Hybrid-permeability log

Abstract

Methods and systems are disclosed. The methods may include collecting matrix training data along a first depth interval of a first well and training a first artificial intelligence (AI) model using the matrix training data. The methods may further include collecting well data along a depth interval of a well, inputting the well data into the first AI model, and producing a predicted matrix permeability log along the depth interval from the first AI model. The methods may still further include collecting an image at a discrete depth within the depth interval of the well, where the image is of a fracture, inputting the image into a second AI model, producing a predicted fracture permeability from the second AI model, and generating the predicted hybrid-permeability log using the predicted matrix permeability log and the predicted fracture permeability.

What is claimed is: 
     
         1 . A method of training a first artificial intelligence (AI) model comprising:
 collecting matrix training data along a first depth interval of a first well,
 wherein the matrix training data comprises training well data and associated training rock core permeability data; and 
   training the first AI model using the matrix training data,
 wherein the first AI model is trained to produce a predicted matrix permeability log from input well data, and 
 wherein the input well data is collected along a second depth interval of the first well or along a depth interval of a second well. 
   
     
     
         2 . The method of  claim 1 , wherein the training well data comprises training geological data. 
     
     
         3 . The method of  claim 1 , wherein the training well data comprises training porosity log data and training rock core porosity data. 
     
     
         4 . The method of  claim 3 , wherein the training porosity log data is within a threshold of the training rock core porosity data. 
     
     
         5 . The method of  claim 1 , wherein the first AI model comprises multi-resolution graph-based clustering (MRGC). 
     
     
         6 . A method of determining a predicted hybrid-permeability log comprising:
 collecting well data along a depth interval of a well;   inputting the well data into a first artificial intelligence (AI) model;   producing a predicted matrix permeability log along the depth interval from the first AI model;   collecting an image at a discrete depth within the depth interval of the well,
 wherein the image is of a fracture; 
   inputting the image into a second AI model;   producing a predicted fracture permeability from the second AI model; and   generating the predicted hybrid-permeability log using the predicted matrix permeability log and the predicted fracture permeability.   
     
     
         7 . The method of  claim 6 , further comprising:
 determining a hydrocarbon production rate based, at least in part, on the predicted hybrid-permeability log; and   determining a production management plan based, at least in part, on the hydrocarbon production rate.   
     
     
         8 . The method of  claim 7 , further comprising:
 taking one or more actions based, at least in part, on the production management plan.   
     
     
         9 . The method of  claim 6 , wherein training the second AI model comprises:
 generating M fracture training pairs by performing steps comprising,
 collecting an mth training image,
 wherein the mth training image is of an mth fracture, 
 
 determining an mth training fracture-identified image using the mth training image, and 
 determining, using a model, an associated mth training fracture permeability for the mth fracture using the mth training fracture-identified image 
 wherein M is an integer greater than or equal to one, 
 wherein m is an integer between 1 and M, inclusive, and 
 wherein each of the M fracture training pairs comprises the mth training image and the associated mth training fracture permeability; and 
   training the second AI model using the M fracture training pairs,
 wherein the second AI model is trained to produce the predicted fracture permeability from the image. 
   
     
     
         10 . The method of  claim 9 , wherein the model comprises Navier-Stokes equations. 
     
     
         11 . The method of  claim 6 , wherein the well data comprises geological data. 
     
     
         12 . The method of  claim 6 , wherein the well data comprises porosity log data and rock core porosity data. 
     
     
         13 . The method of  claim 6 , wherein the second AI model comprises a plurality of convolutional neural networks (CNNs). 
     
     
         14 . The method of  claim 13 , wherein the plurality of CNNs comprises a u-net. 
     
     
         15 . The method of  claim 6 , wherein producing the predicted fracture permeability comprises:
 producing a predicted fracture-identified image from a first CNN;   inputting the predicted fracture-identified image into a second CNN; and   producing the predicted fracture permeability from the second CNN,   wherein the second AI model comprises the first CNN and the second CNN.   
     
     
         16 . A system comprising:
 a computer system configured to:
 receive well data along a depth interval of a well, 
 input the well data into a first artificial intelligence (AI) model, 
 produce a predicted matrix permeability log along the depth interval from the first AI model, 
 receive an image for a discrete depth within the depth interval of the well,
 wherein the image is of a first fracture, 
 
 input the image into a second AI model, 
 produce a predicted fracture permeability from the second AI model, 
 generate a predicted hybrid-permeability log using the predicted matrix permeability log and the predicted fracture permeability, and 
 determine a hydrocarbon production rate based, at least in part, on the predicted hybrid-permeability log; and 
   a production management system configured to:
 determine a production management plan based, at least in part, on the hydrocarbon production rate. 
   
     
     
         17 . The system of  claim 16 , further comprising a first well logging system configured to collect the well data. 
     
     
         18 . The system of  claim 16 , further comprising a second well logging system configured to collect the image. 
     
     
         19 . The system of  claim 16 , further comprising a rock coring system configured to collect rock cores. 
     
     
         20 . The system of  claim 19 , further comprising a permeability system configured to determine associated training rock core permeability data from the rock cores.