Gas pipeline network leakage real-time detection and positioning method based on edge computing and application thereof

Abstract

The application provides a gas pipeline network leakage real-time detection and positioning method based on edge computing and application thereof, aiming at the problems of poor adaptability of a fixed analysis window and high false alarm rate of a pure data-driven model in the prior art, the method synchronously collects pressure and acoustic emission signals, and adaptively generates a dynamic analysis window based on an energy accumulation rate and a pulse density index; a frequency band constraint based on physical characteristics of pipe materials is introduced to extract features in the window, and a multi-level physical proof verification of wave speed-medium space-time matching and sound-pressure homologous coupling mutual checking is performed, and positioning calculation is performed according to a signal arrival time difference after verification. The application realizes high-sensitivity capture of slight leakage, low-delay response of burst leakage and high-trustworthiness filtering of interference events, and is suitable for real-time detection and accurate positioning of gas pipeline network leakage.

Description

Technical Field

[0001] This invention relates to the fields of gas pipeline safety monitoring and industrial Internet of Things edge computing technology, and in particular to a method for real-time detection and location of gas pipeline leaks based on edge computing and its application. Background Technology

[0002] With the continuous expansion of urban gas pipeline networks, distributed leak monitoring technology based on edge computing has gradually become a research hotspot. In typical existing solutions, the edge computing gateway periodically collects pipeline pressure and leak concentration data. After filtering outomas using statistical thresholds, a pre-trained neural network model is used for state assessment, and an alarm is uploaded to the cloud platform when a leak is detected. Other solutions employ infrasound, continuously collecting infrasound signals at both ends of the pipeline. The denoised signals undergo empirical mode decomposition and multi-scale entropy feature extraction. Based on a trained classification model, a leak detection result is output, and the leak point is roughly located based on the signal arrival time difference. All of these solutions, to a certain extent, achieve localized detection and cloud-based collaboration for gas pipeline leaks.

[0003] Therefore, there is an urgent need for a real-time detection and location method for gas pipeline leaks based on edge computing that integrates dynamic signal perception and pipeline physical transmission verification, and its application, in order to solve the problems existing in the current technology. Summary of the Invention

[0004] This invention provides a real-time detection and location method for gas pipeline leaks based on edge computing and its application. It addresses the problems of existing technologies that use a fixed data analysis window, resulting in high false alarm rates for minor leaks and large response delays for sudden leaks. Furthermore, the pure data-driven judgment mode cannot distinguish between actual leaks within the pipeline and external environmental vibrations or operating condition interference from the perspective of physical mechanisms, leading to persistently high false alarm rates.

[0005] The core technology of this invention is to construct an adaptive analysis window driven by both energy accumulation rate and transient pulse density, and combine it with a spatiotemporal consistency verification mechanism based on the physical constraints of pipeline stress wave propagation, so as to realize dynamic matching perception of leakage signals of different intensities and multi-level physical falsification of interference events at the edge side.

[0006] In a first aspect, the present invention provides a method for real-time detection and location of gas pipeline leaks based on edge computing, applied to edge computing gateways deployed at upstream and downstream nodes of gas pipelines, comprising the following steps:

[0007] Simultaneously acquire and preprocess dynamic pressure sensor signals and broadband acoustic emission sensor signals deployed on the pipeline; The energy accumulation rate of the dynamic pressure sensor signal and the transient pulse density of the broadband acoustic emission sensor signal are calculated in real time. A dual triggering factor is generated based on the energy accumulation rate and transient pulse density. The length of the analysis window is dynamically adjusted according to the dual triggering factor to obtain an adaptive analysis window. Within the adaptive analysis window, multi-scale time-frequency features are extracted from dynamic pressure sensor signals and broadband acoustic emission sensor signals. The extracted multi-scale time-frequency features are used to make a preliminary judgment on leakage. For events that are judged to be suspected of leakage, a spatiotemporal consistency verification based on the physical constraints of pipeline stress wave propagation is performed to filter out interference events that do not conform to the physical conduction law. For real leakage events that pass the spatiotemporal consistency verification, the location of the leakage point is calculated based on the signal arrival time difference between upstream and downstream edge computing gateways and the physical parameters of the pipeline, and the location results and confidence level are reported to the cloud platform.

[0008] Furthermore, in the step of generating the dual triggering factor, the dual triggering factor is obtained by weighting and summing the normalized energy accumulation rate and the normalized transient pulse density based on their respective preset weighting coefficients; the length of the adaptive analysis window is continuously and smoothly adjusted according to the value of the dual triggering factor through a preset nonlinear mapping function, so that the window length converges to the preset minimum length when the dual triggering factor increases, and expands to the preset maximum length when the dual triggering factor decreases.

[0009] Furthermore, the nonlinear mapping function is configured as a smooth transition function with an S-shaped response curve; the smooth transition function takes dual triggering factors as input parameters and dynamically maps the window length in combination with a preset center threshold and steepness coefficient; Among them, the center threshold is configured as the critical benchmark for determining window contraction and expansion, and the steepness coefficient is configured as the sensitivity of the adaptive analysis window length to changes in the dual triggering factors.

[0010] Furthermore, in the step of extracting multi-scale time-frequency features, the target analysis frequency band of the broadband acoustic emission sensor signal is adaptively selected based on the pipe material information, and the wavelet packet energy ratio within the target analysis frequency band is calculated as the key feature component.

[0011] Furthermore, spatiotemporal consistency verification includes a wave velocity matching verification step: The theoretical stress wave propagation velocity is calculated based on the pipe material, diameter, wall thickness, and gas density between the upstream and downstream edge calculation gateways. Extract the arrival time difference of the negative pressure wave from the dynamic pressure sensor signal and calculate the apparent propagation distance; When the deviation between the apparent propagation distance and the actual pipeline length between the upstream and downstream edge computing gateways is less than the preset tolerance threshold, the wave speed matching verification is deemed successful.

[0012] Furthermore, the spatiotemporal consistency verification includes the acoustic-pressure homogeneous coupling verification step: Calculate the correlation coefficient between the envelope sequence of the broadband acoustic emission sensor signal and the first-order difference sequence of the dynamic pressure sensor signal within the adaptive analysis window; When the maximum value of the correlation coefficient exceeds the preset coupling threshold, the acoustic-pressure co-source coupling verification is deemed successful.

[0013] Furthermore, in the step of calculating the location of the leak point, when the pipe material is a viscoelastic material, the dispersion correction factor corresponding to the viscoelastic material is obtained, and the theoretical stress wave propagation velocity is corrected using the dispersion correction factor. The location of the leak point is calculated by combining the corrected wave velocity with the signal arrival time difference.

[0014] Secondly, the present invention provides a real-time detection and location device for gas pipeline network leaks based on edge computing, comprising: The signal acquisition module is used to simultaneously acquire and preprocess signals from dynamic pressure sensors and broadband acoustic emission sensors deployed on the pipeline. The dynamic window generation module is used to calculate the energy accumulation rate of the dynamic pressure sensor signal and the transient pulse density of the broadband acoustic emission sensor signal in real time. It generates a dual triggering factor based on the energy accumulation rate and transient pulse density, and dynamically adjusts the length of the analysis window according to the dual triggering factor to obtain an adaptive analysis window. The feature extraction module is used to extract multi-scale time-frequency features from dynamic pressure sensor signals and broadband acoustic emission sensor signals within an adaptive analysis window. The physical verification module is used to make a preliminary judgment on the extracted multi-scale time-frequency features. For events judged as suspected leaks, it performs spatiotemporal consistency verification based on the physical constraints of pipeline stress wave propagation to filter out interference events that do not conform to the physical conduction law. The positioning and communication module is used to calculate the location of the leak point based on the signal arrival time difference between upstream and downstream edge computing gateways and the physical parameters of the pipeline for real leak events that have passed spatiotemporal consistency verification, and to report the positioning results and confidence level to the cloud platform.

[0015] Thirdly, the present invention provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to execute the above-described method for real-time detection and location of gas pipeline leaks based on edge computing.

[0016] Fourthly, the present invention provides a readable storage medium storing a computer program, the computer program including program code for controlling a process to execute the process, the process including the above-described edge computing-based real-time detection and location method for gas pipeline leaks.

[0017] The main contributions and innovations of this invention are as follows: 1. By calculating the energy accumulation rate of dynamic pressure signals and the transient pulse density of broadband acoustic emission signals in real time, dual triggering factors are generated and the analysis window length is dynamically adjusted. This allows the window to automatically expand to fully accumulate weak characteristic energy when facing small, slow leaks, and to rapidly shrink to accurately capture transient impacts when facing sudden, large leaks. This breaks through the limitations of the fixed window mechanism, significantly reduces the false alarm rate of small leaks, and improves the real-time response to sudden leaks.

[0018] 2. In the feature extraction stage, a frequency band constraint based on the physical properties of the pipe is introduced. The target guided wave dispersion band of the acoustic emission signal is adaptively selected according to the pipe material and the wavelet packet energy ratio is calculated. This makes the extracted time-frequency features directly related to the physical nature of the leakage, enhancing the noise resistance and generalization performance of the subsequent judgment model.

[0019] 3. Perform spatiotemporal consistency verification based on the physical constraints of pipeline stress wave propagation, including wave velocity-medium spatiotemporal rigidity matching verification and sound-pressure co-source coupling verification. Utilizing the physical law that negative pressure waves and high-frequency acoustic emissions are strictly synchronized in real leakage events, as well as the constraint of theoretical wave velocity on apparent propagation distance, it can effectively filter out interference events that only meet the data anomaly conditions but do not conform to the physical conduction law of pipelines, such as pressure regulator operation, large flow gas consumption, and vehicle crushing, from the mechanism level, which can significantly reduce the false alarm rate of the system.

[0020] 4. The edge computing gateway only performs deep feature extraction and localization calculations after the dynamic window is triggered and multi-level physical verification is passed. In daily monitoring, it only needs to run low-complexity trigger factor calculations, which significantly reduces the average power consumption and computing load of edge nodes. At the same time, it only uploads confirmed leakage events and related confidence scores to the cloud, which greatly reduces communication bandwidth usage and is suitable for battery-powered long-term deployment scenarios in the field.

[0021] Details of one or more embodiments of the present invention are set forth in the following drawings and description, so that other features, objects and advantages of the invention will be more readily understood. Attached Figure Description

[0022] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is the overall architecture of a gas pipeline network leak real-time detection and location system based on edge computing according to an embodiment of the present invention; Figure 2 This is a logic flowchart of the adaptive adjustment of the dynamic multi-scale sensing window according to an embodiment of the present invention; Figure 3 This is a decision flowchart for spatiotemporal collaborative verification of the physical conduction chain according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with some aspects of one or more embodiments of this specification as detailed in the appended claims.

[0024] It should be noted that the steps of the corresponding methods are not necessarily performed in the order shown and described in this specification in other embodiments. In some other embodiments, the methods may include more or fewer steps than described in this specification. Furthermore, a single step described in this specification may be broken down into multiple steps in other embodiments; and multiple steps described in this specification may be combined into a single step in other embodiments.

[0025] System Architecture and Workflow Overview Figure 1 This paper illustrates the overall architecture of the real-time gas pipeline leak detection and location system based on edge computing, as described in this invention. The system comprises a perception layer, an edge computing layer, and a cloud platform layer. The perception layer consists of multiple monitoring nodes deployed at intervals along the gas pipeline network. Each monitoring node is equipped with a dynamic pressure sensor and a broadband acoustic emission sensor, used to sense changes in fluid pressure inside the pipeline and high-frequency stress wave signals from the pipe wall, respectively. The edge computing layer consists of edge computing gateways deployed at each monitoring node. Each gateway independently runs the leak detection and location method described in this invention, and adjacent gateways exchange data and synchronize time via wireless communication modules. The cloud platform layer receives leak alarm events and related confidence data reported by the edge computing gateways, providing information for maintenance personnel to make decisions and schedule operations.

[0026] Figure 2The logical flow of the adaptive adjustment of the dynamic multi-scale sensing window in the method of this invention is illustrated. The edge computing gateway continuously receives and buffers preprocessed dynamic pressure signals and broadband acoustic emission signals in a circular buffer. It calculates the dual triggering factors in units of a basic observation step size and dynamically adjusts the analysis window length according to the real-time values ​​of the triggering factors through a preset mapping function. When the triggering factor is below the center threshold, the window gradually expands to its maximum length to accumulate weak leakage characteristics; when the triggering factor is above the center threshold, the window rapidly shrinks to its minimum length to capture transient impact details.

[0027] Figure 3 The decision-making process for spatiotemporal collaborative verification of the physical conduction chain in the method of this invention is illustrated. For events initially determined to be suspected leaks, wave velocity-medium spatiotemporal rigidity matching verification and acoustic-pressure co-source coupling verification are performed sequentially. Only when both levels of verification pass can the leak point location calculation stage proceed and the results be reported to the cloud; if either verification fails, the event is marked as operating condition interference or environmental vibration, and no leak alarm is reported, but only the event log is recorded locally.

[0028] Example 1 like Figure 1-3 As shown, this embodiment uses a medium-pressure gas pipeline with a diameter of DN200 as the monitoring object. The pipeline material is PE100, the standard size ratio is SDR11, and the actual length D of the pipeline between two adjacent monitoring nodes A and B is... AB The distance is 500 meters. An edge computing gateway is deployed at node A and node B respectively. Each gateway connects to a dynamic pressure transmitter and a broadband acoustic emission sensor. The dynamic pressure transmitter has a range of 0–1 MPa and a sampling rate of 100 Hz; the broadband acoustic emission sensor has a frequency response range of 5 kHz–300 kHz and a sampling rate of 20 kHz.

[0029] Step S1: Synchronous acquisition and preprocessing of heterogeneous multimodal signals The edge computing gateways at nodes A and B maintain microsecond-level time synchronization via a GPS timing module, synchronously acquiring dynamic pressure sensor signal P(t) and broadband acoustic emission sensor signal A(t) according to a set sampling rate. During the acquisition process, wavelet threshold denoising is performed on P(t) and A(t) respectively. Specifically, the Daubechies wavelet basis db6 is selected to perform a 5-level wavelet decomposition on the original signal, and a soft thresholding function is used to shrink the detail coefficients at each level, with a threshold of [value missing]. Where σ is the estimated standard deviation of noise and N is the signal length. The denoised signal is stored in a circular buffer for subsequent modules to use.

[0030] Step S2: Adaptive generation of dynamic window based on energy-pulse dual triggering Within the circular buffer, the dual triggering factor is continuously calculated at intervals of the basic observation step size Δτ = 1 second.

[0031] Short-time energy accumulation rate The calculation formula is:

[0032] In practical engineering implementation, the difference approximation is used for calculation:

[0033] in This represents the signal energy within the current basic observation step.

[0034] Transient pulse density index The calculation method is as follows: First, determine the adaptive noise floor threshold based on the root mean square value of the acoustic emission signal during historical calm periods. Typically, the root mean square value of the signal during the quiet period is taken as 3 to 4 times; then, the acoustic emission signal A(t) positively crosses + within the current basic observation step size Δτ. And negative travel- Complete number of peaks Pulse density index ,in The sampling rate of the acoustic emission signal.

[0035] right and Normalization is performed. This embodiment uses a sliding window max-min normalization method: within a 60-second historical sliding window, records... and The maximum value ( , ) and minimum value ( , The following formula is used to linearly map the data to the interval [0, 1]:

[0036]

[0037] Historical extreme values ​​are updated using an exponentially weighted moving average to accommodate the slow drift of pipeline operating conditions.

[0038] Dual triggering factors Defined as the weighted sum of the normalized energy accumulation rate and the pulse density exponent: In this embodiment, considering that the PE pipe is sensitive to slow pressure changes and the environmental noise is relatively stable, the weighting coefficients are α=0.5 and β=0.5, which satisfy α+β=1.

[0039] Dynamic analysis window length The calculation is performed using a smooth transition function with an S-shaped response curve. In this embodiment, a Sigmoid function is used.

[0040] The parameters are set as follows: Maximum window length A 20-second window is used to fully accumulate the weak energy characteristics of minute leaks; minimum window length. A timeout of 0.2 seconds is chosen to ensure transient detection capability during sudden large leaks; the steepness coefficient k is set to 10 to ensure moderate transition sensitivity of the window length near the center threshold; the center threshold γ is set to 0.5. When the value is below 0.3, the window length is close to ;when When the value is above 0.7, the window length is close to Smooth transition within the intermediate range.

[0041] Step S3: Physical Prior-Guided Time-Frequency Feature Extraction Using the current moment as the endpoint, extract a length forward. Signal fragments within a dynamic window and Feature extraction is performed.

[0042] For broadband acoustic emission signals The target analysis frequency band is adaptively selected based on the pipe material information recorded in the pipeline GIS system. In this embodiment, the pipe material is PE pipe, and the target guided wave dispersion band is selected as 10kHz to 40kHz. A db6 wavelet basis is used. Perform 5-level wavelet packet decomposition, calculate the wavelet packet energy ratio of each sub-band within the target frequency band, and take the normalized energy values ​​of the three sub-bands with the highest energy ratio to form the characteristic components.

[0043] For dynamic pressure signals The inflection point features of the negative pressure wave slope and the duration features of the falling edge of the waveform are extracted. The inflection point features of the negative pressure wave slope are calculated... The second-order difference sequence Diff 2 The zero-crossing position of [P(t)] is obtained, and the ratio of the change amplitude of the first-order difference before and after the zero-crossing point is recorded as an indicator of the inflection point strength. The duration characteristic of the falling edge of the waveform is obtained in the following way: Identification The local maximum value is recorded, and the moment when the signal drops by 3% from that maximum value is recorded. And the moment when the signal recovers to 50% of that decline or reaches a new steady state. Duration characteristics .

[0044] The feature components extracted from the acoustic emission and pressure signals are concatenated to form a multi-scale time-frequency feature vector, which is then input into a pre-trained lightweight classification model for preliminary leak detection. The classification model can be an Extreme Learning Machine or a lightweight convolutional neural network, outputting a probability value indicating a suspected leak. .

[0045] Step S4: Spatiotemporal Coordination Credibility Verification Based on Physical Transmission Chain when When the threshold exceeds the preset initial judgment threshold (e.g., 0.6), the multi-level physical falsification verification process is initiated.

[0046] First, a wave velocity-medium spatiotemporal rigidity matching verification is performed. The arrival times of the falling edge of the pressure signal at nodes A and B are calculated using the generalized cross-correlation algorithm. and The arrival time difference of the negative pressure wave is obtained. Based on the pipeline network GIS system, the material (PE100), nominal outer diameter D=200mm, wall thickness δ=11.9mm (corresponding to SDR11), and gas density of the pipeline between nodes A and B are obtained. At 20℃ and 0.4MPa, the pressure is approximately 2.8 kg / m³. 3 The bulk modulus of elasticity K = 1.42 × 10⁻⁶ 6 Pa. The elastic modulus of PE100 material is E = 1.1 × 10⁻⁶. 9 Pa, Poisson's ratio ν = 0.4, pipe constraint coefficient C is taken as 1 - ν / 2 = 0.8 for thin-walled pipes. Calculate the theoretical wave velocity based on the stress wave propagation theory formula. :

[0047] Considering that PE pipe, as a viscoelastic material, exhibits a significant dispersion effect (i.e., wave velocity varies with frequency), a dispersion correction factor needs to be introduced. The theoretical wave velocity is corrected. In this embodiment, the frequency dispersion characteristics of SDR11 and PE100 pipes at room temperature are looked up in a table. Taking 0.92, the corrected theoretical wave velocity is... ≈ 424 m / s. Calculate the apparent propagation distance. Set tolerance threshold The actual length of the pipe 5%, or 25 meters. Set a tolerance threshold. The actual length of the pipe 5%, or 25 meters. If If so, the wave speed matching verification is passed.

[0048] Next, perform acoustic-pressure homogeneous coupling verification. This involves verifying the broadband acoustic emission signals of node A (or node B) within the dynamic window. Perform Hilbert transform to extract the envelope sequence. , where Â(t) is the Hilbert transform result of A(t). For dynamic pressure signals within the same time period... The first-order difference sequence Diff(t) = P(t) - P(t-1 / fs) is calculated, with the difference step size consistent with the sampling interval of the pressure signal to highlight the variation characteristics of the pressure falling edge.

[0049] The dynamic window is evenly divided into 10 sliding sub-windows, each sub-window having a length of [missing information]. / 10, step size / 20 (meaning there is a 50% overlap between sub-windows). Calculate separately for each sub-window. The Pearson correlation coefficient with Diff(t) is used to obtain a correlation coefficient sequence. The maximum value of this sequence is taken as the acoustic-pressure synchronization coupling degree. In this embodiment, a coupling threshold is set. It is 0.65. When When the value is greater than 0.65, the acoustic-pressure homogeneous coupling verification is passed.

[0050] If both of the above verifications pass, a real leakage event is confirmed, and the process proceeds to step S5 for location calculation. If either verification fails, the event is marked as "operating condition interference" or "environmental vibration," and the event log is only recorded locally on the edge gateway without reporting a leakage alarm.

[0051] Step S5: Edge-side localization calculation and confidence reporting The arrival time difference of the negative pressure wave will be verified. To improve the accuracy of time difference calculation, a generalized cross-correlation time delay estimation is performed. The corrected theoretical wave velocity is used. The distance from the leak point to node A is calculated using the following formula:

[0052] Normalize the deviation value in wave speed matching verification The normalized coupling degree C2 in the acoustic-pressure coupling verification is equal to... The overall confidence level is obtained by weighted fusion. = 0.4·C1 + 0.6·C2. Locate the leak point. Overall confidence level The feature vector summary is reported to the cloud platform via the wireless communication module.

[0053] In the simulated micro-corrosion perforation leakage scenario of this embodiment, the initial pressure drop rate is approximately 0.1 kPa / min. Traditional fixed-window methods cannot detect the anomaly within a 10-second window. The method of this invention, after continuous monitoring for approximately 120 seconds, achieves a lower energy accumulation rate. Continuously exceeding baseline, dual triggering factors The dynamic window automatically expanded to approximately 15 seconds, reaching 0.72, successfully capturing the characteristic of the acoustic emission signal's energy percentage jumping from 8% to 32% in the 10kHz–40kHz frequency band. In the wave velocity matching verification, the apparent propagation distance was 478 meters, deviating by 4.4% from the actual distance of 500 meters; the acoustic-pressure coupling degree... The value reached 0.78, exceeding the threshold. Ultimately, the system issued a high-confidence alarm, indicating that the positioning error was less than 1% of the total pipeline length.

[0054] Example 2 This embodiment uses the same system architecture and hardware configuration as Embodiment 1. The difference lies in that the monitoring object is a section of DN300 medium-pressure steel pipe, the pipe material is X52 steel grade, the wall thickness is 8mm, and the distance between adjacent nodes is 800 meters. This embodiment focuses on demonstrating the detection effect on sudden and strong leaks and the differences in parameter configuration under steel pipe scenarios.

[0055] In step S2, considering the high sound propagation speed and wider acoustic emission signal bandwidth in steel pipes, the parameter configuration is adjusted as follows: the basic observation step size Δτ is shortened to 0.5 seconds to adapt to the characteristics of fast stress wave propagation speed and more rapid signal changes in steel pipes. Adaptive noise floor threshold. The root mean square value of the acoustic emission signal during the calm period is set to four times to accommodate the relatively high background vibration level in the steel pipe environment. The weighting coefficients are adjusted to α=0.6 and β=0.4, slightly emphasizing the response to the energy accumulation rate, as energy changes are extremely drastic during sudden large leaks. The steepness coefficient k is set to 15 to ensure a faster response to drastic changes in the triggering factor. Maximum window length... Minimum window length (15 seconds) Take 0.1 seconds.

[0056] In step S3, based on the fact that the pipe material is steel, the target guided wave dispersion band is adaptively selected to be 100kHz to 200kHz. A 6-layer wavelet packet decomposition using the db8 wavelet basis is employed to accommodate the richer high-frequency components in the steel pipe signal.

[0057] In step S4, the material (X52 steel), nominal outer diameter D=300mm, wall thickness δ=8mm, and gas density at 20℃ and 0.8MPa are obtained from the pipeline GIS system. =5.6kg / m 3 The bulk modulus of elasticity K = 2.1 × 10⁻⁶6 Pa. The elastic modulus of X52 steel is E = 2.1 × 10⁻⁶. 11 Pa, Poisson's ratio ν = 0.3, pipe constraint coefficient C is taken as 1 - ν / 2 = 0.85 for thin-walled pipes. The theoretical wave velocity is calculated based on the theoretical formula for pressure wave velocity within the pipe. ≈612.3 m / s. The dispersion effect of steel pipe is much weaker than that of PE pipe, with a dispersion correction factor η. steel Take 0.98, and adjust the wave velocity. = ≈ 600m / s. Set tolerance threshold. The actual length of the pipe 5%, or 40 meters. If < If so, the wave speed matching verification is passed.

[0058] In this simulated scenario of a sudden large-scale leak caused by third-party excavation damage, the energy accumulation rate at the moment of the leak is... and pulse density At the same time, it rose sharply. The value jumped from 0.1 to 0.85 within 0.2 seconds, and the dynamic window rapidly contracted to 0.1 seconds, accurately capturing the impact characteristics of the leakage transient. Acoustic-pressure coupling degree With a reliability of up to 0.93 and a wave velocity matching verification deviation of 1.2%, the system issues a high-confidence alarm and completes location within 1 second after the leak occurs, fully demonstrating the invention's rapid response capability to sudden and severe leaks.

[0059] Example 3 This embodiment demonstrates the effective suppression effect of the present invention on the interference of the pressure regulator's operating conditions, and shares the same PE pipeline monitoring node as Embodiment 1.

[0060] During the periodic operation of the gas pressure regulator, the downstream pressure sensor detects a pressure drop edge resembling a leak. The edge computing gateway calculates the energy accumulation rate in step S2. Significantly increased, but the pulse density of the acoustic emission signal... No significant changes, dual triggering factors It is approximately 0.55, slightly above the center threshold, and the dynamic window expands to approximately 8 seconds based on the function mapping.

[0061] In the feature extraction and preliminary judgment in step S3, the falling edge feature of the pressure signal is relatively obvious, but the energy proportion of the acoustic emission signal in the target frequency band does not increase significantly, and the classification model outputs a suspected leakage probability. The value is 0.68, which exceeds the initial judgment threshold, and the process proceeds to the physical verification stage.

[0062] In step S4.1, the wave velocity matching verification showed that the calculated apparent propagation distance was within the theoretical tolerance range, and this step passed. However, in step S4.2, the acoustic-pressure homogeneous coupling verification showed that because the pressure regulator action only caused fluid pressure fluctuations and did not involve high-frequency stress release from the pipe wall, there was no significant synchronous correlation between the envelope sequence of the acoustic emission signal and the first-order difference sequence of the pressure signal. The calculated acoustic-pressure synchronous coupling degree... The value was only 0.21, which is below the coupling threshold of 0.65, and therefore it failed the acoustic-pressure co-source coupling verification.

[0063] Based on the verification failure, the system marks the event as a "pressure regulation condition event" and does not report a leakage alarm to the cloud, but only stores the event record locally for operation and maintenance analysis. This embodiment shows that the present invention, through the physical mechanism of sound-pressure coupling, can effectively distinguish between pressure fluctuations caused by the operation of the pressure regulator and actual leakage events, avoiding such false alarms commonly found in existing pure data-driven solutions.

[0064] Example 4 This embodiment demonstrates the effect of the present invention in suppressing environmental vibration interference caused by heavy vehicle crushing. The monitoring object is the same as in Embodiment 2.

[0065] When a heavy truck drives over the road above the buried pipeline, the vibration energy transmitted through the soil excites the pipeline wall to vibrate. A broadband acoustic emission sensor detects a significant vibration signal with a pulse density of [missing information]. The pressure rises. However, the dynamic pressure sensor did not detect the synchronous negative pressure wave trend of the fluid inside the pipe, and the energy accumulation rate R... E No significant change. Dual triggering factors. Primarily contributed by pulse density, the dynamic window shrinks to approximately 0.5 seconds.

[0066] During the feature extraction and preliminary judgment stages, the acoustic emission signal exhibits certain energy distribution variations within the target frequency band, and the classification model outputs the probability of a suspected leak. The value is 0.61, slightly exceeding the threshold, so we proceed to physical verification.

[0067] In the wave velocity matching verification, since the acoustic emission sensors at node A and node B receive the vibration signals almost simultaneously, the arrival time difference is significant. The calculated apparent propagation velocity is close to zero, far exceeding the theoretical propagation velocity range of pipeline stress waves, thus failing the wave velocity matching verification. Simultaneously, in the acoustic-pressure coupling verification... It only reached 0.12, and also failed.

[0068] The system thus marks the event as "environmental vibration interference" and does not trigger a leakage alarm. This embodiment demonstrates that the present invention, through the dual physical constraints of wave velocity-medium spatiotemporal rigidity matching and sound-pressure homogeneous coupling, can effectively filter out false alarm sources caused by external environmental vibrations, significantly improving the system's operational reliability in complex urban environments.

[0069] Example 5 This embodiment also provides an electronic device, see reference. Figure 4 It includes a memory 404 and a processor 402, wherein the memory 404 stores a computer program and the processor 402 is configured to run the computer program to perform the steps in any of the above method embodiments.

[0070] Specifically, the processor 402 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement embodiments of the present invention.

[0071] Memory 404 may include a mass storage device for data or instructions. For example, and not limitingly, memory 404 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 404 may include removable or non-removable (or fixed) media. Where appropriate, memory 404 may be internal or external to a data processing device. In a particular embodiment, memory 404 is non-volatile memory. In a particular embodiment, memory 404 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), Extended Data Out Dynamic Random-Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0072] The memory 404 can be used to store or cache various data files that need to be processed and / or communicated, as well as possible computer program instructions executed by the processor 402.

[0073] The processor 402 reads and executes computer program instructions stored in the memory 404 to implement any of the edge computing-based real-time detection and location methods for gas pipeline leaks in the above embodiments.

[0074] Optionally, the electronic device may further include a transmission device 406 and an input / output device 408, wherein the transmission device 406 is connected to the processor 402, and the input / output device 408 is connected to the processor 402.

[0075] The transmission device 406 can be used to receive or send data via a network. Specific examples of the network described above may include wired or wireless networks provided by the communication provider of the electronic device. In one example, the transmission device includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission device 406 may be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0076] Input / output device 408 is used to input or output information.

[0077] Example 6 This embodiment also provides a readable storage medium storing a computer program, the computer program including program code for controlling a process to execute the process, the process including the edge computing-based real-time detection and location method for gas pipeline leaks according to Embodiment 1.

[0078] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0079] Generally, various embodiments can be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects of the invention can be implemented in hardware, while others can be implemented by firmware or software executed by a controller, microprocessor, or other computing device, but the invention is not limited thereto. Although various aspects of the invention may be shown and described as block diagrams, flowcharts, or using some other graphical representation, it should be understood that, by way of non-limiting example, these blocks, apparatuses, systems, techniques, or methods described herein can be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or controllers or other computing devices, or some combination thereof.

[0080] Embodiments of the present invention can be implemented by computer software, which may be executable by a data processor of a mobile device, such as a processor entity, or by hardware, or by a combination of software and hardware. Computer software or programs (also referred to as program products) including software routines, applets, and / or macros can be stored in any device-readable data storage medium, and they include program instructions for performing specific tasks. The computer program product may include one or more computer-executable components configured to perform the embodiments when the program is run. The one or more computer-executable components may be at least one piece of software code or a portion thereof. Additionally, it should be noted in this respect that, as Figure 2 Any box in the logical flow can represent a program step, or interconnected logic circuits, boxes and functions, or a combination of program steps and logic circuits, boxes and functions. Software can be stored on physical media such as memory chips or blocks of storage implemented within a processor, magnetic media such as hard disks or floppy disks, and optical media such as DVDs and their data variants, CDs, etc. The physical medium is a non-transient medium.

[0081] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0082] The above embodiments are merely illustrative of several implementations of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the appended claims.

Claims

1. A real-time detection and location method for gas pipeline leaks based on edge computing, applied to edge computing gateways deployed at upstream and downstream nodes of gas pipelines, characterized in that, Includes the following steps: Simultaneously acquire and preprocess dynamic pressure sensor signals and broadband acoustic emission sensor signals deployed on the pipeline; The energy accumulation rate of the dynamic pressure sensor signal and the transient pulse density of the broadband acoustic emission sensor signal are calculated in real time. A dual triggering factor is generated based on the energy accumulation rate and the transient pulse density. The length of the analysis window is dynamically adjusted according to the dual triggering factor to obtain an adaptive analysis window. The dual triggering factor is obtained by weighting and summing the normalized energy accumulation rate and the normalized transient pulse density based on their respective preset weighting coefficients. The length of the adaptive analysis window is continuously and smoothly adjusted according to the value of the dual triggering factor through a preset nonlinear mapping function, so that the window length converges to a preset minimum length when the dual triggering factor increases and expands to a preset maximum length when the dual triggering factor decreases. Within the adaptive analysis window, multi-scale time-frequency features are extracted from the dynamic pressure sensor signal and the broadband acoustic emission sensor signal; The extracted multi-scale time-frequency features are used for preliminary leakage determination. For events identified as suspected leaks, a spatiotemporal consistency verification based on the physical constraints of pipeline stress wave propagation is performed to filter out interfering events that do not conform to the physical conduction laws. The spatiotemporal consistency verification includes a wave velocity matching verification step. The theoretical stress wave propagation velocity is calculated based on the edge calculation of the gas pipeline between upstream and downstream nodes, the pipeline material, diameter, wall thickness, and gas density. Extract the arrival time difference of the negative pressure wave from the dynamic pressure sensor signal and calculate the apparent propagation distance; When the deviation between the apparent propagation distance and the actual pipeline length between the edge computing gateways of the upstream and downstream nodes of the gas pipeline is less than a preset tolerance threshold, it is determined that the wave velocity matching verification has been passed. The spatiotemporal consistency verification also includes an acoustic-pressure homogeneous coupling verification step: Calculate the correlation coefficient between the envelope sequence of the broadband acoustic emission sensor signal and the first-order difference sequence of the dynamic pressure sensor signal within the adaptive analysis window; When the maximum value of the correlation coefficient exceeds the preset coupling threshold, it is determined that the acoustic-pressure homogeneous coupling verification has been passed. For real leak events that pass the spatiotemporal consistency verification, the location of the leak point is calculated based on the signal arrival time difference between the edge computing gateways of the upstream and downstream nodes of the gas pipeline and the physical parameters of the pipeline, and the location result and confidence level are reported to the cloud platform.

2. The method for real-time detection and location of gas pipeline leaks based on edge computing according to claim 1, characterized in that, The nonlinear mapping function is configured as a smooth transition function with an S-shaped response curve; the smooth transition function takes the dual triggering factors as input parameters and dynamically maps the window length in combination with a preset center threshold and steepness coefficient; The center threshold is configured as a critical benchmark for determining window contraction and expansion, and the steepness coefficient is configured as a sensitivity of the length of the adaptive analysis window to changes in the dual triggering factors.

3. The method for real-time detection and location of gas pipeline leaks based on edge computing according to claim 1, characterized in that, In the step of extracting multi-scale time-frequency features, the target analysis frequency band of the broadband acoustic emission sensor signal is adaptively selected based on the pipe material information, and the wavelet packet energy ratio within the target analysis frequency band is calculated as the key feature component.

4. The method for real-time detection and location of gas pipeline leaks based on edge computing according to claim 1, characterized in that, In the step of calculating the location of the leak point, when the pipe material is a viscoelastic material, a dispersion correction factor corresponding to the viscoelastic material is obtained, and the theoretical stress wave propagation velocity is corrected using the dispersion correction factor. The location of the leak point is calculated by combining the corrected wave velocity with the signal arrival time difference.

5. An apparatus for implementing the real-time detection and location method for gas pipeline network leaks based on edge computing as described in any one of claims 1 to 4, characterized in that, include: The signal acquisition module is used to simultaneously acquire and preprocess signals from dynamic pressure sensors and broadband acoustic emission sensors deployed on the pipeline. The dynamic window generation module is used to calculate the energy accumulation rate of the dynamic pressure sensor signal and the transient pulse density of the broadband acoustic emission sensor signal in real time, generate a dual triggering factor based on the energy accumulation rate and the transient pulse density, and dynamically adjust the length of the analysis window according to the dual triggering factor to obtain an adaptive analysis window. The feature extraction module is used to extract multi-scale time-frequency features from the dynamic pressure sensor signal and the broadband acoustic emission sensor signal within the adaptive analysis window; The physical verification module is used to make a preliminary judgment on the extracted multi-scale time-frequency features. For events judged as suspected leaks, a spatiotemporal consistency verification based on the physical constraints of pipeline stress wave propagation is performed to filter out interference events that do not conform to the physical conduction law. The positioning and communication module is used to calculate the location of the leak point based on the signal arrival time difference between the edge computing gateways of the upstream and downstream nodes of the gas pipeline and the physical parameters of the pipeline for real leak events that have passed the spatiotemporal consistency verification, and to report the positioning results and confidence level to the cloud platform.

6. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to execute the edge computing-based real-time detection and location method for gas pipeline leaks as described in any one of claims 1 to 4.

7. A readable storage medium, characterized in that, The readable storage medium stores a computer program, the computer program including program code for controlling a process to execute the process, the process including the real-time detection and location method for gas pipeline network leaks based on edge computing according to any one of claims 1 to 4.

Images