A Novel Pipeline Corrosion Monitoring Method Based on Piezoelectric Active Sensing and CNN

Abstract

In this study, a piezoelectric active sensing-based time reversal method was investigated for monitoring pipeline internal corrosion. An effective method that combines wavelet packet energy with a Convolutional Neural Network (CNN) was proposed to identify the internal corrosion status of pipelines. Two lead zirconate titanate (PZT) patches were pasted on the outer surface of the pipeline as actuators and sensors to generate and receive ultrasonic signals propagating through the inner wall of the pipeline. Then, the time reversal technique was employed to reverse the received response signal in the time domain, and then to retransmit it as an excitation signal to obtain the focused signal. Afterward, the wavelet packet transform was used to decompose the focused signal, and the wavelet packet energy (WPE) with large components was extracted as the input of the CNN model to rapidly identify the corrosion degree inside the pipeline. The corrosion experiments were conducted to verify the correctness of the proposed method. The occurrence and development of corrosion in pipelines were generated by electrochemical corrosion, and nine different depths of corrosion were imposed on the sample pipeline. The experimental results indicated that the classification accuracy exceeded 99.01%. Therefore, this method can quantitatively monitor the corrosion status of pipelines and can pinpoint the internal corrosion degree of pipelines promptly and accurately. The WPE-CNN model in combination with the proposed time reversal method has high application potential for monitoring pipeline internal corrosion.

Advanced Machine Learning Techniques for Corrosion Rate Estimation and Prediction in Industrial Cooling Water Pipelines

“…CNNs have already shown good performances for corrosion prediction; nevertheless, we lack evidence of previous studies employing such models with the same approach as in the present work. In [ 18 ], Yang et al employ a classification approach to determine the corrosion status based on data acquired through the piezoelectric active sensing-based time reversal method, achieving an accuracy of 99.01%. In [ 51 ], Cantero-Chinchilla et al propose a CNN to make estimations about the thickness values (minimum and mean) of corroded profiles from an ultrasonic time-series measurement.…”

“…Many studies have utilized CNNs to study the advancement of corrosion using images [ 16 , 17 ]. Yet, it is less common to encounter works exploring the application of CNNs to numerical corrosion data, despite the existence of some studies such as [ 18 ]. In other time-dependent problems, like those related to energy generation or consumption, the use of these models is more widespread, since this type of network can extract intricate patterns and dependencies from the data, as well as LSTMs [ 19 , 20 ].…”

A pipeline corrosion detecting method using percussion and residual neural network

“…Ai et al [34] proposed a loss recognition method based on 1D-CNN, which can effectively identify minor damages in concrete structures without complex data preprocessing or manual feature extraction. However, with the increase of working conditions, the amount of data becomes larger, and the learning and training of convolutional neural networks becomes very difficult [35,36]. Two-dimensional (2D) data contains richer information than 1D data, making the presentation of the data more intuitive and easier to understand.…”