.png){kind=icon}
.png){kind=icon}
Return
Experimental detection method of lost circulation based on multivariate optimised signal processing
.png){kind=icon}
PRE
AI
DOI：10.1016/j.geoen.2025.214303.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Accurately locating the leakage location can reduce the loss caused by lost circulation, and a method of detecting the leakage location using transient pressure wave is proposed for leakage location. Firstly, the pressure wave signal is denoised using the improved adaptive noise complete ensemble empirical modal decomposition (ICEEMDAN), the joint classification of preferred components by Marginal Distance (MD) and cumulative mean (MASM), the improved wavelet thresholding method, and a multivariate optimisation signal denoising method based on ICEEMDAN-MD-MASM-IWT is proposed; compared with the other four denoising Secondly, the signal features are analysed using the neighbourhood difference method, and the leakage layer is located according to the peak time of the leakage features, which avoids the problems of too strong dependence on time-domain features, difficult time-frequency correspondence, and inconspicuous feature peaks in the traditional lost circulation detection methods; compared with the other traditional lost circulation detection methods, this method has the smallest average error in location. , which is reduced by 2.03 % compared to the time domain analysis method. Further discussion, multiple sets of experiments were conducted for different borehole sizes, and the error range was 3.11 %-7.98 %. The results show that the lost circulation detection method can effectively identify the location of lost circulation.
Keywords:
Lost circulation detection
Pressure wave
Signal denoising
Neighbourhood difference method
Lost circulation location
Journal
G
IF:
4.6
Papers:
191
Citations:
0