System and method for predicting oil production from gas equivalent production stream using machine learning

Abstract

A system and method for predicting oil production is disclosed herein. A computing system receives production data for a plurality of wells in a plurality of geographic regions. The computing system identifies a subset of wells from the plurality of wells. The subset of wells includes wells that have a threshold amount of production data. The computing system trains a machine learning model using the subset of wells. The computing system applies the trained learning model to a target well to predict oil production from the target well.

1 . A method for predicting oil production, the method comprising:
 receiving, by a computing system, production data for a plurality of wells in a plurality of geographic regions;   identifying, by the computing system, a subset of wells from the plurality of wells, wherein the subset of wells comprises wells that have a threshold amount of production data;   training, by the computing system, a machine learning model using the subset of wells; and   applying, by the computing system, the trained machine learning model to a target well to predict oil production from the target well.   
     
     
         2 . The method of  claim 1 , wherein training, by the computing system, the machine learning model using the subset of wells comprises:
 training a plurality of machine learning models using the subset of wells, wherein each of the plurality of machine learning models has a different underlying machine learning algorithm.   
     
     
         3 . The method of  claim 2 , further comprising:
 selecting, from the plurality of machine learning models, the trained machine learning model based on the trained machine learning model achieving a highest level of accuracy.   
     
     
         4 . The method of  claim 1 , wherein identifying the subset of wells comprises: parsing, via an automated script, the production data to identify a subset of wells that have the threshold amount of production data. 
     
     
         5 . The method of  claim 1 , wherein identifying the subset of wells comprises:
 parsing, via an automated script, the production data to identify a subset of wells for which an estimated ultimate recovery is known.   
     
     
         6 . The method of  claim 1 , further comprising:
 generating, by the computing system, a training data set for training the machine learning model, the training data set comprising the subset of wells.   
     
     
         7 . The method of  claim 5 , wherein generating, by the computing system, the training data set comprises:
 clustering the plurality of wells using a clustering algorithm based on distance; and   imputing missing values in the production data of the subset of wells based on wells co-located with the wells in the subset of wells.   
     
     
         8 . A system for predicting oil production, the system comprising:
 a processor; and   a memory having programming instructions stored thereon, which, when executed by the processor, causes the system to perform operations comprising:
 receiving production data for a plurality of wells in a plurality of geographic regions; 
 identifying a subset of wells from the plurality of wells, wherein the subset of wells comprises wells that have a threshold amount of production data; 
 training a machine learning model using the subset of wells; and 
 applying the trained machine learning model to a target well to predict oil production from the target well. 
   
     
     
         9 . The system of  claim 8 , wherein training the machine learning model using the subset of wells comprises:
 training a plurality of machine learning models using the subset of wells, wherein each of the plurality of machine learning models has a different underlying machine learning algorithm.   
     
     
         10 . The system of  claim 9 , further comprising:
 selecting, from the plurality of machine learning models, the trained machine learning model based on the trained machine learning model achieving a highest level of accuracy.   
     
     
         11 . The system of  claim 8 , wherein identifying the subset of wells comprises: parsing, via an automated script, the production data to identify a subset of wells that have the threshold amount of production data. 
     
     
         12 . The system of  claim 8 , wherein identifying the subset of wells comprises:
 parsing, via an automated script, the production data to identify a subset of wells for which an estimated ultimate recovery is known.   
     
     
         13 . The system of  claim 8 , further comprising:
 generating a training data set for training the machine learning model, the training data set comprising the subset of wells.   
     
     
         14 . The system of  claim 13 , wherein generating the training data set comprises:
 clustering the plurality of wells using a clustering algorithm based on distance; and   imputing missing values in the production data of the subset of wells based on wells co-located with the wells in the subset of wells.   
     
     
         15 . A non-transitory computer readable medium comprising instructions, which, when executed by a processor, cause a computing system to perform operations comprising:
 receiving, by the computing system, production data for a plurality of wells in a plurality of geographic regions;   identifying, by the computing system, a subset of wells from the plurality of wells, wherein the subset of wells comprises wells that have a threshold amount of production data;   training, by the computing system, a machine learning model using the subset of wells; and   applying, by the computing system, the trained machine learning model to a target well to predict oil production from the target well.   
     
     
         16 . The non-transitory computer readable medium of  claim 15 , wherein training, by the computing system, the machine learning model using the subset of wells comprises:
 training a plurality of machine learning models using the subset of wells, wherein each of the plurality of machine learning models has a different underlying machine learning algorithm; and   selecting, from the plurality of machine learning models, the trained machine learning model based on the trained machine learning model achieving a highest level of accuracy.   
     
     
         17 . The non-transitory computer readable medium of  claim 15 , wherein identifying the subset of wells comprises: parsing, via an automated script, the production data to identify a subset of wells that have the threshold amount of production data. 
     
     
         18 . The non-transitory computer readable medium of  claim 15 , wherein identifying the subset of wells comprises:
 parsing, via an automated script, the production data to identify a subset of wells for which an estimated ultimate recovery is known.   
     
     
         19 . The non-transitory computer readable medium of  claim 15 , further comprising:
 generating, by the computing system, a training data set for training the machine learning model, the training data set comprising the subset of wells.   
     
     
         20 . The non-transitory computer readable medium of  claim 19 , wherein generating, by the computing system, the training data set comprises:
 clustering the plurality of wells using a clustering algorithm based on distance; and   imputing missing values in the production data of the subset of wells based on wells co-located with the wells in the subset of wells.