Knowledge Graphs for Subsurface Data Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for processing unstructured geophysical data in subsurface exploration are inefficient due to the inability to capture complex relationships and represent multi-dimensional information effectively, leading to uncertainties in decision-making processes for hydrocarbon exploration and reservoir management.
Innovation Solution
The use of knowledge graphs to represent unstructured geophysical data, applying neural and convolutional network techniques to capture relationships between entities and attributes, and integrating this data with machine learning models to infer subsurface knowledge and improve decision-making processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional unstructured data processing methods are used, then data processing simplicity is maintained, but measurement precision and reliability of subsurface information analysis deteriorate
Solution Approach 1:
The patent segments unstructured subsurface data into structured knowledge graphs with distinct nodes (representing geological entities) and edges (representing relationships). This segmentation transforms complex unstructured data into manageable, interconnected components that can be processed systematically, improving measurement precision while organizing complexity in a structured framework.
Solution Approach 2:
The patent introduces knowledge graphs as an intermediary layer between raw unstructured subsurface data and analysis algorithms. This intermediary structure captures complex relationships and multi-dimensional information in a standardized format, enabling more precise analysis without directly increasing the complexity of underlying processing systems.
2Reliability
If complex relationships in subsurface data are captured using knowledge graphs and machine learning, then reliability of decision-making improves, but device complexity increases
Solution Approach 1:
The patent performs preliminary action by pre-structuring subsurface data into knowledge graphs with defined entities, attributes, and relationships before analysis. This pre-processing step captures complex relationships in advance, allowing downstream machine learning models to operate on well-organized data structures, thereby improving decision reliability while managing system complexity through staged processing.
Solution Approach 2:
The patent changes the structural parameters of data representation from unstructured formats to structured knowledge graph formats with specific node types, edge types, and attribute schemas. This parameter transformation enables more reliable capture of complex geological relationships while maintaining manageable system complexity through standardized data structures.
3Productivity
If multi-dimensional subsurface information is processed using machine learning models, then productivity of exploration analysis improves, but loss of information during processing increases
Solution Approach 1:
The patent creates a structured copy of unstructured subsurface data in the form of knowledge graphs, preserving complex relationships and multi-dimensional information in a standardized format. This copying approach enables efficient machine learning processing while maintaining fidelity to the original data's complex relationships, thereby improving productivity without significant information loss.
Solution Approach 2:
The patent implements a nested structure where knowledge graphs contain nested nodes and edges that represent hierarchical geological relationships. This nesting preserves multi-dimensional information at multiple levels of abstraction, allowing machine learning models to process data efficiently while retaining complex relationship information through the nested hierarchical structure.
Data Source
AI summary
A geoscience knowledge system can be obtained, where the geoscience knowledge system can include one or more of publicly available information, industry information, proprietary information, or task specific information. The geoscience knowledge system can be represented as a graph, graph data, network nodes, image data, tokenized data, or textualized data. Subsurface information can be obtained such as from seismic images or other types of sensor data. The subsurface information can be transformed or pre-processed, such as denoising, to make it suitable for use by the geoscience knowledge system. Then subsurface knowledge can be inferred from the subsurface information using the geoscience knowledge system. The subsurface knowledge can provided estimates, approximations, or value of the subterranean formation of interest in order to calculate an economic model parameter, such as a hydrocarbon distribution proximate the subterranean formation of interest.