Study on the Small Leakage Characteristics and Noise Reduction Methods of Liquid Pipelines Based on Internal Acoustic Detection

Abstract

Abstract Timeously detecting pipeline small leakages caused by corrosion, aging, and crack can critically impact structural safety. However, identifying small leaks is challenging and small leaks can gradually develop into significant leakage problems. Although the negative wave method is currently the most common approach used for pipeline leakage detection, it is limited by the working conditions, length, and diameter of the pipeline and cannot be employed to identify small leaks of less than 1 L per min in pipeline. Therefore, to address this problem, this paper established a technique based on passive acoustic internal detection. This study simulated the variation laws between the leakage rates and sound power levels with different leakage apertures and internal pressures using the Ansys Fluent 19.0 software and performed small leakage detection experiments to verify the feasibility of the finite element simulation method. The findings indicated a positive correlation change in the leakage rates and sound power levels in conjunction with the leakage apertures and internal pressures. Furthermore, a denoising algorithm based on a small leakage of self-encoder was proposed, and a denoising model was introduced for signal. This improved denoising efficiency and reduced the dependence on sample data. The signal-to-noise ratio (SNR) of this algorithm exceeded that of traditional denoising methods, while the root mean square error (RMSE) was lower, verifying its feasibility in denoising. It also provided a research basis for denoising in complex operational environments.