Subsurface Lithological Model via Machine Learning Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional subsurface exploration methods are limited by the need for extensive manual surveys and local sensor data, which are impractical for obtaining detailed knowledge of large or global subsurface lithological characteristics.
Innovation Solution
A machine learning system that ingests geoscience and remote sensing data to infer subsurface lithological characteristics, generating a global subsurface model by correlating seismic, drill core, and satellite data using machine learning models, reducing the need for extensive manual surveys.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual surveys and local sensor data are used, then detailed knowledge of subsurface lithology can be obtained, but the coverage area is limited and the process is impractical for large regions
Solution Approach 1:
The patent uses machine learning models as intermediaries to connect remote sensing data with subsurface lithological characteristics. The ML models learn relationships from training data and apply these learned patterns to infer subsurface properties from remotely sensed surface data, enabling coverage of large areas without direct subsurface measurements
Solution Approach 2:
The patent creates a virtual model of subsurface lithology by copying and generalizing patterns from known subsurface data points across the entire study area. The machine learning system learns from limited direct measurements and replicates these patterns to predict subsurface characteristics in areas where no direct measurements exist
2Loss of information
If extensive manual surveys are conducted to obtain detailed subsurface information, then comprehensive lithological data can be collected, but the time and resources required become prohibitive for global or large-scale modeling
Solution Approach 1:
The patent performs preliminary actions by collecting and processing remote sensing data and existing subsurface measurements before the actual inference process. The machine learning models are pre-trained on available data, and then可以快速 apply these models to generate subsurface lithological models for large areas without conducting new manual surveys
Solution Approach 2:
The system uses existing remote sensing data and publicly available subsurface measurements to automatically generate lithological models. The machine learning system serves itself by learning from available data and applying this knowledge to predict subsurface properties, eliminating the need for extensive new data collection efforts
3Measurement precision
If local sensor data is manually placed and maintained, then accurate local subsurface measurements can be obtained, but the complexity and cost of deployment and maintenance increase
Solution Approach 1:
The patent replaces the mechanical system of physical sensors deployed in the subsurface with a computational system based on machine learning. Instead of placing and maintaining physical sensors in difficult-to-access subsurface locations, the system uses algorithms to infer subsurface properties from remotely sensed data, dramatically reducing deployment and maintenance complexity
Data Source
AI summary
This disclosure describes a system and method for generating a subsurface model representing lithological characteristics and attributes of the subsurface of a celestial body or planet. By automatically ingesting data from many sources, a machine learning system can infer information about the characteristics of regions of the subsurface and build a model representing the subsurface rock properties. In some cases, this can provide information about a region using inferred data, where no direct measurements have been taken. Remote sensing data, such as aerial or satellite imagery, gravimetric data, magnetic field data, electromagnetic data, and other information can be readily collected or is already available at scale. Lithological attributes and characteristics present in available geoscience data can be correlated with related remote sensing data using a machine learning model, which can then infer lithological attributes and characteristics for regions where remote sensing data is available, but geoscience data is not.