Real-time estimation of water washing using reservoir gas composition

Abstract

A computer-implemented method for predicting water washing parameters is described. The method includes building, using at least one hardware processor, a training dataset comprising gas composition data and corresponding water washing parameters. The method includes training, using the at least one hardware processor, a machine learning model using the training dataset to output water washing parameters. Additionally, the method includes predicting, using the at least one hardware processor, water washing parameters of an unseen well in real time, wherein inputs to the trained machine learning model are gas composition data associated with the unseen well and the trained machine learning model outputs the water washing parameters.

What is claimed is: 
     
         1 . A computer-implemented method for predicting water washing parameters, the method comprising:
 building, using at least one hardware processor, a training dataset comprising gas composition data and corresponding water washing parameters:   training, using the at least one hardware processor, a machine learning model using the training dataset to output water washing parameters:   predicting, using the at least one hardware processor, water washing parameters of an unseen well in real time, wherein inputs to the trained machine learning model are gas composition data associated with the unseen well and the trained machine learning model outputs the water washing parameters:   creating, using the at least one hardware processor, a water washing intensity index based on the predicted water washing parameters; and   integrating, using the at least one hardware processor, the water washing intensity index with a gross depositional environment map, wherein a resulting integrated map is used to guide water washing of accumulations.   
     
     
         2 . The computer implemented method of  claim 1 , comprising obtaining gas compositions associated with the unseen well in real time, during drilling. 
     
     
         3 . The computer implemented method of  claim 1 , comprising obtaining gas compositions of the training dataset after drilling by analyzing rock samples obtained during drilling. 
     
     
         4 . The computer implemented method of  claim 1 , wherein the gas composition data and corresponding water washing parameters of the training dataset are obtained from offset wells. 
     
     
         5 . The computer implemented method of  claim 1 , wherein predicting water washing parameters of an unseen well in real time comprises predicting a gas oil ratio (GOR), a C7 transformation ratio (Tr1) and a present-day reservoir temperature (PDRT) in real time based on one or more statistical relationships. 
     
     
         6 . The computer implemented method of  claim 1 , wherein the training dataset is built by estimating water washing parameters from gas composition data. 
     
     
         7 . The computer implemented method of  claim 1 , wherein the training dataset is based on historical gas composition data and corresponding water washing parameters. 
     
     
         8 . An apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:
 building a training dataset comprising gas composition data and corresponding water washing parameters;   training a machine learning model using the training dataset to output water washing parameters;   predicting water washing parameters of an unseen well in real time, wherein inputs to the trained machine learning model are gas composition data associated with the unseen well and the trained machine learning model outputs the water washing parameters;   creating a water washing intensity index based on the predicted water washing parameters; and   integrating the water washing intensity index with a gross depositional environment map, wherein a resulting integrated map is used to guide water washing of accumulations.   
     
     
         9 . The apparatus of  claim 8 , comprising obtaining gas compositions associated with the unseen well in real time, during drilling. 
     
     
         10 . The apparatus of  claim 8 , comprising obtaining gas compositions of the training dataset after drilling by analyzing rock samples obtained during drilling. 
     
     
         11 . The apparatus of  claim 8 , wherein the gas composition data and corresponding water washing parameters of the training dataset are obtained from offset wells. 
     
     
         12 . The apparatus of  claim 8 , wherein predicting water washing parameters of an unseen well in real time comprises predicting a gas oil ratio (GOR), a C7 transformation ratio (Tr1) and a present-day reservoir temperature (PDRT) in real time based on one or more statistical relationships. 
     
     
         13 . The apparatus of  claim 8 , wherein the training dataset is built by estimating water washing parameters from gas composition data. 
     
     
         14 . The apparatus of  claim 8 , wherein the training dataset is based on historical gas composition data and corresponding water washing parameters. 
     
     
         15 . A system, comprising:
 one or more memory modules;   one or more hardware processors communicably coupled to the one or more memory modules, the one or more hardware processors configured to execute instructions stored on the one or more memory models to perform operations comprising:   building a training dataset comprising gas composition data and corresponding water washing parameters;   training a machine learning model using the training dataset to output water washing parameters;   predicting water washing parameters of an unseen well in real time, wherein inputs to the trained machine learning model are gas composition data associated with the unseen well and the trained machine learning model outputs the water washing parameters;   creating a water washing intensity index based on the predicted water washing parameters; and   integrating the water washing intensity index with a gross depositional environment map, wherein a resulting integrated map is used to guide water washing of accumulations.   
     
     
         16 . The system of  claim 15 , comprising obtaining gas compositions associated with the unseen well in real time, during drilling. 
     
     
         17 . The system of  claim 15 , comprising obtaining gas compositions of the training dataset after drilling by analyzing rock samples obtained during drilling. 
     
     
         18 . The system of  claim 15 , wherein the gas composition data and corresponding water washing parameters of the training dataset are obtained from offset wells. 
     
     
         19 . The system of  claim 15 , wherein predicting water washing parameters of an unseen well in real time comprises predicting a gas oil ratio (GOR), a C7 transformation ratio (Tr1) and a present-day reservoir temperature (PDRT) in real time based on one or more statistical relationships. 
     
     
         20 . The system of  claim 15 , wherein the training dataset is built by estimating water washing parameters from gas composition data.