Time-lapse Survey Noise Reduction via Geometry Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geophysical surveys, particularly time-lapse surveys, face noise issues due to differences in survey geometry between repeated measurements, which hinder accurate detection of subsurface changes in marine environments.
Innovation Solution
A system and method that utilize difference modules, estimate modules, and multi-model adaptive subtraction to generate noise models and adjust time-lapse difference data, reducing noise by subtracting geometry-related artifacts and preserving actual subsurface changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If time-lapse surveys are conducted to detect subsurface changes, then monitoring capability is improved, but noise from geometry differences between surveys increases
Solution Approach 1:
The patent introduces noise models as intermediary representations that capture the geometric differences between surveys. These noise models serve as mediators between the actual survey data and the desired time-lapse difference, allowing the system to separate true subsurface changes from geometric artifacts through mathematical modeling and subtraction.
Solution Approach 2:
The patent creates synthetic copies of survey data by applying estimated noise models to replicate the geometric differences. These synthetic datasets are then used to generate noise models that can be subtracted from actual survey differences, effectively copying and removing the harmful geometric variations while preserving true signals.
2Measurement precision
If noise models are generated and subtracted to reduce geometry-related noise, then measurement precision is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary actions by estimating noise models before generating the final time-lapse difference. The system pre-processes survey data to create noise models that capture geometric differences, allowing these models to be subtracted from actual survey differences in a systematic manner that reduces computational complexity.
Solution Approach 2:
The patent changes parameters by transforming survey data into different mathematical representations. The system estimates noise models as corrections to survey geometry, then applies these parameter changes to adjust the time-lapse difference calculations, improving precision through controlled parameter transformation rather than complex computational methods.
Data Source
AI summary
Techniques are disclosed for reducing noise when computing time-lapse differences between two or more geophysical surveys performed over the same region. In some computer-implemented embodiments, a time-lapse difference is determined between first and second data representing the first and second surveys, respectively. Based on geometry information corresponding to the second survey, first estimated data is generated representing how the first data would have looked if the second survey geometry had been used during the first survey. A noise model is generated based on differences between the first data and the first estimated data. The time-lapse difference is then adjusted using the noise model, thereby reducing noise in the time-lapse difference caused by differences between the geometries of the first and second surveys.