Bayesian Risk Modification for Seismic Anomaly Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods face challenges in accurately valuing seismic information across multiple segments in seismic anomaly analysis, leading to potential misclassification of direct fluid indicators as hydrocarbon indicators, which affects the estimation of prospect chance of success in hydrocarbon exploration.
Innovation Solution
A computer-implemented method that computes prior probabilities of success and failure scenarios for segments, determines likelihoods of anomalies, classifies segments into direct fluid indicator dependency groups, and calculates a posterior chance of success based on the correlation between anomalies using a Bayesian Risk Modification approach, incorporating a DFI correlation parameter to model dependency between segments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If seismic anomalies are treated as direct hydrocarbon indicators without considering fluid conditions, then the prospect chance of success is overestimated, but the classification accuracy deteriorates due to false positives from fluid conditions
Solution Approach 1:
The patent segments the prospect into multiple discrete segments and classifies anomalies into different dependency groups (DFI-dependent, DFI-independent, and mixed groups). This segmentation allows separate analysis of segments with different fluid indicator dependencies, preventing false positives from propagating across the entire prospect while maintaining accurate classification for each segment group.
Solution Approach 2:
The patent introduces a DFI correlation parameter (k) that quantifies the degree of correlation between anomalies in different segments. By adjusting this parameter based on the specific dependency group, the method dynamically changes the interpretation of seismic anomaly likelihoods, allowing accurate differentiation between true hydrocarbon indicators and false positives caused by fluid conditions.
2Productivity
If Bayesian Risk Modification is applied to multiple segments without accounting for anomaly correlation, then the computational complexity is reduced, but the prospect chance of success estimation becomes unrealistic due to overcounting independent anomalies
Solution Approach 1:
The patent divides the prospect into segmented dependency groups based on DFI relationships. By processing each group separately with appropriate correlation parameters, the method maintains computational efficiency while avoiding the unrealistic overestimation that would result from treating all segments as completely independent.
Solution Approach 2:
The patent introduces the DFI correlation parameter k as an intermediary that mediates between complete independence (k=0) and complete dependence (k=1) of anomalies. This parameter allows the Bayesian Risk Modification to account for partial correlations in a computationally efficient manner, producing realistic prospect chance of success estimates without requiring complex multi-segment joint probability calculations.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Methods, systems, and computer-readable media for determining a chance of success for a prospect including two or more segments. The method may include receiving seismic data indicative of a plurality of anomalies in a prospect. The prospect may include a plurality of segments. Prior probabilities of success and failure scenarios may be computed for at least one of the segments of the prospect. Likelihoods of the anomalies may be determined given the success and failure scenarios for the at least one of the segments. At least two of the segments may be classified into a direct fluid indicator dependency group. A degree of correlation may be determined between the anomalies for the direct fluid indicator dependency group. A posterior chance of success may be determined for the prospect based at least in part on the degree of correlation between the anomalies.