Geosteering via Reconciled Subsurface Parameters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for classifying subsurface rock types ahead of a drill bit are not fast or accurate enough, leading to inefficiencies in well planning and geosteering, as they rely on reconciling varying quality data from different sources like LWD, EM, and seismic surveys.
Innovation Solution
Implementing a deep learning method that reconciles physical parameters from LWD, EM, and seismic data to classify rock types using machine learning networks, such as neural networks and restricted Boltzmann machines, to enhance geosteering decisions by improving data consistency and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional methods are used to classify subsurface rock types from reconciled data, then the classification can be performed, but the accuracy and speed are insufficient for efficient well planning and geosteering
Solution Approach 1:
The patent replaces traditional mechanical classification methods with machine learning networks (neural networks and restricted Boltzmann machines) that automatically learn patterns from reconciled physical parameters. This substitution enables both high accuracy through sophisticated pattern recognition and high speed through automated processing, resolving the contradiction between classification accuracy and processing speed.
2Measurement precision
If data from multiple sources (LWD, EM, seismic) is reconciled to improve quality, then measurement precision improves, but the process complexity increases
Solution Approach 1:
The patent introduces machine learning networks as intermediaries that automatically reconcile physical parameters from multiple data sources (LWD, EM, seismic surveys). These networks learn the relationships between different data types and produce consistent, high-quality estimates without requiring complex manual reconciliation processes, thus improving data quality while managing process complexity.
Solution Approach 2:
The patent transforms raw physical parameters from multiple sources into reconciled estimates by applying machine learning-based parameter transformations. The networks learn optimal parameter relationships and transformations that reconcile discrepancies between different measurement sources, improving overall data quality through automated parameter adjustment.
Data Source
AI summary
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.