Hydrocarbon phase behavior modeling for compositional reservoir simulation

A data processing system for hydrocarbon phase behavior modeling for compositional reservoir simulation, the data processing system comprising:

• at least one processor;
• and
• a memory storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising:
• estimating phase properties of a hydrocarbon sample based on a mole-fraction weighted mixing rule;
• determining contributions of individual phase components to the mole-fraction weighted phase properties;
• generating input data for a machine learning model configured to perform phase identification and phase split determination in combination, the machine learning model including a first sub-network including a first number of layers each including respective neurons, and a second sub-network including a second number of layers each including respective neurons, the input data including the contributions from the individual phase components to the phase properties,
• wherein the first sub-network and the second sub-network are trained in parallel to enable the machine learning model to satisfy a performance threshold that indicates a recall threshold and a precision threshold;
• generating, based on processing the input data using the neurons of the first number of layers of the first sub-network of the machine learning model, first output data including probability values for each potential phase state of the hydrocarbon sample, the probability values including a first value representing a liquid state for a grid block, a second value for a vapor state for the grid block, a third value for a two-phase state of the grid block, and a fourth value representing a critical state of the grid block;
• processing the first output data including the probability values and the input data including the contributions from the individual phase components to the phase properties by the neurons of the second number of layers of the second sub-network of the machine learning model;
• generating, by the second sub-network, second output data including fractional values for equilibrium of the individual phase components, vapor fraction, vapor compressibility, and liquid compressibility for the hydrocarbon sample;
• and
• outputting, for the grid block of a compositional reservoir simulation, the second output data, wherein the first output data and the second output data satisfy the performance threshold,
• wherein training the first sub-network and the second sub-network comprises:
• receiving training data comprising phase properties values;
• determining a categorical cross-entropy error from the first sub-network;
• generating a probabilities vector based on the probability values and the categorical cross-entropy error;
• processing the probability vector by the second sub-network;
• determining, based on the processing, a mean-squared error (MSE) between predicted output values and output values of the output data;
• and
• training the first sub-network and the second sub-network simultaneously by minimizing the MSE value over a plurality of training epochs.

One or more non-transitory computer readable media storing instructions for hydrocarbon phase behavior modeling for compositional reservoir simulation, the instructions, when executed by at least one processor, being configured to cause the at least one processor to perform operations comprising:

• estimating phase properties of a hydrocarbon sample based on a mole-fraction weighted mixing rule;
• determining contributions of individual phase components to the mole-fraction weighted phase properties;
• generating input data for a machine learning model configured to perform phase identification and phase split determination in combination, the machine learning model including a first sub-network including a first number of layers each including respective neurons, and a second sub-network including a second number of layers each including respective neurons, the input data including the contributions from the individual phase components to the phase properties,
• wherein the first sub-network and the second sub-network are trained in parallel to enable the machine learning model to satisfy a performance threshold that indicates a recall threshold and a precision threshold;
• generating, based on processing the input data using the neurons of the first number of layers of the first sub-network of the machine learning model, first output data including probability values for each potential phase state of the hydrocarbon sample, the probability values including a first value representing a liquid state for a grid block, a second value for a vapor state for the grid block, a third value for a two-phase state of the grid block, and a fourth value representing a critical state of the grid block;
• processing the first output data including the probability values and the input data including the contributions from the individual phase components to the phase properties by the neurons of the second number of layers of the second sub-network of the machine learning model;
• generating, by the second sub-network, second output data including fractional values for equilibrium of the individual phase components, vapor fraction, vapor compressibility, and liquid compressibility for the hydrocarbon sample;
• and
• outputting, for the grid block of a compositional reservoir simulation, the second output data, wherein the first output data and the second output data satisfy the performance threshold,
• wherein training the first sub-network and the second sub-network comprises:
• receiving training data comprising phase properties values;
• determining a categorical cross-entropy error from the first sub-network;
• generating a probabilities vector based on the probability values and the categorical cross-entropy error;
• processing the probability vector by the second sub-network;
• determining, based on the processing, a mean-squared error (MSE) between predicted output values and output values of the output data;
• and
• training the first sub-network and the second sub-network simultaneously by minimizing the MSE value over a plurality of training epochs.