Method for ruggedizing integrated computational elements for analyte detection in the oil and gas industry
A method for ruggedizing an ICE design, fabrication and application with neural networks as disclosed herein includes selecting a database for integrated computational element (ICE) optimization is provided. The method includes adjusting a plurality of ICE operational parameters according to an environmental factor recorded in the database and simulating environmentally compensated calibration inputs. The method includes modifying a plurality of ICE structure parameters to obtain an ICE candidate structure having improved performance according to a first algorithm applied to the database and validating the ICE candidate structure with an alternative algorithm applied to the database. Further, the method includes determining a plurality of manufacturing ICEs based on the validation with the first algorithm and the alternative algorithm, and fabricating one of the plurality of manufacturing ICEs. A method for determining a fluid characteristic using a calibrated ICE fabricated as above and supplemental elements is also provided.
1. A method, comprising:
selecting a database for integrated computational element (ICE) design;
adjusting a plurality of ICE operational parameters according to an environmental factor recorded in the database;
simulating environmentally compensated calibration inputs;
modifying a plurality of ICE structure parameters to obtain an ICE candidate structure having improved performance according to a first algorithm applied to the database;
validating the ICE candidate structure with an alternative algorithm applied to the database;
determining a plurality of manufacturing ICEs based on the validation with the first algorithm and the alternative algorithm; and
fabricating one of the plurality of manufacturing ICEs.
2. The method of claim 1 , wherein modifying the plurality of ICE structure parameters includes at least one of modifying a number of layers of material in the ICE candidate structure or modifying a thickness of an existing layer of material in the ICE candidate structure.
3. The method of claim 1 , wherein validating the ICE candidate structure includes estimating at least one of a standard error of prediction (SEP), a standard error of calibration (SEC), a variance, a sensitivity and a specificity value associated with the ICE candidate structure.
4. The method of claim 1 , wherein validating the ICE candidate structure includes using an algorithm selected from one of a neural network model, a linear model, or a polynomial model relating tool input to the fluid characteristic of interest.
5. The method of claim 1 , wherein selecting the database includes selecting an optical-PVT spectroscopy and analyte database including a plurality of optical-PVT data from light and medium oils, gas condensates, gas, heavy oils, water and multiphase mixtures.
6. The method of claim 1 , wherein adjusting the plurality of ICE operational parameters according to an environmental factor includes a temperature correction of high and low refraction index constants of ICE materials in simulating an ICE response complying with an environmental condition of at least one of a fluid spectroscopy and an analyte in the database.
7. The method of claim 1 , wherein adjusting the plurality of ICE operational parameters according to an environmental factor includes a temperature correction of a substrate material in simulating an ICE response complying with an environmental condition of at least one of a fluid spectroscopy and an analyte in the database.
8. The method of claim 1 , wherein adjusting the plurality of ICE operational parameters according to environmental factors further includes correcting a temperature effect on an optical response of at least one of an optical filter, a window and a detector related to an ICE response.
9. The method of claim 1 , wherein adjusting the ICE operational parameters according to an environmental factor includes measuring fabricated ICE transmittance spectra under different temperature settings.
10. The method of claim 1 , wherein the validating the ICE candidate structure with an alternative algorithm includes using at least one of a neural network algorithm, a linear algorithm, and a polynomial data mapping algorithm.
11. The method of claim 1 , wherein fabricating one of the plurality of manufacturing ICEs further includes:
measuring a thickness of a deposited layer;
incorporating the measured layer thickness with a plurality of target film thicknesses on subsequent layers of the one of the plurality of manufacturing ICEs in a fluid prediction model to predict a performance of an ICE being fabricated; and
adjusting a thickness on a subsequent layer of the manufacturing ICE according to the predicted performance of the ICE being fabricated.
12. A non-transitory, computer readable medium storing commands which, when executed by a processor in a controller, cause the controller to perform a method comprising:
selecting a database for integrated computational element (ICE) design;
adjusting a plurality of ICE operational parameters according to an environmental factor recorded in the database;
simulating environmentally compensated calibration inputs;
modifying a plurality of ICE structure parameters to obtain an ICE candidate structure having improved performance according to a first algorithm applied to the database;
validating the ICE candidate structure with an alternative algorithm applied to the database;
determining a plurality of manufacturing ICEs based on the validation with the first algorithm and the alternative algorithm; and
fabricating one of the plurality of manufacturing ICEs.
13. The non-transitory, computer-readable medium of claim 12 , wherein the commands for modifying a plurality of ICE structure parameters include commands for at least one of modifying a number of layers of material in the ICE structure or modifying a thickness of an existing layer of material in the ICE structure.
14. The non-transitory, computer-readable medium of claim 12 , wherein commands for validating the ICE candidate structure include commands for estimating at least one of a standard error of prediction (SEP) a standard error of calibration (SEC), a variance, a sensitivity and a specificity value associated with the ICE candidate structure.
15. The non-transitory, computer-readable medium of claim 12 , wherein commands for validating the ICE candidate structure include commands for using an algorithm selected from one of a neural network model, a linear model, or a polynomial model relating tool input to the fluid characteristic of interest.
16. The non-transitory, computer-readable medium of claim 12 , wherein commands for selecting a database include commands for selecting an optical-Pressure, Volume, and Temperature (PVT) spectroscopy and analyte database including a plurality of optical-PVT data from light and medium oils, gas condensates, gas, heavy oils, water and multiphase mixtures.
17. The non-transitory, computer-readable medium of claim 12 , wherein commands for adjusting a plurality of ICE operational parameters according to an environmental factor include commands for performing a temperature correction of at least one ICE material according to an environmental condition of a fluid spectroscopy and an analyte in the database, and wherein the at least one ICE material includes a material in one of a high refraction index layer, a low refraction index layer, or a substrate.