Published April 2026, Pg. 54-60

Section: Сorrosion protection

UOT: 622.197

DOI: 10.37474/0365-8554/2026-04-54-60

Monitoring of pipeline corrosion condition using piezoceramic sensors

N.I. Abbasov - SOCAR Transportation Department

M.S. Karimova PhD in Tech. Sc. - Azerbaijan State Oil and Industry University

This paper presents a comprehensive monitoring method based on piezoceramic sensors for diagnosing the internal corrosion condition of main oil pipelines. The proposed approach integrates active sensing, the time reversal method, feature extraction based on wavelet packet energy, and automatic diagnosis using a convolutional neural network. To improve the reliability of ultrasonic signals, the time reversal method was applied, resulting in focused signals. Informative features were extracted from the focused signals based on wavelet packet energy, and internal corrosion levels were automatically identified using artificial intelligence techniques. Experimental results demonstrate that the proposed method achieves high diagnostic accuracy.

References:

1. Luo Z., Han Q., Wang X. Corrosion monitoring of pipelines using time reverssal guided waves. NDT & E International, 2018, v. 98, pp. 123-131.

2. Zhang Y., Li X., Sun H. Wavelet packet energy-based feature extraction for corrosion detection in pipelines. – Measurement, 2019, v. 146, pp. 675-684.

3. Zhao Y., Sun Z., Ye L. Corrosion damage classification using ultrasonic guided waves and convolutional neural networks. – Structural Health Monitoring, 2020, v. 19, № 5, pp. 1681-1694.

4. Yang D., Zhang X., Zhou T., Wang T., Li J. A novel pipeline corrosion monitoring method based on piezoelectric active sensing and convolutional neural networks. – Sensors, 2023, v. 23, № 8, pp. 3855-3874.

5. Wang T., Zhou Z., Wang C. Time reversal guided wave-based damage imaging for pipelines. – Ultrasonics, 2019, v. 96, pp. 110-119.

6. Peng Z., He C., Yang Z. Ultrasonic guided wave-based corrosion monitoring using piezoelectric sensors. – Sensors, 2021, v. 21, № 8, pp. 2765-2780.

7. Zhou J., Liu Y., Li H. Wavelet packet analysis for feature extraction of non-stationary signals in pipeline inspection. –
Mechanical Systems and Signal Processing, 2018, v. 99, pp. 1-15.

8. LeCun Y., Bengio Y., Hinton G. Deep learning. – Nature, 2015, c. 521, № 7553, pp. 436-444.

9. Sun L., Wang C., Chen X. Internal corrosion assessment of pipelines using ultrasonic signals and machine learning. //
Journal of Loss Prevention in the Process Industries, 2022, v. 75, pp. 104698.