In-situ evaluation of gauges
Methods for evaluating sensor data to predict when the sensor should be recalibrated are described. The methods include a model that utilizes current wellbore data as input for the recalibration prediction.
1. A method for determining when to calibrate a wellbore sensor comprising:
a) modeling drift thresholds for one or more wellbore sensors, wherein the modeling of the drift thresholds is performed by combining field data and calibration data;
b) collecting field data from the one or more wellbore sensors;
c) comparing the collected field data to the drift thresholds; and
d) recalibrating the one or more wellbore sensors if the collected field data exceeds the drift thresholds.
2. The method of claim 1 , wherein the field data comprises currently collected field data and historical field data.
3. The method of claim 1 , wherein the calibration data comprises laboratory testing data.
4. The method of claim 1 , wherein the one or more wellbore sensors are conveyed as part of a wellbore tool.
5. The method of claim 4 , wherein the wellbore tool is a quartz gauge.
6. The method of claim 4 , wherein the wellbore tool is a testing, wireline, or drilling tool.
7. The method of claim 1 , wherein the field data comprises data based on job environment and health conditions.
8. The method of claim 1 , wherein the field data comprises operational measurements.
9. The method of claim 1 , wherein the field data comprises pressure measurements.
10. The method of claim 1 , wherein the field data comprises temperature measurements.
11. The method of claim 1 , further comprising processing the field data collected from the one or more wellbore sensors for comparison to the drift thresholds.
12. The method of claim 1 , further comprising providing an alert if the collected field data exceeds the drift thresholds.
13. The method of claim 12 , wherein the alert is an automatic email.
14. The method of claim 1 , further comprising updating the drift thresholds modeling with the collected field data.
15. The method of claim 1 , wherein said modelling drift thresholds uses machine learning techniques.
16. The method of claim 15 , wherein said machine learning techniques are selected from a group comprising gradient boosted trees, random forest, and/or support vector machine.
17. A method for determining the drift status of a wellbore sensor comprising:
a) creating a dynamic model defining drift thresholds for one or more wellbore sensors, the dynamic model created by combining field data collected by the one or more wellbore sensors and laboratory calibration data for the one or more wellbore sensors;
b) inputting field data collected by the one or more wellbore sensors into the dynamic model;
c) comparing the input field data to the defined drift thresholds for the one or more wellbore sensors;
d) updating the dynamic model with the input field data; and
e) calibrating the one or more wellbore sensors if the input field data exceeds the defined drift thresholds.
18. The method of claim 17 , further comprising updating the dynamic model based on the calibrating of the one or more wellbore sensors.
19. A method for determining when to calibrate a wellbore sensor comprising:
a) modeling drift thresholds for one or more wellbore sensors;
b) collecting field data from the one or more wellbore sensors;
c) comparing the collected field data to the drift thresholds;
d) recalibrating the one or more wellbore sensors if the collected field data exceeds the drift thresholds; and
f) updating the drift thresholds modeling with the collected field data.
20. The method of claim 19 , wherein said modelling drift thresholds uses machine learning techniques.
21. The method of claim 20 , wherein said machine learning techniques are selected from a group comprising gradient boosted trees, random forest, and/or support vector machine.