Oilfield water and well management using modeled predictions of oilfield water production or hydrocarbon production

Abstract

A method for managing oilfield water. Oilfield water data is grouped into discrete and non-overlapping groups. Outliers are removed from the data. Features of the remaining data are analyzed to identify most discriminative feature. The data is separated into training data and testing data, and the training data is fit into a model that shows the best precision, accuracy and recall. The model is confirmed using the testing data. Upon confirmation of the accuracy of the model, the model is applied to data for a new proposed oilfield well, and a new proposed project is implemented or disapproved based on a result of the identified model that predicts water production of the new proposed oilfield well.

1 - 20 . (canceled) 
     
     
         21 . A method of oilfield water management, comprising the steps of:
 obtaining oilfield water data;   discretizing the oilfield water data into a plurality of oilfield water data groups, each oilfield water data group representing oilfield water data taken from a discrete and non-overlapping predetermined time period and representing at least one observation of water use or water production from a plurality of well completions;   separating the oilfield water data into training data and testing data;   generating a plurality of models based on the training data using a plurality of different classification models;   identify a preferred model among the plurality of models, the preferred model exhibiting, among the plurality of models, at least one of higher precision, accuracy, and recall in modelling the training data;   running the preferred model on the testing data to confirm accuracy of the preferred model in estimating the oilfield water production of an existing oilfield well;   applying the preferred model to determine an estimated oilfield water production range of the existing oilfield well;   monitoring actual oilfield water production of the existing oilfield well by sensor networks configured to monitor oilfield water production; and   determining whether the actual oilfield water production of the existing oilfield well deviates from the estimated oilfield water production range.   
     
     
         22 . The method of  claim 21 , further comprising
 alerting an operator via electronic communication protocol when the actual oilfield water production of the existing oilfield well deviates from an estimated oilfield water production range.   
     
     
         23 . The method of  claim 21 , wherein identifying the preferred model is performed using a grid search function. 
     
     
         24 . The method of  claim 21 , further comprising updating the preferred model based on updated oilfield water data. 
     
     
         25 . The method of  claim 21 , further comprising applying the preferred model to project water usage requirements of a new proposed oilfield well drilling operation. 
     
     
         26 . The method of  claim 21 , further comprising estimating water resources required to complete a new proposed oilfield well drilling operation based on a result of the preferred model. 
     
     
         27 . A method of oilfield hydrocarbon production management, comprising:
 obtaining oilfield hydrocarbon production data;   discretizing the oilfield hydrocarbon production data into a plurality of oilfield hydrocarbon production data groups, each oilfield hydrocarbon production data group representing oilfield hydrocarbon production data taken from a discrete and non-overlapping predetermined time period and representing at least one observation of oilfield hydrocarbon production from a plurality of well completions;   separating the oilfield hydrocarbon production data into training data and testing data;   
       generating a plurality of models based on the training data using a plurality of different classification models;
 identifying a preferred model among the plurality of models, the preferred model exhibiting, among the plurality of models, at least one of higher precision, accuracy, and recall in modelling the training data; 
 running the preferred model on the testing data to confirm accuracy of the preferred model in predicting oilfield hydrocarbon production in an existing oilfield well; 
 applying the preferred model to determine an estimated oilfield hydrocarbon production range of the existing oilfield well; 
 monitoring actual oilfield hydrocarbon production of the existing oilfield well by sensor networks configured to monitor the oilfield hydrocarbon production; and 
 determining whether the actual oilfield hydrocarbon production of the existing oilfield well deviates from the estimated oilfield hydrocarbon production range. 
 
     
     
         28 . The method of  claim 27 , further comprising
 alerting an operator via electronic communication protocol when the actual oilfield hydrocarbon production of the existing oilfield well deviates from an estimated oilfield hydrocarbon production.   
     
     
         29 . The method of  claim 27 , wherein identifying the preferred model is performed using a grid search function. 
     
     
         30 . The method of  claim 27 , further comprising updating the preferred model based on updated oilfield hydrocarbon production data. 
     
     
         31 . The method of  claim 27 , further comprising applying the preferred model to a proposed oilfield well to project hydrocarbon production. 
     
     
         32 . A system of monitoring oilfield water production by predicting oilfield water production, comprising:
 a database of oilfield water data;   sensor networks configured to monitor oilfield water production; and   a computer system with a processor for executing instructions stored in a computer-readable medium for:   obtaining oilfield water data,   storing the oilfield water data in the database,   discretizing the oilfield water data into a plurality of oilfield water data groups, each oilfield water data group representing oilfield water data taken from a discrete and non-overlapping predetermined time period and representing at least one observation of water use or water production from a plurality of well completions,   separating the oilfield water data into training data and testing data,   generating a plurality of models based on the training data using a plurality of different classification models;   
       identifying a preferred model among a plurality of models, the preferred model exhibiting, among the plurality of models, at least one of higher precision, accuracy, and recall in modelling the training data,
 running the preferred model on the testing data to confirm accuracy of the preferred model in estimating oilfield water production of an existing oilfield well, 
 applying the preferred model to determine an estimated oilfield water production range of the existing oilfield well, 
 receiving data of actual oilfield water production of the existing oilfield well from the sensor networks, and 
 determining whether the actual oilfield water production of the existing oilfield well deviates from the estimated oilfield water production range. 
 
     
     
         33 . The system of  claim 32 , wherein the instructions stored in the computer-readable medium further comprise instructions for alerting an operator via electronic communication protocol when the actual oilfield water production of the existing oilfield well deviates from an estimated oilfield water production range. 
     
     
         34 . The system of  claim 32 , wherein identifying the preferred model is performed using a grid search function. 
     
     
         35 . The system of  claim 32 , wherein the instructions stored in the computer-readable medium further comprise instructions for updating the preferred model based on updated oilfield water data. 
     
     
         36 . The system of  claim 32 , wherein the instructions stored in the computer-readable medium further comprise instructions for applying the preferred model to project water usage requirements of a new proposed oilfield well drilling operation. 
     
     
         37 . The system of  claim 32 , wherein the instructions stored in the computer-readable medium further comprise instructions for estimating water resources required to complete a new proposed oilfield well drilling operation based on a result of the preferred model.