Leak detection for galvanized steel pipes due to loosening of screw thread connections based on acoustic emission and neural networks

Abstract

Galvanized steel pipes with screw thread connections are widely used in indoor gas transportation. In contrast with the failure of pipe tubes, leakage in this system is prone to occur in the screw thread connections. Aiming at this specific engineering application, a method based on acoustic emission (AE) and artificial neural networks (ANNs) is proposed to detect small gas leaks. Experiments are conducted on a specifically designed galvanized steel pipe system with the manipulated leak occurring in the screw thread connection to acquire the raw AE data. The features in the time and frequency domains are extracted and selected to establish an ANN model for leak detection. It has been validated that the developed ANN-based leak detector can achieve an identification accuracy of over 98%. It is also verified that the proposed model is effective even when the AE signals due to a small leak pass over two screw thread connections or an elbow connection.

Experimental validation of gas leak detection in screw thread connections of galvanized pipe based on acoustic emission and neural network

“…As indicated by the previous work, the features such as the peak value, root mean square, and frequency centroid present a good performance for leak detection with unknown leak flow for galvanized steel pipes. Given the stability of frequency spectrum characteristics and the convergence of neural networks, to achieve quantitative detection, the signal frequency spectrum of 2–30 kHz is discretized into 14 frequency bands with each bandwidth of 2 kHz.…”

“…Moreover, the shape and abrasion degree of screw thread connection can disturb the leak signal and increase its nonstationarity and complexity. The BP neural network was initially applied to detect the leak in the screw thread of galvanized gas pipe and proved to be effective …”

“…14,15 Although the monitoring effectiveness for plastic pipes degrades seriously due to the attenuation, 16 it is still a reliable detection method for steel pipes. 17 Besides, as a noninvasive method, AE method has been demonstrated to be promising because it can achieve reliable detection results dynamically and efficiently based on a simple principle. Moreover, it is sensitive to defects and suitable for inaccessible environments.…”

“…The BP neural network was initially applied to detect the leak in the screw thread of galvanized gas pipe and proved to be effective. 17 On the basis of the previous work, 17 experimental investigations and quantitative analysis are conducted to improve the leak detector and clarify its scope of application. Thus, an indoor experimental platform is built to simulate the typical leak scenarios of the most commonly used galvanized pipes.…”

Acoustic Emission Burst Extraction for Multi-Level Leakage Detection in a Pipeline

“…Furthermore, AE sensors provide enough sensitivity to instantly identify any abnormal state in pipelines. AE-based methods have been employed in numerous studies to detect leaks by modeling them as classification problems [8][9][10][11]. These data-driven approaches are reasonable because the AE signal from leakage is non-stationary and becomes difficult to find in an explicit model for leak detection.…”