Related Experiment Video
Updated: Aug 13, 2025
06:17
Quantifying the Relative Thickness of Conductive Ferromagnetic Materials Using Detector Coil-Based Pulsed Eddy Current Sensors
Published on: January 16, 2020
5.8K
A Novel Pipeline Corrosion Monitoring Method Based on Piezoelectric Active Sensing and CNN
Dan Yang1,2, Xinyi Zhang1,3, Ti Zhou4
1Key Laboratory for Metallurgical Equipment and Control of Ministry of Education, Wuhan University of Science and Technology, Wuhan 430081, China.
Sensors (Basel, Switzerland)
|January 21, 2023
Summary
This study introduces a novel method using piezoelectric active sensing and time reversal to monitor pipeline corrosion. The technique accurately identifies internal corrosion status with over 99% accuracy, offering a promising solution for pipeline integrity.
Area of Science:
- Materials Science
- Mechanical Engineering
- Signal Processing
Background:
- Pipeline integrity is crucial for infrastructure safety and operational efficiency.
- Internal corrosion poses a significant threat to pipelines, leading to potential failures.
- Existing monitoring methods often lack the precision for early and accurate detection of internal corrosion.
Purpose of the Study:
- To develop and validate a piezoelectric active sensing-based time reversal method for monitoring internal pipeline corrosion.
- To propose an effective approach combining wavelet packet energy and a Convolutional Neural Network (CNN) for corrosion status identification.
- To quantitatively assess the pipeline's internal corrosion degree promptly and accurately.
Main Methods:
- Utilized two lead zirconate titanate (PZT) patches as actuators and sensors for ultrasonic signal generation and reception.
- Employed time reversal technique to focus ultrasonic signals at the corrosion site.
- Applied wavelet packet transform to extract wavelet packet energy (WPE) from the focused signal.
- Integrated WPE as input for a CNN model to classify pipeline internal corrosion degrees.
Main Results:
- Experimental validation using electrochemically induced corrosion with nine different depths.
- Achieved a classification accuracy exceeding 99.01% for identifying internal corrosion status.
- Demonstrated the method's capability for quantitative monitoring and precise localization of corrosion.
Conclusions:
- The proposed time reversal method combined with WPE-CNN offers a highly accurate and efficient solution for pipeline internal corrosion monitoring.
- This approach has significant potential for real-time, non-destructive evaluation of pipeline integrity.
- The study confirms the method's effectiveness in promptly and accurately pinpointing internal corrosion degrees.
