Deep Neural Network for Electromagnetic Pipe Inspection Artifact Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing pipe inspection technologies face challenges in accurately detecting metal loss due to corrosion in nested tubulars within wellbores, as inversion artifacts can incorrectly indicate metal loss, leading to false positives and reduced image clarity.
Innovation Solution
A deep neural network (DNN) with convolutional layers is applied to corrected inverted images from electromagnetic pipe inspection tools, enhancing the accuracy of tubular integrity property estimation by filtering out inversion artifacts and refining feature representation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If electromagnetic pipe inspection tools are used to detect metal loss in nested tubulars, then inspection capability is provided, but inversion artifacts cause false positives and reduced measurement precision
Solution Approach 1:
A deep neural network is introduced as an intermediary between the electromagnetic inspection tool and the final interpretation. The DNN processes the raw inspection data and identifies inversion artifacts, acting as a mediator that filters out false positives while preserving genuine metal loss detections, thereby resolving the contradiction between inspection reliability and measurement precision
Solution Approach 2:
The patent replaces traditional mechanical inversion algorithms with a data-driven deep neural network approach. This substitution allows the system to learn complex patterns and distinguish between actual metal loss and inversion artifacts, improving both reliability and precision simultaneously by moving from deterministic to intelligent processing
2Measurement precision
If traditional inversion methods are used to process inspection data, then processing speed is maintained, but image clarity and feature representation are degraded due to artifacts
Solution Approach 1:
Traditional mathematical inversion methods are replaced with a deep neural network that uses learned features to enhance image clarity. The DNN's ability to recognize patterns and suppress artifacts directly improves measurement precision, while the increased processing complexity is acceptable given the significant gain in image quality and detection accuracy
Solution Approach 2:
The patent transforms the processing approach by changing from fixed algorithmic parameters to adaptive learned parameters. The deep neural network adjusts its internal parameters during training to optimize image clarity and artifact suppression, allowing the system to achieve superior measurement precision while managing complexity through parameter optimization rather than algorithmic complexity
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 solution provides a more accurate representation of tubular integrity properties, effectively correcting inversion artifacts and improving the detection of metal loss, leading to enhanced inspection reliability and reduced false positives.
Implementation Method 1
An electromagnetic inversion can be applied to the measurements to provide inverted tubular integrity properties
Data Source
AI summary
Methods and systems for inspecting the integrity of multiple nested tubulars are provided. A method for inspecting the integrity of multiple nested tubulars can comprise conveying an electromagnetic pipe inspection tool inside the innermost tubular of the multiple nested tubulars; taking measurements of the multiple nested tubulars with the electromagnetic pipe inspection tool; inverting the measurements for a tubular integrity property of each individual tubular of the multiple nested tubulars to provide inverted tubular integrity properties; arranging the inverted integrity properties into an inverted image representative of an estimated tubular integrity property of each individual tubular; and feeding the inverted image to a pre-trained deep neural network (DNN) to produce a corrected image, wherein the DNN comprises at least one convolutional layer, and wherein the corrected image comprises a representation of the tubular integrity property of each individual tubular of the multiple nested tubulars.