstract:To monitor and optimize process industry's production,the large history database are analyzed by time sequential pattern based on FTP-graph(frequent time pattern graph).Firstly,the topological relations of the time sequential patterns are expounded.FTP-Graph algorithm is studied by joining frequent temporal graphs,subtracting candidate temporal graph and scanning process instances in database.The exceptional equipment parameters are found and fuzzed by fuzzy theory.The parameters in real-time database are scattered by time-window approach.The time sequential database is given.According to Apriori algorithm,the FTP-Graph algorithm is studied and implemented in process industry.
Keywords:data mining   time sequential pattern   FTP-graph   Apriori algorithm   process industry
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号