View in EDS HTML Full Text PDF Full Text

Anomaly detection method for TBM construction based on improved VMD-XGBoost-BILSTM combined model.

Saved in:
Bibliographic Details
Title: Anomaly detection method for TBM construction based on improved VMD-XGBoost-BILSTM combined model.
Authors: Lu, Zhipeng1 (AUTHOR), Shi, Kebin1 (AUTHOR) xndsg@sina.com
Source: Earth Science Informatics. Dec2023, Vol. 16 Issue 4, p4273-4284. 12p.
Subject Terms: *Intrusion detection systems (Computer security), *Anomaly detection (Computer security), *Metaheuristic algorithms, *Water tunnels, *Traffic monitoring, *Rock bursts, *Boring & drilling (Earth & rocks), *System safety
Geographic Terms: Xinjiang Uygur Zizhiqu (China), China
Abstract: TBM method construction is an important accompanying construction form vigorously developed in China, and its anomaly detection is an important link to provide a basis for system operation and maintenance decision-making. The abnormal operation state of TBM caused by different surrounding rock geology, jamming or rock burst will directly affect the tunneling speed and ability, and then affect the system safety and construction progress. In this paper, a TBM construction anomaly detection method based on historical tunneling speed and other construction monitoring data is proposed. Firstly, the preprocessing steps based on outlier removal and correlation analysis are used to remove the noise in the original data and select the best features. Secondly, the variational mode decomposition is used to decompose the data into multiple modal components to extract the periodic and aperiodic features of TBM tunneling. Furthermore, an improved VMD-XGBoost-BILSTM combination model is constructed, and the characteristics of combination weighting, attention mechanism and improved whale optimization algorithm are used to realize the normal prediction of accurate and stable tunneling speed. Finally, by comparing with the actual measured values, the set rules are used to judge the anomalies. The experiments are carried out on the actual mining data of YE long-distance water conveyance tunnel in Xinjiang. The results show that the method proposed in this paper improves the RMSE by more than 20% compared with the single BILSTM or XGBoost model. The attention mechanism and IWOA algorithm bring 5.53% and 10.73% results improvement respectively, which can achieve the effect of early warning of different geological information changes. [ABSTRACT FROM AUTHOR]
Database: Energy & Power Source
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: enr
DbLabel: Energy & Power Source
An: 174096721
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Anomaly detection method for TBM construction based on improved VMD-XGBoost-BILSTM combined model.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Lu%2C+Zhipeng%22">Lu, Zhipeng</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Shi%2C+Kebin%22">Shi, Kebin</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> xndsg@sina.com</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Earth+Science+Informatics%22">Earth Science Informatics</searchLink>. Dec2023, Vol. 16 Issue 4, p4273-4284. 12p.
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: *<searchLink fieldCode="DE" term="%22Intrusion+detection+systems+%28Computer+security%29%22">Intrusion detection systems (Computer security)</searchLink><br />*<searchLink fieldCode="DE" term="%22Anomaly+detection+%28Computer+security%29%22">Anomaly detection (Computer security)</searchLink><br />*<searchLink fieldCode="DE" term="%22Metaheuristic+algorithms%22">Metaheuristic algorithms</searchLink><br />*<searchLink fieldCode="DE" term="%22Water+tunnels%22">Water tunnels</searchLink><br />*<searchLink fieldCode="DE" term="%22Traffic+monitoring%22">Traffic monitoring</searchLink><br />*<searchLink fieldCode="DE" term="%22Rock+bursts%22">Rock bursts</searchLink><br />*<searchLink fieldCode="DE" term="%22Boring+%26+drilling+%28Earth+%26+rocks%29%22">Boring & drilling (Earth & rocks)</searchLink><br />*<searchLink fieldCode="DE" term="%22System+safety%22">System safety</searchLink>
– Name: SubjectGeographic
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Xinjiang+Uygur+Zizhiqu+%28China%29%22">Xinjiang Uygur Zizhiqu (China)</searchLink><br /><searchLink fieldCode="DE" term="%22China%22">China</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: TBM method construction is an important accompanying construction form vigorously developed in China, and its anomaly detection is an important link to provide a basis for system operation and maintenance decision-making. The abnormal operation state of TBM caused by different surrounding rock geology, jamming or rock burst will directly affect the tunneling speed and ability, and then affect the system safety and construction progress. In this paper, a TBM construction anomaly detection method based on historical tunneling speed and other construction monitoring data is proposed. Firstly, the preprocessing steps based on outlier removal and correlation analysis are used to remove the noise in the original data and select the best features. Secondly, the variational mode decomposition is used to decompose the data into multiple modal components to extract the periodic and aperiodic features of TBM tunneling. Furthermore, an improved VMD-XGBoost-BILSTM combination model is constructed, and the characteristics of combination weighting, attention mechanism and improved whale optimization algorithm are used to realize the normal prediction of accurate and stable tunneling speed. Finally, by comparing with the actual measured values, the set rules are used to judge the anomalies. The experiments are carried out on the actual mining data of YE long-distance water conveyance tunnel in Xinjiang. The results show that the method proposed in this paper improves the RMSE by more than 20% compared with the single BILSTM or XGBoost model. The attention mechanism and IWOA algorithm bring 5.53% and 10.73% results improvement respectively, which can achieve the effect of early warning of different geological information changes. [ABSTRACT FROM AUTHOR]
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=174096721
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s12145-023-01101-9
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 12
        StartPage: 4273
    Subjects:
      – SubjectFull: Intrusion detection systems (Computer security)
        Type: general
      – SubjectFull: Anomaly detection (Computer security)
        Type: general
      – SubjectFull: Metaheuristic algorithms
        Type: general
      – SubjectFull: Water tunnels
        Type: general
      – SubjectFull: Traffic monitoring
        Type: general
      – SubjectFull: Rock bursts
        Type: general
      – SubjectFull: Boring & drilling (Earth & rocks)
        Type: general
      – SubjectFull: System safety
        Type: general
      – SubjectFull: Xinjiang Uygur Zizhiqu (China)
        Type: general
      – SubjectFull: China
        Type: general
    Titles:
      – TitleFull: Anomaly detection method for TBM construction based on improved VMD-XGBoost-BILSTM combined model.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Lu, Zhipeng
      – PersonEntity:
          Name:
            NameFull: Shi, Kebin
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 12
              Text: Dec2023
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-print
              Value: 18650473
          Numbering:
            – Type: volume
              Value: 16
            – Type: issue
              Value: 4
          Titles:
            – TitleFull: Earth Science Informatics
              Type: main
ResultId 1