Method and system for forecasting real-time well mass flow in green energy generation utilizing digital twin technology

Abstract

A method of managing a well system includes: obtaining, by a digital twin manager and based on a predetermined monitoring criterion, well mass flow data of the well system; obtaining, by the digital twin manager, modeled well mass flow data for the well system using a physics-based model; training, by the digital twin manager, a physics constrained machine learning model using one or more machine learning algorithms based on the well mass flow data and the modeled well mass flow data as inputs; obtaining, by the digital twin manager, real-time well mass flow data of the well system; outputting, by the digital twin manager, predicted well mass flow data using the real-time well mass flow data and the trained physics constrained machine learning model; and transmitting, by the digital twin manager, a command to the well system that adjusts a well operation based on the predicted well mass flow data.

What is claimed is: 
     
         1 . A method of managing a well system, comprising:
 obtaining, by a digital twin manager and based on a predetermined monitoring criterion, well mass flow data of the well system;   obtaining, by the digital twin manager, modeled well mass flow data for the well system using a physics-based model;   training, by the digital twin manager, a physics constrained machine learning model using one or more machine learning algorithms based on the well mass flow data and the modeled well mass flow data as inputs;   obtaining, by the digital twin manager, real-time well mass flow data of the well system;   outputting, by the digital twin manager, predicted well mass flow data using the real-time well mass flow data and the trained physics constrained machine learning model; and   transmitting, by the digital twin manager, a command to the well system that adjusts a well operation based on the predicted well mass flow data.   
     
     
         2 . The method of  claim 1 ,
 wherein the predicted well mass flow data has higher time resolution than the real-time well mass flow data from the well system and captures non-linear dynamics behavior of the well system,   wherein the well system comprises interconnected subsystems that include a compressor subsystem and a sales header subsystem.   
     
     
         3 . The method of  claim 1 , 
 wherein the physics-based model emulates components of well mass flow behavior of the well system that are below a predetermined frequency, and   wherein the physics constrained machine learning model is trained to predict components of the well mass flow behavior of the well system that are above, below, and include the predetermined frequency.   
     
     
         4 . The method of  claim 1 ,
 wherein the physics constrained machine learning model is trained based on at least six months of the well mass flow data, and   wherein the well mass flow data includes data for both normal operational conditions and shut-down conditions.   
     
     
         5 . The method of  claim 1 :
 wherein the physics constrained machine learning model is obtained using a machine learning algorithm selected from a group consisting of a Levenberg-Marquardt algorithm, a Gauss-Newton algorithm, a steepest descent algorithm, and an artificial neural network.   
     
     
         6 . The method of  claim 1 :
 wherein the physics constrained machine learning model uses a misfit function which includes a well mass flow prediction error, and   wherein the well mass flow prediction error is selected from a group consisting of integral square error (ISE), mean error (ME), normalized ISE, and normalized ME.   
     
     
         7 . A well system, comprising:
 a well site;   a physics-based modeling server that outputs modeled well mass flow data for the well site based on a physics-based model; and   a digital twin manager, coupled to the physics-based modeling server and the well site, that includes a processor,   wherein the digital twin manager:
 obtains, based on a predetermined monitoring criterion, well mass flow data of the well site; 
 obtains modeled well mass flow data for the well site using the physics-based model; 
 trains a physics constrained machine learning model using one or more machine learning algorithms based on the well mass flow data and the modeled well mass flow data as inputs; 
 obtains real-time well mass flow data of the well site; 
 outputs predicted well mass flow data using the real-time well mass flow data and the trained physics constrained machine learning model; and 
 transmits a command to the well site that adjusts a well operation based on the predicted well mass flow data. 
   
     
     
         8 . The well system of  claim 7 ,
 wherein the predicted well mass flow data has higher time resolution than the real-time well mass flow data from the well system and captures non-linear dynamics behavior of the well system,   wherein the well system comprises interconnected subsystems that include a compressor subsystem and a sales header subsystem.   
     
     
         9 . The well system of  claim 7 ,
 wherein the physics-based model emulates components of well mass flow behavior of the well system that are below a predetermined frequency, and   wherein the physics constrained machine learning model is trained to predict components of the well mass flow behavior of the well system that are above, below, and include the predetermined frequency.   
     
     
         10 . The well system of  claim 7 ,
 wherein the physics constrained machine learning model is trained based on at least six months of the well mass flow data, and   wherein the well mass flow data includes data for both normal operational conditions and shut-down conditions.   
     
     
         11 . The well system of  claim 7 :
 wherein the physics constrained machine learning model is obtained using a machine learning algorithm selected from a group consisting of a Levenberg-Marquardt algorithm, a Gauss-Newton algorithm, a steepest descent algorithm, and an artificial neural network.   
     
     
         12 . The well system of  claim 7 :
 wherein the physics constrained machine learning model uses a misfit function which includes a well mass flow prediction error, and   wherein the well mass flow prediction error is selected from a group consisting of integral square error (ISE), mean error (ME), normalized ISE, and normalized ME.   
     
     
         13 . A non-transitory computer readable medium storing instructions executable by a computer processor, the instructions comprising functionality for:
 obtaining well mass flow data of a well system based on a predetermined monitoring criterion;   obtaining modeled well mass flow data for the well system using a physics-based model;   training a physics constrained machine learning model using one or more machine learning algorithms based on the well mass flow data and the modeled well mass flow data as inputs;   obtaining real-time well mass flow data of the well system;   outputting predicted well mass flow data using the real-time well mass flow data and the trained physics constrained machine learning model; and   transmitting a command to the well system that adjusts a well operation based on the predicted well mass flow data.   
     
     
         14 . The non-transitory computer readable medium of  claim 13 ,
 wherein the predicted well mass flow data has higher time resolution than the real-time well mass flow data from the well system and captures non-linear dynamics behavior of the well system,   wherein the well system comprises interconnected subsystems that include a compressor subsystem and a sales header subsystem.   
     
     
         15 . The non-transitory computer readable medium of  claim 13 ,
 wherein the physics-based model emulates components of well mass flow behavior of the well system that are below a predetermined frequency, and   wherein the physics constrained machine learning model is trained to predict components of the well mass flow behavior of the well system that are above, below, and include the predetermined frequency.