Acoustic method of high-pressure natural gas pipelines leakage detection: Numerical and applications

Highlights

• The acoustic characteristics in high pressure natural gas pipeline are analyzed by multi-physics coupled.

• The laws of pressure perturbations changing with many factors can be obtained.

• Infrasound detection technology based on Möhring acoustic analogy theory is applied to field pipeline.

• The method reduces the response time and improves the positioning accuracy.

Abstract

Leak detection is one of the most important issues in natural gas pipelines operation, major accidents can result from pipe gas leaks in pipelines that transport high-pressure gases. Natural gas pipeline leakage can cause huge economic losses and affect public health. Currently, pipeline leak detection relies on the semi-empirical formula based on the traditional basis. However, such methods have obvious limitations. It is impossible to predict the new process in a forward-looking way, which leads to a sharp increase in cost. The infrasonic wave method offers the possibility of continuous acoustic monitoring of pipelines and remote detection of leaks. In this work, the time-frequency signal of pipe leakage acoustic wave was studied by using the method of “acoustic - pipe and acoustic - pressure” multi-physical field coupling. Finally, the acoustic leakage monitoring method was applied to the field pipeline. The results show that: 1) natural gas pipeline leakage is a kind of broadband noise. With the increase of frequency, the energy tends to oscillate and decay; 2) the acoustic wave propagates from the leakage hole, and the amplitude decreases rapidly. 3) the leakage acoustic energy is mainly concentrated below 20 Hz, and the mean sound pressure increases with the increase of internal pressure and leakage diameter; 4) ultra-low frequency sound pressure level (SPL) is of great significance to the detection of natural gas pipeline leakage by the acoustic method. The field experiment demonstrates that this method can improve the detection ability of small leakage. Hence, the proposed method provides a new approach for the detection of pipeline leakage in technology popularization and engineering applications.

Introduction

In China, the length of natural gas pipelines will reach 104,000 km by 2020 and 163,000 km by 2025 [1]. According to the rough statistics of the service age of natural gas pipelines, the service age of pipelines over 20 years accounts for 17%. As pipelines will enter a highly aging stage, pipeline accidents will also occur frequently over time [2]. As a result of corrosion, construction defects, mechanical failure, ground movement, third-party interference, etc. Pipeline leakage is one of the most common accidents. As natural gas is usually combustible, explosive, and diffusible, there have been vast amounts of natural gas pipeline leakage accidents causing considerable death toll and environmental damage. Therefore, to prevent accidents which can lead to more damage, it is necessary to detect the leakage in a short time [3,4]. Leakage is one of the main causes of natural gas pipeline failure. Currently, many leak detection methods have been developed for pipelines, such as methods based on negative pressure wave, flow balance, leak detection cable, distributed optical fiber, and acoustic waves, etc [[5], [6], [7], [8], [9], [10], [11], [12], [13]].

Generally, among the current fault diagnosis methods for the natural gas pipeline leakage, some are only suitable for the specific operating environment. For example, the negative pressure wave method uses the sensors at both ends of the pipeline to collect the leakage information, and then the leakage is located according to the transmission speed of the negative pressure wave and the time difference between the arrival of the sensors at both ends of the pipeline. It can only detect large leaks and is insensitive to small leaks. The flow balance method detects pipeline leaks by measuring the flow at the inlet and outlet of the pipeline. However, it is not possible to locate the leak. The leak detection cable method is to install gas-sensitive cables on the outer wall of the pipeline for leak detection. Besides, it cannot detect slow leaks and costs too much. Distributed optical fiber method can detect small leakage quickly and has high positioning accuracy. However, it needs to lay optical fiber along the pipeline, which has high costs and is difficult to lay the pipeline already built, etc. These methods have their limitations so that the detection and positioning of pipeline leakage cannot be timely and accurate. Among them, the acoustic method is superior to the traditional ones, such as higher sensitivity, longer detection distance, lower false alarm rate, quicker leak detection, real-time, etc.

Much work has been done on pipeline leak acoustic detection; these works have developed lots of experience. Muggleton [14] established the propagation model of sound waves in pipelines and verified the correctness of the model and attenuation characteristics in buried pipelines through experiments, and the experimental results were in good agreement with the theoretical model. However, the acoustic characteristics of pipeline leakage have not been studied systematically. Kim [15] carried out a time-frequency analysis on the acoustic wave and concluded that the cut-off frequency of the acoustic wave in the pipeline was obtained, which provided a choice for the filter bandwidth required for pipeline leak detection. The propagation law of leakage acoustic waves along the pipeline is not analyzed. Ayed [16] carried out a transient analysis on the condition of the pipeline with two leakage points and obtained the pressure signal characteristics and leakage location results. However, the influence of acoustic characteristics of the leakage hole is not considered. Park [17] uses the minimum variance inverse frequency method to accurately estimate the arrival time difference in the noise. Yang [18] proved the feasibility of acoustic wave leakage detection technology through the processing of laboratory test data. It has not been applied to the actual field, and the field operability is unknown. Guo [19] discussed the law of pressure wave propagation and attenuation when pipeline leakage occurs by using the pressure wave excitation mode of mechanical impact. There is no further analysis of acoustic attenuation characteristics of high-pressure natural gas pipelines. Chen [20] proposed a method using an acoustic technique based on instantaneous energy (IE) distribution and correlation analysis. The results show that the leakages are successfully detected and the average recognition rate reaches 93.3%. However, the time-frequency characteristics of the leakage sound wave are not considered. Zhang [21] designed a novel leak detection monitoring system (LDMS) of long-distance oil pipelines based on dynamic pressure transmitter (DPT). Wavelet transform and empirical mode decomposition (EMD) are used to process the collected leakage signals. The system can correctly detect leaks with a rate of 96.7% and the greatest localization error is 101 m. Whether the detection system is suitable for the high-pressure gas pipeline is not considered.

In the aforementioned methods, scholars have carried out a lot of work and research on pipeline leakage monitoring. Nevertheless, it mainly focuses on acoustic signal processing, acoustic sensor technology, and research in the laboratory stage. There are few studies on the characteristics of natural gas pipeline leakage sound sources and the propagation law of leakage sound waves along the pipeline. In addition, most of the study focuses on the liquid delivery pipeline, but it is invalid to the gas pipeline. There are still many difficulties in the generation, propagation, attenuation, and reception of acoustic waves in gas pipeline leakage, which limit the popularization of acoustic wave leakage detection method in natural gas pipelines. Thus, based on previous studies, acoustic leak detection is studied by the combination of simulation and experimental methods in this work. The propagation law of the sound field in the leakage of the high-pressure natural gas pipeline is analyzed. This is of great significance to the implementation and popularization of acoustic waves leakage detection technology.