Source-Receiver Compression for Seismic Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Full-Waveform Inversion (FWI) seismic data processing is hindered by high computational costs due to large data sets and redundancy in survey data, particularly from a non-optimally designed survey, which increases noise sensitivity and requires numerous iterations.
Innovation Solution
A method for compressing survey data by calculating a weighted sum of physical transmitters and receivers, reducing the number of representative sources and receivers through singular-value decomposition, thereby reducing the computational load and noise sensitivity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of sources and receivers are used in FWI survey, then measurement precision and coverage are improved, but computational cost and data processing time increase significantly
Solution Approach 1:
The patent extracts and removes redundant sources and receivers from the survey data by identifying and eliminating those that do not contribute unique information. This is achieved through statistical analysis of the survey design to determine which sources/receivers provide redundant measurements, allowing their removal while maintaining adequate coverage and precision for the inversion process.
Solution Approach 2:
The patent applies partial action by using only the necessary subset of sources and receivers required for effective inversion, rather than processing all collected data. By calculating the minimum required number of sources and receivers based on the survey geometry and target resolution, the method reduces computational load while maintaining sufficient measurement precision for accurate reservoir characterization.
2Measurement precision
If a large number of sources and receivers are used in FWI survey, then measurement precision is improved, but noise sensitivity increases
Solution Approach 1:
The patent removes redundant sources and receivers that contribute to noise accumulation without adding useful information. By statistically analyzing the survey design, the method identifies and eliminates sources/receivers whose measurements are redundant or whose signal-to-noise ratio is insufficient, thereby reducing noise sensitivity while preserving essential measurement coverage.
Solution Approach 2:
The patent changes the parameters of the survey design by recalculating the optimal number and distribution of sources and receivers based on the actual survey geometry and target resolution requirements. This parameter optimization ensures that only necessary measurements are included, reducing the overall noise floor while maintaining adequate precision for inversion.
3Manufacturing precision
If standard FWI method is used with optimally designed survey, then inversion accuracy is improved, but computational overhead increases
Solution Approach 1:
The patent extracts and removes redundant sources and receivers from the survey data before inversion, reducing the dimensionality of the problem. This extraction step identifies and eliminates measurements that do not contribute unique information, thereby reducing computational overhead while maintaining inversion accuracy through the use of only essential data.
Solution Approach 2:
The patent applies partial action by processing only the necessary subset of survey data required for accurate inversion, rather than all collected measurements. By calculating the minimum required number of sources and receivers based on survey geometry and target resolution, the method reduces computational overhead while maintaining sufficient precision for accurate reservoir characterization.
Data Source
AI summary
Source-receiver compression is used to help design surveys and mitigate the computational costs of data set inversion. The source-receiver compression is based on data redundancy and sensitivity. More particularly, a compressed source array is produced for minimum redundancy and maximum sensitivity to reservoir model parameters. The synthesized transmitter array has a reduced number of sources, thereby reducing the number of forward model simulations needed to carry out the inversion. Furthermore, the data collected at the receivers employed in the survey can be compressed. This has the implication of reducing the computational cost of constructing the Jacobian matrix and inverting the corresponding Hessian matrix.