US 12,248,114 B2
Correction of nuclear magnetic resonance data in high vibration environments
Rebecca Jachmann, Houston, TX (US); and Jie Yang, Houston, TX (US)
Assigned to Halliburton Energy Services, Inc., Houston, TX (US)
Filed by Halliburton Energy Services, Inc., Houston, TX (US)
Filed on Jul. 20, 2021, as Appl. No. 17/380,315.
Claims priority of provisional application 63/185,979, filed on May 7, 2021.
Prior Publication US 2022/0365242 A1, Nov. 17, 2022
Int. Cl. G01V 3/34 (2006.01); G01R 33/44 (2006.01); G01R 33/50 (2006.01); G01R 33/561 (2006.01); G01V 3/32 (2006.01); G01N 24/08 (2006.01)
CPC G01V 3/34 (2013.01) [G01R 33/448 (2013.01); G01R 33/50 (2013.01); G01R 33/5615 (2013.01); G01V 3/32 (2013.01); G01N 24/081 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method of generating a log of spectra using nuclear magnetic resonance, the method comprising:
acquiring data signals from operation of a nuclear magnetic resonance logging (NMR) tool in a borehole;
processing the data signals to generate processed data;
selecting, from the processed data, a window of datasets acquired consecutively in time, the window of data comprising multiple datasets each having data points, wherein the multiple datasets comprise overlapping data points;
wherein each data point has a corresponding index count representing a position of that data point from a beginning of the dataset to an end of the dataset;
from each dataset, selecting data points having a same index count across the datasets;
comparing the selected data points to one another;
from the compared data points, selecting a minimal data point;
for the index count being compared, generating filtered datasets by replacing the data on the index count with the minimal data point, wherein the index count labels the position of every dataset from the multiple datasets; and
outputting the filtered datasets as a log.