Microseismic Event Depth Inversion From Seismic Image Attributes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current microseismic event location methods in subterranean formations face challenges due to the need for formation velocity information, uneven receiver distribution, and high data volume and low signal-to-noise ratios, especially in single well monitoring, making accurate event detection and location difficult.
Innovation Solution
A process and system utilizing a dense seismic receiver array to convert seismic data into an image domain, perform edge detection, characterize linear segments, and estimate microseismic event depth based on slowness attributes without requiring a velocity model, using techniques like Gaussian denoising and Hough transforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional time picking or energy scanning methods are used for microseismic event location, then the processing can be performed with available data, but the method requires formation velocity information and evenly distributed receivers which are unavailable in single well monitoring
Solution Approach 1:
The patent inverts the conventional approach by not requiring velocity model information. Instead of using travel time and velocity to calculate location, the method uses the image domain representation where location information is directly encoded in the spatial distribution of seismic energy, eliminating the need for velocity model inversion
Solution Approach 2:
The patent replaces the conventional mechanical/geometric approach (time picking, velocity modeling, ray tracing) with an image processing approach. By transforming seismic data into the image domain and applying edge detection algorithms, the method substitutes complex seismic inversion mechanics with image analysis techniques that directly reveal event locations
2Area of stationary object
If optical fiber distributed acoustic sensing (DAS) is used to increase monitoring coverage, then the data volume increases 2-4 times and signal to noise ratio decreases, making current processing techniques inadequate
Solution Approach 1:
The patent merges multiple seismic traces into the image domain representation, where coherent signals from multiple receivers constructively interfere to form distinct image events while incoherent noise averages out. This combining process inherently enhances the signal-to-noise ratio while processing the full DAS data volume
Solution Approach 2:
The patent introduces the image domain as an intermediary representation between raw seismic data and event location results. This intermediate domain transforms the problem by encoding spatial and temporal information in a form that is more robust to noise and更适合 for automated detection using edge detection algorithms
3Measurement precision
If receivers are evenly distributed in 3D space for appropriate accuracy, then location precision improves, but this is not possible for single well monitoring configurations
Solution Approach 1:
The patent accepts and exploits the asymmetric, linear configuration of single well receivers rather than requiring symmetric 3D distribution. The image domain transformation and edge detection methodology is specifically designed to extract maximum location information from linear arrays, turning the geometric limitation into a workable configuration
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Processes and systems for microseismic event detection. In some embodiments, the process can include acquiring seismic data with a dense seismic receiver array; converting seismic data into an image domain dataset; detecting one or more seismic waveform edges; characterizing within the image domain dataset one or more linear segments which each define a portion of one or more of the one or more seismic waveform edges, and wherein the linear segments are functions of one or more seismic arrival times and one or more subterranean depths; and estimating the subterranean depth of the one or more microseismic events within a subterranean formation based on the one or more linear segments.