ML Classifier for Geophysical Noise Attenuation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Geophysical surveys face challenges in noise filtration during marine oil and gas exploration, where traditional manual quality control is time-consuming and inefficient, especially with increasing data volumes, leading to suboptimal seismic imaging and interpretation accuracy.
Innovation Solution
Implementing a machine learning classifier trained on subsets of survey data to automate quality control decisions, using filtering techniques like singular spectrum analysis and independent component analysis to distinguish between acceptable, mild, and harsh filtering, thereby reducing human intervention and improving data processing efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual quality control is used to check filtration, then filtering quality can be assessed, but the process is time-consuming and inefficient
Solution Approach 1:
The patent replaces manual mechanical quality control with an automated machine learning classification system. The system uses trained classifiers to automatically evaluate filtration quality by analyzing sensor data characteristics, eliminating the need for time-consuming manual inspection while maintaining or improving assessment accuracy.
Solution Approach 2:
The system enables self-service quality control where the filtration system automatically evaluates its own output using machine learning models. The classification engine autonomously determines whether filtration is acceptable, mild, or harsh based on data patterns, without requiring external manual intervention for each assessment.
2Ease of operation
If traditional manual quality control is used, then filtering decisions can be made, but human intervention is required which reduces efficiency
Solution Approach 1:
The patent replaces human operational intervention with an automated machine learning system. The classification engine processes sensor data and makes filtration quality decisions automatically, eliminating the need for human operators to manually assess each data set while significantly increasing processing throughput and efficiency.
Solution Approach 2:
The system introduces a machine learning classification engine as an intermediary between data collection and quality control decisions. This intermediary automatically analyzes sensor data characteristics and provides objective filtration quality assessments, replacing subjective human judgment and enabling scalable automated decision-making.
3Reliability
If filtration is applied to remove noise, then signal quality improves, but harsh filtration may cause distortions in desired signals
Solution Approach 1:
The system implements feedback-based filtration control where the machine learning classifier continuously monitors sensor data characteristics and provides feedback on filtration quality. Based on this feedback, the system can adjust filtration parameters in real-time to maintain signal quality while avoiding excessive filtering that would cause distortion.
Solution Approach 2:
The system employs dynamic filtration where filtering intensity is adjusted based on real-time data characteristics rather than applying fixed aggressive filtering. The machine learning model dynamically determines appropriate filtration levels for different data segments, preserving signal integrity while effectively removing noise.
Data Source
AI summary
Techniques are disclosed relating to machine learning in the context of noise filters for sensor data, e.g., as produced by geophysical surveys. In some embodiments, one or more filters are applied to sensor data, such a harsh filter determined to cause a threshold level of distortion in measured reflections, a mild filter determined to leave a threshold level of remaining noise signals, or an acceptable filter. In some embodiments, the system trains a machine learning classifier based on outputs of the filtering procedures and uses the classifier to determine whether other filtered sensor data from the same survey exhibits acceptable filtering. This may improve accuracy or performance in detecting unacceptable filtering, in some embodiments.