Geosteering using reconciled subsurface physical parameters

Abstract

Systems and methods for geosteering using reconciled subsurface physical parameters are disclosed. The methods include obtaining reconciled physical parameters at each of a plurality of locations within a subsurface; training at least one machine learning network to classify the reconciled physical parameters into a rock type based, at least in part, on the reconciled physical parameters; classifying the reconciled physical parameters into the rock type with the at least one machine learning network; and interpreting the rock type to form a subsurface geology model and inform a geosteering decision.

What is claimed: 
     
         1 . A method, comprising:
 obtaining reconciled physical parameters at each of a plurality of locations within a subsurface;   training at least one machine learning network to classify the reconciled physical parameters into a rock type based, at least in part, on the reconciled physical parameters;   classifying the reconciled physical parameters into the rock type with the at least one machine learning network; and   interpreting the rock type to form a subsurface geology model and inform a geosteering decision.   
     
     
         2 . The method of  claim 1 , wherein obtaining reconciled physical parameters further comprises preprocessing the reconciled physical parameters to remove outliers. 
     
     
         3 . The method of  claim 1 , wherein the at least one machine learning network comprises a deep belief network. 
     
     
         4 . The method of  claim 3 , wherein the deep belief network comprises at least one restricted Boltzmann machine (RBM). 
     
     
         5 . The method of  claim 1 , wherein the training further comprises incorporating expert information. 
     
     
         6 . The method of  claim 5 , wherein incorporating the expert information comprises assigning values to nodes. 
     
     
         7 . The method of  claim 6 , wherein the values assigned to the nodes are based on a determination of at least one Shapley value. 
     
     
         8 . The method of  claim 1 , wherein the reconciled physical parameters are obtained from logging-while-drilling (LWD) data. 
     
     
         9 . The method of  claim 8 , wherein the LWD data comprises at least one selected from the group consisting of: neutron porosity data, borehole caliber data, nuclear magnetic resonance data, gamma ray data, weight on bit data, rate of penetration data, inclination data, measured depth data, true vertical depth data, bearing data, temperature data, and pressure data. 
     
     
         10 . A non-transitory computer-readable memory comprising computer-executable instructions stored thereon that, when executed on a processor, cause the processor to perform the steps of:
 obtaining reconciled physical parameters at each of a plurality of locations within a subsurface;   training at least one machine learning network to classify the reconciled physical parameters into a rock type based, at least in part, on the reconciled physical parameters;   classifying the reconciled physical parameters into the rock type with the at least one machine learning network; and   interpreting the rock type to form a subsurface geology model and inform a geosteering decision.   
     
     
         11 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 10 , wherein obtaining reconciled physical parameters further comprises preprocessing the reconciled physical parameters to remove outliers. 
     
     
         12 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 10 , wherein the at least one machine learning network comprises a deep belief network. 
     
     
         13 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 12 , wherein the deep belief network comprises at least one restricted Boltzmann machine (RBM). 
     
     
         14 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 10 , wherein the training further comprises incorporating expert information. 
     
     
         15 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 14 , wherein incorporating the expert information comprises assigning values to nodes. 
     
     
         16 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 15 , wherein the values assigned to the nodes are based on a determination of at least one Shapley value. 
     
     
         17 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 10 , wherein the reconciled physical parameters are determined from physical parameters that are estimated from at least one selected from the group consisting of: logging while drilling (LWD) data, seismic data, and electromagnetic data. 
     
     
         18 . The non-transitory computer-readable memory comprising computer-executable instructions stored thereon of  claim 17 , wherein the LWD data comprises at least one selected from the group consisting of: neutron porosity data, borehole caliber data, nuclear magnetic resonance data, gamma ray data, weight on bit data, rate of penetration data, inclination data, measured depth data, true vertical depth data, bearing data, temperature data, and pressure data. 
     
     
         19 . A system, comprising:
 a computer, configured to:   obtain reconciled physical parameters at each of a plurality of locations within a subsurface,   train at least one machine learning network to classify the reconciled physical parameters into a rock type based, at least in part, on the reconciled physical parameters,   classify the reconciled physical parameters into the rock type with the at least one machine learning network, and   interpret the rock type to form a subsurface geology model; and   a geosteering system, configured to guide a drill bit through the subsurface.   
     
     
         20 . The system of  claim 19 , wherein the drill bit is guided according to an interpreted rock type.