A gas pipeline leakage positioning method based on wavelet packet transform and coherence function

Abstract

This invention discloses a gas pipeline leak location method based on wavelet packet transform and coherence function, comprising the following steps: for the characteristic frequency range of the acquired signal, by calculating the coherence function between signals, a frequency range greater than a threshold is selected as the effective frequency bandwidth; the acquired leak sound signal is decomposed into multiple scales using wavelet packet transform, and based on the effective frequency bandwidth, appropriate frequency components are selected for reconstruction to obtain the target signal; the cross-correlation function between the target functions is calculated to extract time delay information, and the influence of temperature on the propagation speed of the leak sound signal is considered and corrected; the leak point is located by combining the time delay, thus improving the accuracy of leak location. According to this invention, the signal-to-noise ratio of the signal is effectively improved, and it is applied to the problem of locating gas pipeline leak points in complex environmental conditions. Combined with the corrected propagation speed of the leak sound signal, the accuracy of locating gas pipeline leak points is improved, which is of great significance to urban public safety.

Description

Technical Field

[0001] This invention relates to the technical field of gas pipeline leak location, and in particular to a gas pipeline leak location method based on wavelet packet transform and coherence function. Background Technology

[0002] As a transportation medium for essential urban supplies such as water and natural gas, pressurized pipelines are widely used in lifeline systems like water supply and gas systems, providing material support for the normal operation and development of society. However, the pipeline laying environment is complex, and most older pipelines have been in service for a long time. Due to aging, corrosion, and third-party damage, gas pipelines frequently leak, resulting in significant energy waste and potential threats to personnel and property safety. Therefore, accurately locating gas pipeline leaks and inspecting and repairing them before potential hazards occur is of great significance to urban public safety.

[0003] Gas pipelines only have valve wells or localized exposed openings at various stations that can be directly accessed. Localized point detection methods, such as acoustic-based gas pipeline leak location methods, show promising development prospects. These methods primarily use cross-correlation estimation to determine the time delay of the leak sound signal propagating along the pipeline to sensors deployed upstream and downstream of the leak point, and combine this with the propagation speed of the leak sound signal to locate the leak point. It is evident that the propagation speed of the leak sound wave and the accuracy of the time delay estimation determine the accuracy of the leak point location. However, under complex and variable environmental noise interference, the signal-to-noise ratio of the leak sound signal is low, resulting in poor accuracy of the time delay estimation. Furthermore, accurately and reasonably selecting an appropriate propagation speed for the leak sound signal also determines the accuracy of the leak point location. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the present invention aims to provide a gas pipeline leak location method based on wavelet packet transform and coherence functions. This method analyzes the coherence functions between signals collected by sensors deployed upstream and downstream of the leak point, selects the effective frequency bandwidth of the leak sound signal, extracts the leak sound signal using wavelet packet transform to improve the signal-to-noise ratio, estimates the time delay of the extracted leak sound signal using cross-correlation analysis, and accurately locates the leak point by combining this with the corrected propagation velocity of the leak sound signal. To achieve the above-mentioned objectives and other advantages of the present invention, a gas pipeline leak location method based on wavelet packet transform and coherence functions is provided, comprising: S1. Obtain the leakage sound signals upstream and downstream of the leakage point through the sensor, and select the effective frequency bandwidth of the signal through a pre-set threshold. S2. Decompose the leakage sound signals upstream and downstream of the leakage point by wavelet packet transform, and reconstruct the target leakage sound signal based on the effective frequency bandwidth by selecting appropriate frequency band components. S3. Calculate the cross-correlation function of the reconstructed leakage sound signal. The time delay corresponding to the peak value of the function is the time delay of the leakage sound signal propagating to the two sensors. S4. Calculate the cutoff frequency of the plane wave in the gas pipeline and estimate the propagation speed of the leakage sound signal in the pipeline based on the ambient temperature during signal acquisition. S5. Calculate the distance between the leak point and the sensor according to the leak point location formula.

[0005] Preferably, step S1 specifically involves acquiring leakage sound signals upstream and downstream of the leakage point using a sound pressure sensor. and ， The distances from the sensor to the leak point are respectively and Calculate signal and The coherence function selects the effective frequency bandwidth of the signal by using a pre-set threshold.

[0006] Preferably, the signal is estimated using Welch. and power spectral density and Further calculation of the signal and coherence function Based on signals and coherence function Select a value greater than the set threshold. frequency range This serves as the effective frequency bandwidth for this group of signals.

[0007] Preferably, step S2 specifically involves selecting 'sym17' as the basis function for the wavelet transform, and the number of decomposition levels is determined based on the signal's sampling rate and effective frequency bandwidth. To confirm; Based on effective frequency bandwidth The selected frequency band components after wavelet packet decomposition belong to the effective frequency bandwidth. The coefficients within the range are used to reconstruct the target signal. and .

[0008] Compared with the prior art, the advantages and positive effects of the present invention are: Based on the coherence between the collected leakage sound signals, the effective frequency bandwidth of the leakage sound signals was selected. Wavelet packet transform was used to perform multi-scale decomposition of the signals, and effective frequency band components were selected in combination with the effective frequency bandwidth. The target leakage sound signal was then reconstructed, reducing the interference of environmental noise and improving the signal-to-noise ratio. The time delay information was obtained by calculating the cross-correlation function between the target signals. At the same time, the influence of temperature on the propagation speed of the leakage sound signal was considered, and the propagation speed was corrected, thereby improving the accuracy of leakage location. Attached Figure Description

[0009] Figure 1 A flowchart of the gas pipeline leak location method based on wavelet packet transform and coherence function according to the present invention; Figure 2 The signal of the gas pipeline leak location method based on wavelet packet transform and coherence function according to the present invention and A schematic diagram of the coherence function and effective frequency bandwidth; Figure 3 The wavelet packet decomposition tree diagram is shown for the gas pipeline leak location method based on wavelet packet transform and coherence function according to the present invention. Figure 4 The image shows the cross-correlation function between the original signals in the gas pipeline leak location method based on wavelet packet transform and coherence function according to the present invention. Figure 5 The image shows the cross-correlation function between signals after effective frequency bandwidth bandpass filtering, according to the gas pipeline leak location method based on wavelet packet transform and coherence function of the present invention. Figure 6 This is a graph showing the cross-correlation function between target signals extracted according to the gas pipeline leak location method based on wavelet packet transform and coherence function of the present invention. (Detailed implementation details follow.) The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0010] This invention collects signals by deploying sound pressure sensors upstream and downstream of the leak point; calculates the coherence function between the collected signals, and selects the frequency range where the coherence function value is greater than a preset threshold as the effective frequency bandwidth of the signal set; performs multi-scale decomposition on the two sets of leakage sound signals through wavelet packet transform, extracts wavelet coefficients within the effective frequency bandwidth, and reconstructs the target signal; uses cross-correlation analysis to estimate the time delay of the reconstructed signal; and, considering the influence of temperature, corrects the propagation speed of the leakage sound signal, combining the time delay to locate the leak point.

[0011] Reference Figure 1 A gas pipeline leak localization method based on wavelet packet transform and coherence function includes: Step 1: Use a sound pressure sensor to acquire the leakage sound signals upstream and downstream of the leak point. and The distances from the sensor to the leak point are respectively and Calculate signal and The coherence function selects the effective frequency bandwidth of the signal by using a pre-set threshold; The acoustic signal of a gas pipeline leak was collected using the gas pipeline leak detection and location test platform at Tongji University. The gas pipeline consisted of a DN150 spiral steel pipe with a wall thickness of 4.5mm, and the pipeline pressure ranged from 0.1 to 0.3 MPa. Compressed air was used to simulate gas, and a gas leak was simulated by directly drilling a hole in a replaceable standard pipe section. The distance between the sensor and the leak point was [missing information]. and The sampling frequency was 4000Hz and the acquisition time was 2min.

[0012] 1) Estimating the signal using Welch and power spectral density and Further calculation of the signal and coherence function :

[0013] in It is a signal and Cross-spectral density; 2) Based on signal and coherence function Select values ​​greater than a given threshold. frequency range This serves as the effective frequency bandwidth of the signal group. Figure 2Given a set of measured data, the coherence function is given, and a threshold value for the coherence function is set. The search selects a frequency range where the coherence function value is greater than 0.1. Based on this, the effective frequency bandwidth selected in this embodiment is 10Hz~140Hz. Step 2: Decompose the signal using wavelet packet transform. and Based on the effective frequency bandwidth selected in step 1, appropriate frequency band components are selected to reconstruct the target leakage sound signal. and ; Figure 3 A schematic diagram of a three-layer wavelet packet decomposition tree is given. In each layer of the decomposition tree, the low-frequency and high-frequency components are decomposed simultaneously. The signal is finally decomposed into eight frequency band components, A4~A7 and D4~D7. Different frequency band components contain different frequency components of the original signal. Based on the effective frequency bandwidth selected in step 1, only the frequency band components falling within the effective frequency bandwidth are selected for reconstruction to obtain the target signal, while other interference signals are filtered out.

[0014] 1) The basis function for wavelet transform is 'sym17', and the number of decomposition levels is determined based on the signal's sampling rate and effective frequency bandwidth. To confirm:

[0015] in Sampling rate, This is the floor operator; 2) Based on effective frequency bandwidth The selected frequency band components after wavelet packet decomposition belong to the effective frequency bandwidth. The coefficients within the range are used to reconstruct the target signal. and ; Step 3, Calculate the target signal and cross-correlation function The time delay corresponding to the peak value of the function is the time delay for the leakage sound signal to propagate to the two sensors. ; Figure 4 The method for directly calculating the original signal is given. and The cross-correlation coefficient results show that the peak cross-correlation coefficient is 0.24, indicating a low signal-to-noise ratio and high susceptibility to environmental noise. Figure 5The results of calculating the cross-correlation coefficients between signals after bandpass filtering the original signal within its effective frequency bandwidth are presented. Compared to directly calculating the cross-correlation coefficients from the original signal, the peak value of the cross-correlation coefficient obtained by this method is 0.47, showing a significant improvement. Furthermore, the method based on wavelet packet transform and coherence function proposed in this invention processes the original signal to obtain the target signal. and And calculate their cross-correlation coefficients, the results are as follows Figure 6 As shown, the peak value of the cross-correlation coefficient reached 0.61, indicating a further improvement, corresponding to a lower latency. As can be seen, the method proposed in this invention further improves the signal-to-noise ratio and reduces the interference of environmental noise, thus achieving better time delay estimation accuracy even when the signal-to-noise ratio is low.

[0016] Step 4, the cutoff frequency of the plane wave in the gas pipeline is:

[0017] in Where is the radius of the gas pipeline. This refers to the propagation speed of the leakage sound signal; for commonly used gas pipelines, The sound velocity of a gas leak can reach several kiloHz. In gas pipelines, the sound signal primarily propagates in the form of a plane wave. This propagation speed is mainly affected by ambient temperature; the propagation speed of the leak sound signal within the pipeline can be estimated based on the ambient temperature at the time of signal acquisition. :

[0018] in The absolute temperature (K) The adiabatic index of the gas is . The gas constant is 8.314 J / mol K. The value is the molar mass of the gas (kg / mol); in this embodiment, the gas is air, the ambient temperature is 25°C, and the calculated corrected propagation speed is 346 m / s.

[0019] Step 5: Calculate the distance from the leak point to the sensor according to the leak point location formula. and :

[0020]

[0021] in The distance between sensor 1 and sensor 2 is given.

[0022] Based on the delay obtained in step 3 The propagation speed obtained in step 4 The calculated distances from the leak point to the sensor are as follows: and The absolute error between the actual distance and the actual distance is within 0.2m, which fully demonstrates the effectiveness of the method designed in this invention.

[0023] In summary, this invention extracts time delay information by calculating the cross-correlation function between objective functions, and corrects for the effect of temperature on the propagation speed of the leakage sound signal. Combining this with the time delay analysis improves the accuracy of leak location. This invention effectively improves the signal-to-noise ratio, making it applicable to locating gas pipeline leaks in complex environments. Furthermore, by incorporating the corrected propagation speed of the leakage sound signal, it further enhances the accuracy of gas pipeline leak location, which is of great significance to urban public safety.

[0024] The number of devices and processing scale described herein are for simplification purposes. Applications, modifications, and variations of this invention will be readily apparent to those skilled in the art. Although embodiments of the invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for this invention, and further modifications can be readily implemented by those skilled in the art. Therefore, without departing from the general concept defined by the claims and their equivalents, this invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A method for locating gas pipeline leaks based on wavelet packet transform and coherence function, characterized in that, Includes the following steps: S1. Obtain the leakage sound signals upstream and downstream of the leakage point through the sensor, and select the effective frequency bandwidth of the signal through a pre-set threshold. S2. Decompose the leakage sound signals upstream and downstream of the leakage point by wavelet packet transform, and reconstruct the target leakage sound signal based on the effective frequency bandwidth by selecting appropriate frequency band components. S3. Calculate the cross-correlation function using the target leakage sound signal. The time delay corresponding to the peak value of the function is the time delay for the leakage sound signal to propagate to the two sensors. S4. Calculate the cutoff frequency of the plane wave in the gas pipeline and estimate the propagation speed of the leakage sound signal in the pipeline based on the ambient temperature during signal acquisition. S5. Calculate the distance between the leak point and the sensor according to the leak point location formula.

2. The gas pipeline leak location method based on wavelet packet transform and coherence function as described in claim 1, characterized in that, Specifically, step S1 involves acquiring leakage sound signals upstream and downstream of the leakage point using a sound pressure sensor. and ， The distances from the sensor to the leak point are respectively and Calculate signal and The coherence function selects the effective frequency bandwidth of the signal by using a pre-set threshold.

3. The gas pipeline leak location method based on wavelet packet transform and coherence function as described in claim 2, characterized in that, Estimating signals using Welch and power spectral density and Further calculation of the signal and coherence function Based on signals and coherence function Select a value greater than the set threshold. frequency range This serves as the effective frequency bandwidth for this group of signals.

4. The gas pipeline leak location method based on wavelet packet transform and coherence function as described in claim 2, characterized in that, Step S2 specifically involves selecting the wavelet transform basis function 'sym17', and determining the number of decomposition levels based on the signal's sampling rate and effective frequency bandwidth. To confirm; Based on effective frequency bandwidth The selected frequency band components after wavelet packet decomposition belong to the effective frequency bandwidth. The coefficients within the range are used to reconstruct the target signal. and .

Images