2-D inversion of multi-component induction logging data to resolve anisotropic resistivity structure

Abstract

High Definition Induction Logging (HDIL) tools can provide reliable information about the vertical and radial variations of resistivity structure in isotropic media. The focusing technique provides quantitative information about the resistivity variation and qualitative information about invasion at the well site. This type of logging tool utilizes transmitter-receiver arrays coaxial with the borehole and thus cannot provide information about anisotropy in vertical wells. This greatly limits the application of array induction tools in the characterization of reservoirs with finely laminated sand/shale sequences. A multi-component induction tool, 3DEX™, has been developed by Baker Atlas and Royal Dutch Shell. It provides the much needed ability to detect anisotropy for sand-shale laminated reservoirs. Data from such a logging tool are inverted to give an estimate of vertical and horizontal resistivity in a vertical borehole. 3DEX™, however, lacks the radial resolution provided by array induction tools. Thus 3DEX™may encounter difficulties in looking through an invaded zone and detecting the anisotropy in the formations. Joint inversion of HDIL and 3DEX™ data is able to identify parameters of the invaded zone as well as of the anisotropic formations.

What is claimed is:  
     
         1 . A method of logging of subsurface formations including a plurality of layers each having a horizontal resistivity and a vertical resistivity, the method comprising: 
 (a) using a multi-component electromagnetic logging tool in a borehole in the subsurface formations for obtaining multi-component measurements indicative of said resistivities of said layers, said borehole associated with an invaded zone in said layers;    (b) defining an initial model of said plurality of layers, said initial model including, for each of said plurality of layers, 
 (i) a horizontal resistivity,  
 (ii) a vertical resistivity,  
 (iii) a length of the invaded zone, and  
 (iv) a resistivity of said invaded zone;  
   (c) determining expected responses of the multi-component logging tool to said model;    (d) defining a data objective function related to a difference between said expected responses and said measurements made with the multi-component logging tool;    (e) iteratively updating said model thereby reducing a global objective function, said global objective function comprising a sum of said data objective function and a model objective function related to changes in said model at each iteration.    
     
     
         2 . The method of  claim 1  wherein said electromagnetic logging tool is conveyed on one of (i) a wireline, (ii) a drillstring, and (iii) coiled tubing.  
     
     
         3 . The method of  claim 1  wherein said measurements made with said multi-component logging tool measures comprise H xx , H zz  and H xz , measurements.  
     
     
         4 . The method of  claim 1  wherein said multi-component measures further comprise H yy  and H xy  measurements.  
     
     
         5 . The method of  claim 1  wherein defining said initial model further comprises using measurements from an array logging tool and deriving the initial model from said array resistivity measurements.  
     
     
         6 . The method of  claim 1  wherein determining said expected responses further comprises using a forward modeling program.  
     
     
         7 . The method of  claim 1  wherein defining said global objective function further comprises estimating a covariance of noise present in said multi-component measurements.  
     
     
         8 . The method of  claim 1  wherein said model objective finction includes a relative weight between resistivity components and length components.  
     
     
         9 . The method of  claim 1  wherein iteratively updating the model further comprises defining a sensitivity matrix relating the observations to model parameters.  
     
     
         10 . The method of  claim 1  wherein iteratively updating said model further comprises windowing of said measurements.  
     
     
         11 . A method of logging of subsurface formations including a plurality of layers each having a horizontal resistivity and a vertical resistivity, the method comprising: 
 (a) using a multi-component logging tool in a borehole in the subsurface formations for obtaining multi-component measurements indicative of said resistivities of said layers, said borehole associated with an invaded zone in said layers;    (b) using an array logging tool in said borehole for obtaining additional measurements primarily indicative of said horizontal resistivity and properties of said invaded zone;    (c) defining an initial model of said plurality of layers, said initial model including, for each of said plurality of layers, 
 (i) a horizontal resistivity,  
 (ii) a vertical resistivity,  
 (iii) a length of the invaded zone, and  
 (iv) a resistivity of said invaded zone;  
   (c) determining expected responses of the multi-component logging tool and the array logging tool to said model;    (d) defining a data objective function related to a difference between said expected responses and said measurements made with the multi-component logging tool and the array logging tool;    (e) iteratively updating said model thereby reducing a global objective function, said global objective finction comprising a sum of said data objective function and a model objective function related to changes in said model at each iteration.    
     
     
         12 . The method of  claim 1  wherein said multi-component logging tool and said array logging tool are conveyed on one of (i) a wireline, (ii) a drillstring, and (iii) coiled tubing.  
     
     
         13 . The method of  claim 11  wherein said measurements made with said multi-component logging tool measures comprise H xx , H zz  and H xz  measurements  
     
     
         14 . The method of  claim 12  wherein said measurements made with said multi-component logging tool further comprise H yy , and H xy  measurements  
     
     
         15 . The method of  claim 11  wherein defining said initial model further comprises using measurements from the array logging tool.  
     
     
         16 . The method of  claim 11  wherein determining said expected responses further comprises using a forward modeling program.  
     
     
         17 . The method of  claim 11  wherein defining said global objective function further comprises estimating a covariance of noise present in said multi-component and array logging measurements.  
     
     
         18 . The method of  claim 11  wherein said model objective function includes a relative weight between resistivity components and length components.  
     
     
         19 . The method of  claim 11  wherein iteratively updating the model further comprises defining a sensitivity matrix relating the observations to model parameters.  
     
     
         20 . The method of  claim 11  wherein iteratively updating said model further comprises windowing of said measurements