摘要
轨道交通通信系统中的集中告警系统不判定每个子系统故障数据是否正确,也不判定某一子系统故障与另一子系统故障的关联性,对系统故障的智能识别不起作用。为了实现对系统故障的智能识别,在对Apriori算法的最小支持度和最小置信度进行改进的基础上,实现了动态生成最小支持度和最小置信度;对子系统的故障和子系统间的故障进行分析,挖掘其相互关联关系,提前识别出系统故障。在地铁项目上的使用表明,改进的算法可以有效识别出子系统的部分故障,以及子系统间的故障关联关系。
The centralized alarm system widely used in the rail transit communication system does not determine whether the fault data of each subsystem is correct,nor does it determine whether the fault of one subsystem is related to another.It has no effect on the intelligent identification of system faults.In order to realize the intelligent identification of system faults,based on the improvement of the minimum support and minimum confidence of the Apriori algorithm,the minimum support and minimum confidence are dynamically generated,and the faults of subsystems or the faults between subsystems are analyzed.The research tries to mine their interrelationships to identify system faults in advance.Through the use of actual subway projects,the results show that the improved algorithm can effectively identify partial faults of the subsystems,and can also effectively identify the relationship between the faults of the subsystems.
作者
李文锋
闫涛
LI Wen-feng;YAN Tao(Jinling Institute of Technology, Nanjing 211169, China;China Railway Signal & Communication Shanghai Engineering Bureau Group Co. , Ltd. , Shanghai 200436, China)
出处
《金陵科技学院学报》
2021年第2期7-11,32,共6页
Journal of Jinling Institute of Technology
基金
金陵科技学院高层次人才科研启动基金(jit-b-202109)。
关键词
数据挖掘
智能
告警
故障关联
data mining
intelligence
alarm
fault interrelationships
