Algorithm for cost effective thermodynamic fluid property predictions using machine-learning based models
A method for determining isothermal phase behavior for reservoir simulation includes generating a training data set using negative flash calculations, training a first machine learning algorithm to identify a supercritical phase and a subcritical phase, training a second machine learning algorithm to identify a number of stable phases in the subcritical phase, and training a third machine learning algorithm to determine a phase split of the subcritical phase that has more than one identified stable phase.
1. A computer-implemented method, comprising:
generating a training data set using negative flash calculations;
training, using the training data set, a first machine learning algorithm to identify a supercritical phase and a subcritical phase based on a tie-line length (t) of a ternary diagram of data under analysis, wherein a value of t equal to 0 is indicative of a supercritical phase and a value of t greater than 0 is indicative of a subcritical phase;
training, using the training data set, a second machine learning algorithm to identify a number of stable phases in the subcritical phase based on a fraction (β) of moles in a vapor phase relative to a total number of moles in the data under analysis, wherein a value of β between 0 and 1 is indicative of more than one stable phase; and
training a third machine learning algorithm to determine a phase split of the subcritical phase that has more than one identified stable phase based on a mole fraction and a molar composition for each of the more than one stable phase.
2. The computer-implemented method of claim 1 , wherein the training data set includes input data points and output data points, and the output data points are calculated using the negative flash calculations based on the input data points.
3. The computer-implemented method of claim 2 , wherein the output data points include supercritical data points, subcritical single-phase data point, and subcritical two-phase data points.
4. The computer-implemented method of claim 1 , wherein the first machine learning algorithm is a relevance vector machine (RVM) algorithm.
5. The computer-implemented method of claim 1 , wherein the first machine learning algorithm is trained by solving a classification problem.
6. The computer-implemented method of claim 1 , wherein the second machine learning algorithm is a RVM algorithm.
7. The computer-implemented method of claim 1 , wherein the second machine learning algorithm is trained by solving a multi-class classification problem.
8. The computer-implemented method of claim 1 , wherein the third machine learning algorithm is an artificial neural network algorithm.
9. The computer-implemented method of claim 1 , further comprising using the trained first machine learning algorithm, the second trained machine learning algorithm, and the third trained machine learning algorithm to calculate isothermal phase equilibria.
10. The computer-implemented method of claim 9 , further comprising constructing one or more Jacobian matrices for reservoir simulation based on the calculated isothermal phase equilibria.
11. A non-transitory, computer-readable medium storing one or more instructions executable by a computer system to perform operations comprising:
generating a training data set using negative flash calculations;
training, using the training data set, a first machine learning algorithm to identify a supercritical phase and a subcritical phase based on a tie-line length (t) of a ternary diagram of data under analysis, wherein a value of t equal to 0 is indicative of a supercritical phase and a value of t greater than 0 is indicative of a subcritical phase;
training, using the training data set, a second machine learning algorithm to identify a number of stable phases in the subcritical phase based on a fraction (β) of moles in a vapor phase relative to a total number of moles in the data under analysis, wherein a value of β between 0 and 1 is indicative of more than one stable phase; and
training a third machine learning algorithm to determine a phase split of the subcritical phase that has more than one identified stable phase based on a mole fraction and a molar composition for each of the more than one stable phase.
12. The non-transitory, computer-readable medium of claim 11 , wherein the output data points include supercritical data points, subcritical single-phase data point, and subcritical two-phase data points.
13. The non-transitory, computer-readable medium of claim 11 , wherein the first machine learning algorithm is a relevance vector machine (RVM) algorithm, the second machine learning algorithm is a RVM algorithm, and a third machine learning algorithm is an artificial neural network algorithm.
14. The non-transitory, computer-readable medium of claim 11 , further comprising using the trained first machine learning algorithm, the second trained machine learning algorithm, and the third trained machine learning algorithm to calculate isothermal phase equilibria.
15. A computer-implemented system, comprising:
one or more computers; and
one or more computer memory devices interoperably coupled with the one or more computers and having tangible, non-transitory, machine-readable media storing instructions that, when executed by the one or more computers, perform operations comprising:
generating a training data set using negative flash calculations;
training, using the training data set, a first machine learning algorithm to identify a supercritical phase and a subcritical phase based on a tie-line length (t) of a ternary diagram of data under analysis, wherein a value of t equal to 0 is indicative of a supercritical phase and a value of t greater than 0 is indicative of a subcritical phase;
training, using the training data set, a second machine learning algorithm to identify a number of stable phases in the subcritical phase based on a fraction (β) of moles in a vapor phase relative to a total number of moles in the data under analysis, wherein a value of β between 0 and 1 is indicative of more than one stable phase; and
training a third machine learning algorithm to determine a phase split of the subcritical phase that has more than one identified stable phase based on a mole fraction and a molar composition for each of the more than one stable phase.
16. The computer-implemented system of claim 15 , wherein the training data set includes input data points and output data points, and the output data points are calculated using the negative flash calculations based on the input data points, and wherein the output data points include supercritical data points, subcritical single-phase data point, and subcritical two-phase data points.
17. The computer-implemented system of claim 15 , wherein the first machine learning algorithm is a relevance vector machine (RVM) algorithm, the second machine learning algorithm is a RVM algorithm, and a third machine learning algorithm is an artificial neural network algorithm.