Joint Inversion of Seismic Data Across Time Scales
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current seismic data analysis methods face challenges in accurately interpreting datasets expressed in different time scales, particularly in 4D seismic and multi-component seismic, due to time shifts and variations in wave propagation, leading to difficulties in characterizing underground reservoirs effectively.
Innovation Solution
A method for joint stratigraphic inversion of seismic data, using a scale factor constrained by physical laws to align seismic events across different time scales, allowing for the construction of physically plausible images of underground reservoirs by minimizing the difference between synthetic and recorded seismic data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If seismic data from different time scales are analyzed separately, then each dataset can be processed independently, but the interpretation accuracy deteriorates due to time shifts and propagation velocity variations
Solution Approach 1:
The patent combines multiple seismic datasets expressed in different time scales into a unified inversion framework. By integrating 4D seismic data and multi-component seismic data simultaneously, the method resolves time shifts and propagation velocity variations through joint optimization, achieving accurate reservoir characterization without separate processing that would compromise interpretation accuracy.
Solution Approach 2:
The invention transforms the time scale parameter by introducing a scaling factor that relates different time bases. This parameter transformation allows datasets with different propagation velocities and time scales to be reconciled through mathematical scaling, enabling accurate comparison and integration of seismic events across multiple datasets while maintaining interpretation precision.
2Ease of operation
If time shifts are corrected using conventional alignment methods, then seismic events can be aligned, but the physical plausibility deteriorates due to lack of constraints on propagation velocity variations
Solution Approach 1:
The patent implements a feedback mechanism where the inversion process continuously adjusts the scaling factor based on the mismatch between observed and synthesized seismic data. This iterative feedback ensures that time shift corrections remain physically plausible by constraining propagation velocity variations to realistic ranges, while still achieving effective alignment of seismic events across different time scales.
Solution Approach 2:
The invention introduces a scaling factor as an intermediary parameter that mediates between datasets with different time scales. This intermediary enables alignment while maintaining physical plausibility by acting as a bridge that transforms time measurements in a physically consistent manner, rather than applying direct rigid alignment that would violate physical constraints.
3Quantity of substance
If multiple seismic datasets are integrated without time scale normalization, then data redundancy is preserved, but the characterization reliability deteriorates due to time base mismatches
Solution Approach 1:
The patent applies parameter transformation by scaling time measurements to a common reference frame. This allows multiple seismic datasets to be integrated with their time scale differences accounted for through the scaling factor, preserving the quantity and redundancy of data while eliminating time base mismatches that would otherwise compromise characterization reliability.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The method involves carrying out sequential inversion of seismic data to determine estimations of combination of physical quantities from the data expressed in time scales (t0, t1). A scale factor to express synthetic data described in the scale (t0) is defined in the scale (t1) by a differential equation that relates travel time variations of seismic wave differential to the combination of physical quantities. An image is constructed by carrying out a joint inversion in which a cost function using the factor is minimized to evaluate a difference between the synthetic and seismic data.