Seismic Imaging Framework with Gradient Conditioning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing seismic data processing methods struggle with accurately modeling subsurface structures due to issues like near-surface heterogeneity, coverage gaps, and illumination deficiencies, leading to false anomalies and incomplete subsurface imaging.

Innovation Solution

A method and system are developed to generate synthetic seismic data using a wavelet and velocity model, apply a gradient conditioner, and perform iterative inversion to enhance the accuracy of subsurface modeling, incorporating features like seismic attributes and machine learning for improved data interpretation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If traditional seismic data processing methods are used, then the processing workflow is simpler, but the imaging resolution is lower and false anomalies increase due to near-surface heterogeneity and illumination deficiencies

Engineering Contradiction:
Improveimaging resolutionVSAvoidprocessing workflow complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The processing workflow is segmented into distinct modules: gradient conditioner generation, iterative inversion process, and final imaging. This allows complex processing to be broken down into manageable steps that can be applied systematically to improve imaging resolution while maintaining workflow organization

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A gradient conditioner is generated in advance before the main inversion process. This preliminary action prepares the data by pre-processing gradients to account for near-surface heterogeneity and illumination variations, thereby improving final imaging resolution without adding complexity during the main processing stage

Inventive Principle:
Principle #10Preliminary action

2Reliability

If conventional inversion processes are used, then the computational cost is lower, but coverage gaps and illumination deficiencies lead to incomplete subsurface imaging

Engineering Contradiction:
Improvesubsurface imaging completenessVSAvoidcomputational cost
Core Design Contradiction:
ReliabilityVSUse of energy by stationary object

Solution Approach 1:

The iterative inversion process uses feedback loops where synthetic seismic data is continuously compared with actual seismic data, and the velocity model is updated based on the misfit. This feedback mechanism ensures that coverage gaps and illumination deficiencies are progressively addressed, improving imaging completeness while managing computational resources efficiently

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The inversion process iteratively changes velocity model parameters to minimize the difference between synthetic and observed seismic data. By systematically adjusting these parameters across multiple iterations, the method achieves more complete subsurface imaging without requiring excessive computational resources

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If velocity models without gradient conditioning are used, then the processing is faster, but near-surface heterogeneity causes false anomalies

Engineering Contradiction:
Improvevelocity model accuracyVSAvoidprocessing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The gradient conditioner is generated in advance to pre-process the velocity model data, accounting for near-surface heterogeneity effects. This preliminary conditioning prevents false anomalies from developing during the main inversion process, improving velocity model accuracy without requiring excessive processing time during the critical inversion stage

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP4621447A1Seismic imaging framework
Publication Date: 2025.09.24 SERVICES PETROLIERS SCHLUMBERGER SA
  • EP4621447A1 patent drawingFigure 1
  • EP4621447A1 patent drawingFigure 2
  • EP4621447A1 patent drawingFigure 3

AI summary

A method can include performing a simulation to generate synthetic seismic data using a wavelet and a velocity model of a subsurface geologic region, where the velocity model includes representations of subsurface structural features and associated geophysical parameters; generating a velocity gradient conditioner for the velocity model based at least in part on the synthetic seismic data; performing an iterative inversion process using acquired seismic data for the subsurface geologic region, the velocity model, and the velocity gradient conditioner; and outputting a final velocity model of the subsurface geologic region as a result of the iterative inversion process.