Dynamic Detection of Overburden Thickness Based on Hilbert-Huang Transform
CSTR:
Author:
Affiliation:

Clc Number:

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    Accurately revealing the thickness of soil and its changes in the process of dynamic settlement is of great significance for reasonably determining the thickness of overburden and scientifically evaluating the stability of reclaimed soil. Through field tests, combined with ground penetrating radar data, signal processing was carried out based on Hilbert-Huang transform (HHT), and the time domain range of overburden was determined according to Hilbert spectrum. Extracting the time instantaneous frequency information of multiple intrinsic mode functions (IMF) obtained by HHT, the relationship model between IMF2 component and relative dielectric constant was established, and the detection of overburden thickness was realized. The established relationship model was further applied to the detection of overburden thickness on time series, so as to realize the dynamic detection of overburden thickness. The results showed that the subsidence of overburden could cause the change of dielectric constant difference between layered media, and then cause the change of amplitude at the interface. Based on HHT method, the time-domain position of the overburden could be better obtained. In different reclamation times, the time-domain range of the overburden was in dynamic change. There was a high correlation between the average instantaneous frequency and the relative dielectric constant, and the R2 modeled by polynomial regression was 0.8870. The combined time-frequency analysis considering the amplitude and instantaneous frequency changes in the time domain can effectively detect the thickness of the overburden layer;with the increase of the thickness of overburden layer and the passage of time, the detection accuracy was decreased slightly, and the average relative error was 3.65%.

    Reference
    Related
    Cited by
Get Citation
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:June 06,2022
  • Revised:
  • Adopted:
  • Online: March 10,2023
  • Published:
Article QR Code