Optical computing device diagnostics and treatment

Patent metadata

Field	Value
Publication number	US-9612361-B2
Application number	US-201414780780-A
Country	US
Kind code	B2
Filing date	Dec 12, 2014
Priority date	Dec 12, 2014
Publication date	Apr 4, 2017
Grant date	Apr 4, 2017

Abstract

An example method includes performing validation testing on a tool using a plurality of reference fluids, the tool having a calibrated optical sensor installed therein that includes one or more optical elements. One or more tool sensor responses from the calibrated optical sensor may be obtained and pre-processed, and the one or more tool sensor responses may be compared with calibrated optical sensor responses derived from the calibrated optical sensor during calibration and thereby detecting one or more optical sensor anomalies. The one or more optical sensor anomalies may be evaluated through performance analysis with one or more candidate models, and an alternative candidate model may be selected to mitigate the one or more optical sensor anomalies. One or more remedial options may be pursued when the alternative candidate model fails to mitigate the one or more optical sensor anomalies.

First claim

What is claimed is:

1. A method, comprising:
performing validation testing on a tool having a calibrated optical sensor that includes one or more optical elements, wherein performing the validation testing comprises:
circulating a plurality of reference fluids through the tool at one or more calibration set points used to calibrate the calibrated optical sensor during an optical sensor calibration; and
collecting optical sensor responses from an optical channel of each optical element at the one or more calibration set points and thereby obtaining one or more tool sensor responses;

obtaining and pre-processing the one or more tool sensor responses;
comparing the one or more tool sensor responses with calibrated optical sensor responses derived from the calibrated optical sensor during the optical sensor calibration, and detecting one or more optical sensor anomalies in response to a difference between the one or more tool sensor responses and the calibrated optical sensor responses;
evaluating the one or more optical sensor anomalies through performance analysis with one or more candidate models; and
selecting an alternative candidate model to mitigate the one or more optical sensor anomalies.

2. The method of claim 1 , wherein the one or more calibration set points comprise a same or reduced number of specific temperature and pressure set used to calibrate the calibrated optical sensor.

3. The method of claim 1 , wherein obtaining and pre-processing one or more tool responses from the calibrated optical sensor comprises:
directing light to interact with the plurality of reference fluids to produce interacted light;
transmitting the interacted light as energy toward the one or more optical elements of the calibrated optical sensor;
generating detector responses over the one or more optical elements as the one or more tool sensor responses, wherein the one or more tool sensor responses are baseline corrected and normalized to provide normalized tool sensor responses; and
converting the normalized tool sensor responses from a tool data space to a calibration data space.

4. The method of claim 1 wherein comparing the one or more tool sensor responses with calibrated optical sensor responses comprises:
calculating a normalized signal response from an optical channel of each optical element during the optical sensor calibration of the calibrated optical sensor to obtain normalized optical sensor signal responses;
calculating a normalized signal response from an optical channel of each optical element during validation testing of the tool to obtain normalized tool sensor responses;
averaging the normalized optical sensor signal responses under one or more temperature and pressure calibration set points;
averaging the normalized tool sensor responses under the one or more temperature and pressure calibration set points; and
computing a percentage difference between the normalized optical sensor signal responses and the normalized tool sensor responses with respect to a full dynamic range of optical sensor calibration data on each channel.

5. The method of claim 1 , wherein evaluating the one or more optical sensor anomalies through performance analysis with candidate models comprises:
calculating a normalized signal response from an optical channel of each optical element during validation testing of the tool to obtain normalized tool sensor responses;
providing the normalized tool sensor responses into at least one instrument standardization model;
using the at least one instrument standardization model transformed tool sensor responses as candidate inputs of pre-calibrated fluid characterization models;
predicting fluid properties of the plurality of reference fluids with a different number of inputs; and
determining if one or more remedial options are needed to mitigate the one or more optical sensor anomalies.

6. The method of claim 5 , wherein determining if the one or more remedial options are needed comprises comparing relative prediction variation of tool testing with a variation of optical sensor calibration and optical pressure-volume-temperature (PVT) calibration for each fluid property using the different number of inputs.

7. The method of claim 1 , further comprising pursuing one or more remedial options when the alternative candidate model fails to mitigate the one or more optical sensor anomalies.

8. The method of claim 7 , wherein pursuing the one or more remedial options comprises selecting a remedial option that requires replacement of one or more of the one or more optical elements.

9. The method of claim 8 , further comprising:
substituting only anomalous optical elements in the calibrated optical sensor; and
performing a new validation testing on the tool using the plurality of reference fluids.

10. The method of claim 8 , wherein the one or more optical elements are disposed on one or more sensor wheels arranged within the calibrated optical sensor, the method further comprising substituting at least one of the one or more sensor wheels in the tool.

11. The method of claim 10 , wherein substituting the at least one of the one or more sensor wheels comprises:
removing the at least one of the one or more sensor wheel from the tool;
replacing at least one of the one or more sensor wheels with a newly calibrated sensor wheel and an associated light source and components;
performing a new validation testing on the tool using the plurality of reference fluids.

12. The method of claim 7 , wherein pursuing the one or more remedial options comprises selecting a remedial option that does not require replacement of one or more of the one or more optical elements.

13. The method of claim 12 , further comprising rebuilding the one or more candidate models using less affected optical channels as inputs.

14. The method of claim 13 , wherein rebuilding the one or more candidate models using less affected optical channels as inputs comprises:
removing anomalous optical channels; and
reconstructing a candidate model base using other available optical channels through at least one of backward stepwise input selection or forward stepwise input selection, while maintaining instrument standardization models unchanged.

15. The method of claim 12 , further comprising rebuilding instrument standardization models.

16. The method of claim 15 , wherein rebuilding the instrument standardization models comprises using as model inputs at least one of the one or more tool sensor responses and the calibrated optical sensor responses with questionable optical channels removed, while maintaining fluid predictive models unchanged.

17. The method of claim 12 , further comprising developing a new candidate model to compensate for a difference between the one or more tool sensor responses and the calibrated optical sensor responses derived from the calibrated optical sensor during optical sensor calibration.

18. The method of claim 17 , wherein developing the new candidate model comprises generating a linear or non-linear mapping function on optical channels where relative differences between the one or more tool sensor responses and the calibrated optical sensor responses are large, while maintaining sensor-based instrument standardization models and fluid predictive models unchanged.

19. The method of claim 1 , further comprising:
running the tool into a wellbore as part of a drill string; and
drilling a portion of the wellbore with a drill bit coupled to an end of the drill string.

20. The method

Assignees

• Halliburton Energy Services Inc

Inventors

• Chen Dingding
• Perkins David L
• Soltmann William
• Gascooke Darren
• Shen Jing

Classifications

• G01V13/00 Primary

  Manufacturing, calibrating, cleaning, or repairing instruments or devices covered by groups G01V1/00 – G01V11/00 · CPC title

• G01V8/02

  Prospecting · CPC title

• G01V8/00

  Prospecting or detecting by optical means · CPC title

• G01N21/274 Primary

  Calibration, base line adjustment, drift correction · CPC title

• E21B49/082

  Wire-line fluid samplers (E21B49/083 takes precedence) · CPC title