Uphole Velocity Modeling Using ML for Stable Seismic Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional seismic surveys face challenges in accurately imaging subsurface reservoirs due to complex physical parameter distributions in the near-surface weathering layer, particularly in arid regions, leading to unreliable calibrations and distorted deep seismic images, which can result in dry wells or missed exploration targets.
Innovation Solution
Employing a computer-implemented method using supervised machine learning, specifically a feed-forward artificial neural network, to interpret uphole travel times and construct a robust velocity model by removing anomalies, discretizing data, and training the model with synthetic data to determine interval velocities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional seismic acquisition layouts are used, then deep reservoir imaging is optimized, but near-surface velocity measurement precision deteriorates
Solution Approach 1:
The patent segments the velocity modeling process into two distinct components: near-surface velocity modeling using uphole survey data and deep reservoir velocity modeling using conventional seismic data. This segmentation allows each component to be optimized independently, with the near-surface model calibrated using specialized uphole measurements and the deep reservoir model calibrated using conventional seismic acquisition, thereby resolving the contradiction between near-surface measurement precision and deep reservoir imaging reliability.
Solution Approach 2:
The patent introduces an intermediary approach by using uphole survey data as a bridge between surface measurements and deep subsurface velocity structures. The uphole surveys provide direct vertical velocity measurements through the weathering layer, which then serve as calibration data for the overlying seismic acquisition, enabling accurate deep imaging without compromising near-surface velocity measurement precision.
2Difficulty of detecting and measuring
If refraction seismology is used, then velocity structure analysis is performed, but hidden low velocity layers cannot be detected
Solution Approach 1:
The patent inverts the conventional refraction seismology approach by using uphole surveys that measure vertical travel times directly. Instead of relying on refracted waves that bypass hidden layers, the uphole method measures travel times through the entire vertical column including hidden low-velocity layers, thereby detecting information that would be lost in conventional refraction surveys.
Solution Approach 2:
The patent creates a simplified copy of the velocity measurement process by using vertical uphole surveys that directly sample the weathering layer and near-surface structure. This copy of the measurement approach bypasses the limitations of conventional seismic methods and directly captures velocity information from hidden layers, preserving information that would otherwise be lost.
3Measurement precision
If detailed log-type velocity-depth profiles are generated, then velocity resolution is improved, but noise and error propagation increase
Solution Approach 1:
The patent applies partial action by generating velocity-depth profiles at an optimal level of detail rather than maximum detail. The uphole survey data is processed to create velocity models with sufficient resolution for calibration purposes without over-processing that would amplify noise and errors. This selective level of detail maintains reliability while providing adequate precision for the intended application.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method provides a stable and accurate velocity model, reducing imaging errors and enabling precise subsurface analysis, thereby improving the accuracy of seismic imaging and drilling operations.
Implementation Method 1
seismic surveys are performed to produce images of the various rock formations in the earth... The seismic surveys obtain seismic data indicating the response of the rock formations to the travel of elastic wave seismic energy
Implementation Method 2
a source is lowered within a shallow borehole and the uphole times are recorded by seismic receivers (for example, geophones) located on the surface
Data Source
AI summary
Constructing a velocity model from uphole seismic survey data using a statistical approach or a machine learning (ML) model. The uphole travel time vs. depth data from the uphole seismic survey is processed by fitting a smoothing function and removing outliers to form an uphole travel time vs. depth function that is then discretized to depth intervals. In the statistical approach, the discretized uphole travel time vs. depth function is segmented by piecewise linear functions, and the linear segments are used to interpret the interval velocities at the corresponding depth intervals. In the machine learning approach, a machine learning model is trained using synthetic uphole travel time data. The trained machine learning model is provided to determine interval velocities from the uphole travel time vs. depth data from the uphole seismic survey.