System and method for automatic well log normalization in multiple wells

Abstract

Systems and methods for automatic well log normalization are disclosed. The methods include acquiring a target log in a target well and a reference log in a reference well; identifying a stratigraphic interval in the target well and in the reference well; projecting the target log from the target well onto a pseudo target log in a pseudo target well and projecting the reference log from the reference well onto a pseudo reference log in a pseudo reference well; identifying lithologies from a first histogram of the pseudo target log and a second histogram of the pseudo reference log; determining, using the first histogram and the second histogram a regression relationship between the pseudo target log and the pseudo reference log; applying the regression relationship to the pseudo target log to generate a normalized pseudo target log; and determining a reservoir quality of the reservoir that produces hydrocarbons.

What is claimed is: 
     
         1 . A method, comprising:
 acquiring, using a well logging tool, a target log in a target well and a reference log in a reference well; and   using a well log interpretation system:
 identifying a stratigraphic interval in the target well and in the reference well, the target well and reference well being drilled into a reservoir, 
 projecting the target log covering the stratigraphic interval from the target well onto a pseudo target log in a pseudo target well and projecting the reference log over the stratigraphic interval from the reference well onto a pseudo reference log in a pseudo reference well, 
 identifying lithologies from a first histogram of the pseudo target log and a second histogram of the pseudo reference log, 
 determining, using the first histogram and the second histogram a regression relationship between the pseudo target log and the pseudo reference log, 
 applying the regression relationship to the pseudo target log to generate a normalized pseudo target log, and 
 determining, based on the normalized pseudo target log, a reservoir quality of the reservoir that produces hydrocarbons. 
   
     
     
         2 . The method of  claim 1 , wherein the target log and the reference log each comprise a gamma ray log. 
     
     
         3 . The method of  claim 1 , wherein the target well and the reference well are horizontal or deviated wells. 
     
     
         4 . The method of  claim 1 , wherein the regression relationship is a linear relationship. 
     
     
         5 . The method of  claim 1 , wherein the pseudo target well and the pseudo reference well are perpendicular to the stratigraphic interval. 
     
     
         6 . The method of  claim 1 , wherein identifying lithologies comprises fitting Gaussian mixture models to the first histogram and to the second histogram. 
     
     
         7 . The method of  claim 6 , wherein peak values of the Gaussian mixture models are associated with the lithologies. 
     
     
         8 . The method of  claim 6 , wherein modes of the Gaussian mixture models are used to fit the regression relationship. 
     
     
         9 . The method of  claim 1 , wherein the stratigraphic interval is a hydrocarbon reservoir. 
     
     
         10 . The method of  claim 1 , wherein projecting the target log and projecting the reference log comprise interpolating onto a uniform sampling grid. 
     
     
         11 . A system, comprising:
 a well logging tool, configured to acquire a target log in a target well and a reference log in a reference well, the target well and reference well being drilled into a reservoir; and   a well log interpretation system, configured to:
 identify a stratigraphic interval in the target well and in the reference well, 
 project the target log covering the stratigraphic interval from the target well onto a pseudo target log in a pseudo target well and projecting the reference log over the stratigraphic interval from the reference well onto a pseudo reference log in a pseudo reference well, 
 identify lithologies from a first histogram of the pseudo target log and a second histogram of the pseudo reference log, 
 determine, using the first histogram and the second histogram a regression relationship between the pseudo target log and the pseudo reference log, 
 apply the regression relationship to the pseudo target log to generate a normalized pseudo target log, and 
 determine, based on the normalized pseudo target log, a reservoir quality of the reservoir that produces hydrocarbons. 
   
     
     
         12 . The system of  claim 11 , wherein the target log and the reference log each comprise a gamma ray log. 
     
     
         13 . The system of  claim 11 , wherein the target well and the reference well are horizontal or deviated wells. 
     
     
         14 . The system of  claim 11 , wherein the regression relationship is a linear relationship. 
     
     
         15 . The system of  claim 11 , wherein the pseudo target well and the pseudo reference well are perpendicular to the stratigraphic interval. 
     
     
         16 . The system of  claim 11 , wherein identifying lithologies comprises fitting Gaussian mixture models to the first histogram and to the second histogram. 
     
     
         17 . The system of  claim 16 , wherein peak values of the Gaussian mixture models are associated with the lithologies. 
     
     
         18 . The system of  claim 16 , wherein modes of the Gaussian mixture models are used to fit the regression relationship. 
     
     
         19 . The system of  claim 11  wherein the stratigraphic interval is a hydrocarbon reservoir. 
     
     
         20 . The system of  claim 11 , wherein projecting the target log and projecting the reference log comprise interpolating onto a uniform sampling grid.