Controlling operation of a steam-assisted gravity drainage oil well system by adjusting controls based on forecast emulsion production

A computer program product for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system comprising a computer readable storage medium for storing computer readable program code which, when executed, causes a computer:

• to generate a causal model of the SAGD oil well system;
• to train the causal model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system, the historical time series data being obtained from a plurality of sensors in the SAGD oil well system;
• to utilize the causal model to determine a forecast emulsion production and a forecast set of control parameters associated with one or more of the SAGD production sites of the SAGD oil well system for a given time step, the forecast set of control parameters comprising respective amounts of steam allocated to the one or more SAGD production sites of the SAGD oil well system;
• and
• to adjust a set of controls of the SAGD oil well system based on the forecast emulsion production and the forecast set of control parameters and subject to one or more constraints associated with the SAGD oil well system;
• to operate the SAGD oil well system utilizing the adjusted set of controls;
• to collect data from the plurality of sensors in the SAGD oil well system while operating the SAGD oil well system utilizing the adjusted set of controls;
• and
• to refine the causal model of the SAGD oil well system utilizing the collected data from the plurality of sensors in the SAGD oil well system;
• wherein the utilizing, adjusting, operating, collecting and refining are repeated for multiple time steps;
• wherein the causal model comprises a machine learning model utilizing a set of input parameters, a set of output parameters and at least one hidden layer;
• and
• wherein the machine learning model comprises a neural network model comprising a nonlinear autoregressive exogenous model.

A method for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system comprising:

• generating a causal model of the SAGD oil well system;
• training the causal model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system, the historical time series data being obtained from a plurality of sensors in the SAGD oil well system;
• utilizing the causal model to determine a forecast emulsion production and a forecast set of control parameters associated with one or more of the SAGD production sites of the SAGD oil well system for a given time step, the forecast set of control parameters comprising respective amounts of steam allocated to the one or more SAGD production sites of the SAGD oil well system;
• and
• adjusting a set of controls of the SAGD oil well system based on the forecast emulsion production and the forecast set of control parameters and subject to one or more constraints associated with the SAGD oil well system;
• operating the SAGD oil well system utilizing the adjusted set of controls;
• collecting data from the plurality of sensors in the SAGD oil well system while operating the SAGD oil well system utilizing the adjusted set of controls;
• and
• refining the causal model of the SAGD oil well system utilizing the collected data from the plurality of sensors in the SAGD oil well system;
• wherein the utilizing, adjusting, operating, collecting and refining steps are repeated for multiple time steps;
• wherein the causal model comprises a machine learning model utilizing a set of input parameters, a set of output parameters and at least one hidden layer;
• wherein the machine learning model comprises a neural network model comprising a nonlinear autoregressive exogenous model;
• and
• wherein the method is performed by at least one processing device comprising a processor coupled to a memory, the processing device being operatively coupled to the SAGD oil well system.