长江科学院院报 ›› 2022, Vol. 39 ›› Issue (8): 58-64.DOI: 10.11988/ckyyb.20210457

• 工程安全与灾害防治 • 上一篇    下一篇

基于关联规则的库岸边坡监测数据挖掘方法

陈波1,2, 詹明强1,2, 黄梓莘3   

  1. 1.河海大学 水利水电学院,南京 210098;
    2.河海大学 水文水资源与水利工程科学国家重点实验室,南京 210098;
    3.中国电建集团中南勘测设计研究院有限公司,长沙 410014
  • 收稿日期:2021-05-08 修回日期:2021-07-08 出版日期:2022-08-01 发布日期:2022-08-26
  • 作者简介:陈 波(1986-),男,浙江绍兴人,教授,博士,主要从事大坝安全监控研究。E-mail:chenbo@hhu.edu.cn
  • 基金资助:
    国家重点研发计划项目(2018YFC0407104);国家自然科学基金项目(52079049);中央高校基本科研业务费项目(B200202160)

Data Mining Method for Bank Slope Monitoring Based on Association Rules

CHEN Bo1,2, ZHAN Ming-qiang1,2, HUANG Zi-shen3   

  1. 1. College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China;
    2. State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering,Hohai University,Nanjing210098,China;
    3. Power China Zhongnan Engineering Corporation Limited,Changsha 410014,China
  • Received:2021-05-08 Revised:2021-07-08 Online:2022-08-01 Published:2022-08-26

摘要: 库岸边坡失稳灾害会对工程自身效益和周边生命财产安全造成巨大损失,而边坡运行监测资料记录了失稳灾害孕育的全过程信息。针对库岸边坡监测数据库的数据挖掘方法运行速度慢的问题,将FP-Growth关联规则算法引入到边坡安全监测数据挖掘中,通过FP-Growth关联规则算法开展边坡监测数据中的因果关联规则和空间关联规则挖掘工作,分别挖掘了边坡监测环境量和效应量之间的因果性、多测点效应量之间的关联性,从包含多测点多项目的边坡时空监测数据中提取了反映边坡运行性状的有效信息,并提供有效知识帮助。计算实例表明,FP-Growth关联规则算法实现过程简单,挖掘结果可靠,为类似库岸边坡的监测数据挖掘提供了一条良好的思路。

关键词: 库岸边坡监测, 数据挖掘, FP-Growth算法, 因果关联规则, 空间关联规则

Abstract: The monitoring data of slope records the whole process information of slope instability disaster.The data mining of bank slope monitoring database runs slowly.FP-Growth association rule algorithm is introduced into slope safety monitoring data mining.The causal association rule and spatial association rule in slope monitoring data are mined by FP-Growth association rule algorithm.The causality between environmental variables and affected variables of slope monitoring and the correlation among affected variables of multiple measuring points are mined respectively.Effective information reflecting slope operation characteristics is extracted from slope spatio-temporal monitoring data containing multi-measuring points and multi-items.Calculation example shows that FP-Growth association rule algorithm is simple in implementation and reliable in mining results,hence providing a good approach for monitoring data mining of similar reservoir bank slopes.

Key words: reservoir bank slope monitoring, data mining, FP-Growth algorithm, cause-effect association rule, spatial association rule

中图分类号: