A method and system for inverting parameters of pipeline natural gas leakage sources

Abstract

This application relates to the field of natural gas pipeline technology, and provides a method and system for inverting parameters of pipeline natural gas leakage sources. The method first acquires natural gas concentration data at different monitoring points at time t during a pipeline natural gas leak, and determines the predicted natural gas concentration dataset C at time t+Δt based on a pipeline natural gas leakage diffusion proxy model constructed using intrinsic orthogonal decomposition and artificial neural networks. Then, a pipeline natural gas leakage source inversion model constructed using Bayesian networks and an improved particle filter algorithm is used to estimate the parameters of the pipeline natural gas leakage source. This allows for the extraction of complex concentration distribution characteristics unique to natural gas leakage scenarios based on the leakage diffusion model. Finally, the leakage source parameter information is iteratively calculated through the leakage source inversion model, effectively solving problems related to pipeline natural gas leakage source location and parameters, and providing technical support for ensuring pipeline safety.

Description

Technical Field

[0001] This application relates to the field of natural gas pipeline technology, and in particular to a method and system for inverting parameters of natural gas leakage sources in pipelines. Background Technology

[0002] Natural gas, an indispensable part of modern energy supply, is widely used as a low-carbon fuel in industry, power generation, and residential life. However, pipeline materials deteriorate over time due to environmental factors, leading to cracks or holes and potentially causing gas leaks. Especially in confined spaces like pipelines, the diffusion of natural gas is hindered, making the consequences of leaks even more severe.

[0003] Initially, the detection of natural gas leaks relied primarily on manual inspections and visual observation, involving regular checks of pipelines and the surrounding environment to detect signs such as odors or withered vegetation. With technological advancements, methods such as gas detectors and acoustic detection were gradually introduced. Gas detectors can monitor changes in gas concentration around pipelines in real time, enabling timely leak detection. Acoustic detection technology identifies leak locations by analyzing the propagation of sound waves through the pipeline. While these methods have significantly improved the accuracy and efficiency of detection, issues such as missed detections and inaccurate location still exist. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for inverting parameters of pipeline natural gas leakage sources, so as to solve or alleviate the problems existing in the prior art.

[0005] To achieve the above objectives, this application provides the following technical solution:

[0006] This application provides a method for inverting parameters of pipeline natural gas leakage source, including: step S101, obtaining natural gas concentration data at time t at different monitoring points when pipeline natural gas leakage occurs;

[0007] Step S102: Based on the pipeline natural gas leakage diffusion proxy model constructed by intrinsic orthogonal decomposition and artificial neural network, obtain the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks, according to the natural gas concentration data;

[0008] Step S103: Based on the Bayesian network and the improved particle filter algorithm, the leakage source inversion model of the pipeline natural gas is constructed, and the parameters of the leakage source of the pipeline natural gas are estimated according to the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks.

[0009] Preferably, in step S102,

[0010] The gas concentration data is dimensionality-reduced based on intrinsic orthogonal decomposition, and the dimensionality-reduced dataset of the pipeline natural gas leak is input into the artificial neural network to obtain the predicted natural gas concentration dataset C at time t+Δt during the pipeline natural gas leak; where...

[0011]

[0012] In the formula, The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j The predicted concentration of natural gas at time t+Δt is given by i = {1,…,m} and j = {1,…,n}, where m and n are both positive integers.

[0013] Preferably, in step S103, based on the leakage source inversion model, the leakage coordinates in the leakage field at the time of the pipeline natural gas leakage are cyclically resampled according to the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks and the natural gas concentration data at time t at different monitoring points.

[0014] Based on Gaussian perturbation and Markov transition matrix, the leakage source parameters corresponding to the leakage coordinates after cyclic resampling are... Update; among them, The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j The predicted concentration of natural gas at time t+Δt.

[0015] Preferably, the step of cyclically resampling the leakage coordinates within the leakage field at the time of the pipeline natural gas leak based on the leakage source inversion model and the predicted natural gas concentration dataset C at time t+Δt during the pipeline natural gas leak, along with the natural gas concentration data at different monitoring points at time t, includes: based on the Bayesian network in the leakage source inversion model, and according to the leakage coordinates (s) within the leakage field at the time of the pipeline natural gas leak... i ,x j Predicted concentration of natural gas at time t+Δt Using the natural gas concentration data at time t, calculate the leakage coordinates (s) at time t+Δt within the leakage field when the pipeline natural gas leaks. i ,x j Position weights Where i = {1, ..., m}, j = {1, ..., n}, and m and n are both positive integers;

[0016] Based on the leakage coordinates (s) at time t+Δt within the leakage field during the pipeline natural gas leak. i ,x j The position weights are determined according to the formula:

[0017]

[0018] Calculate the effective number of leakage coordinates Neff at time t+Δt within the leakage field during the pipeline natural gas leak; where w′ ij The leakage coordinates (s) at time t+Δt within the leakage field during a natural gas pipeline leak are described. i ,x j Position weights The weight values ​​after normalization;

[0019] In response to the fact that the effective number of leakage coordinates Neff at time t+Δt in the leakage field during the pipeline natural gas leak is less than a preset threshold, the leakage coordinates in the leakage field during the pipeline natural gas leak are cyclically resampled until the effective number of leakage coordinates Neff in the leakage field during the pipeline natural gas leak is greater than or equal to the preset threshold.

[0020] Preferably, the step of cyclically resampling the leakage coordinates in the leakage field at time t+Δt in response to the pipeline natural gas leak being less than a preset threshold, until the effective number of leakage coordinates in the leakage field is greater than or equal to the preset threshold, includes:

[0021] In response to the fact that the number of valid leakage coordinates (Neff) at time t+Δt within the leakage field during the pipeline natural gas leak is less than a preset threshold, the leakage coordinates at time t+Δt within the leakage field during the pipeline natural gas leak are filtered, and the leakage coordinates at time t+Δt are retained. i ,x j Leakage coordinates (s) whose position weight is greater than or equal to a preset weight threshold i ,x j ), and the leakage coordinates (s) at time t+Δt within the leakage field during the retained pipeline natural gas leakage. i ,x j Add noise information;

[0022] Based on the position weight of the leakage coordinates after adding noise information at time t+Δt in the leakage field during the pipeline natural gas leak, the effective number Neff of the leakage coordinates after adding noise information at time t+Δt in the leakage field during the pipeline natural gas leak is cyclically judged until the effective number Neff of the leakage coordinates in the leakage field during the pipeline natural gas leak is greater than or equal to a preset threshold.

[0023] Preferably, it further includes: calculating the leakage coordinates (s) of the added noise information based on the leakage diffusion proxy model. i+Δi ,x j+ΔjPredicted concentration of natural gas at )

[0024] The leakage coordinates (s) will be added to the noise information. i+Δi ,x j+Δj Predicted concentration of natural gas at ) The leak coordinates (s) at time t+Δt are compared with the natural gas concentration data at time t to determine the leak coordinates (s) within the leak field at time t+Δt after adding noise information during the pipeline natural gas leak. i+Δi ,x j+Δj The position weight of ).

[0025] Preferably, according to the formula:

[0026]

[0027] The leakage coordinates (s) at time t+Δt within the leakage field during the pipeline natural gas leak. i ,x j Position weights After normalization, the weight values ​​w are obtained. ij ;

[0028] In the formula, The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j The concentration of natural gas at time t. The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j Predicted natural gas concentration at time t+Δt;

[0029] σ is the leakage coordinate (s) within the leakage field when the pipeline natural gas leaks. i ,x j The measurement error of the natural gas concentration 0 at time t at point t;

[0030] i = {1, ..., m}, j = {1, ..., n}, where m and n are both positive integers.

[0031] Preferably, the leakage source parameters corresponding to the cyclically resampled leakage coordinates are based on Gaussian perturbation and Markov transition matrix. Updates will be made, including:

[0032] According to the formula:

[0033]

[0034] Leak source parameters corresponding to the leak coordinates after cyclic resampling Update; among them,

[0035]

[0036] In the formula, The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j ) at t+

[0037] Predicted natural gas concentration at time Δt;

[0038] The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j The leakage source parameters at time t. The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j Predicted natural gas concentration at time t; For the parameters of the leakage source Leakage source parameters after adding Gaussian perturbation;

[0039] The coordinates of the leak at time t (s) represent the leak field during the natural gas leak in the pipeline. i ,x j Leakage coordinates (s) at time t+Δt i+Δi ,x j+Δj The Markov transition matrix of ); S is the leakage source parameter. The covariance matrix constructed after Cholesky decomposition;

[0040] The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i+Δi ,x j+Δj Likelihood estimate at time t+Δt; The coordinates of the leak within the leak area when the pipeline natural gas leaks (s) i ,x j Likelihood estimate at time t;

[0041] μ∈[0,1], is a random function; i={1,…,m}, j={1,…,n}, where m and n are both positive integers.

[0042] Preferably, this application also provides a pipeline natural gas leak source parameter inversion system, including:

[0043] The data acquisition unit is configured to acquire natural gas concentration data at time t at different monitoring points when a pipeline natural gas leak occurs;

[0044] The simulation prediction unit is configured as a pipeline natural gas leakage diffusion proxy model constructed based on intrinsic orthogonal decomposition and artificial neural network, and obtains the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks, based on the natural gas concentration data.

[0045] The inversion estimation unit is configured to construct an inversion model of the pipeline natural gas leakage source based on a Bayesian network and an improved particle filter algorithm, and to estimate the parameters of the pipeline natural gas leakage source based on the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks.

[0046] Beneficial effects:

[0047] The pipeline natural gas leak source inversion method provided in this application acquires natural gas concentration data at different monitoring points at time t during a pipeline natural gas leak. Based on an intrinsic orthogonal decomposition and artificial neural network-constructed pipeline natural gas leak diffusion surrogate model, it determines the predicted natural gas concentration dataset C at time t+Δt during the pipeline natural gas leak. Furthermore, based on a Bayesian network and improved particle filter algorithm, a pipeline natural gas leak source inversion model is constructed to estimate the parameters of the pipeline natural gas leak source. This allows for the extraction of complex concentration distribution characteristics unique to natural gas leak scenarios based on the leak diffusion model. Then, the leak source parameter information is iteratively calculated through the leak source inversion model, effectively solving problems related to pipeline natural gas leak source location and parameters, and providing technical support for ensuring pipeline safety. Attached Figure Description

[0048] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments of this application and their descriptions are used to explain this application and do not constitute an undue limitation of this application.

[0049] in:

[0050] Figure 1 This is a flowchart illustrating a pipeline natural gas leak source inversion method according to some embodiments of this application;

[0051] Figure 2 This is a schematic diagram comparing simulation results and CFD results of a leakage diffusion proxy model in an obstacle-free scenario provided according to some embodiments of this application;

[0052] Figure 3 This is a schematic diagram comparing simulation results and CFD results of a leakage diffusion proxy model in an obstacle-filled scenario provided according to some embodiments of this application;

[0053] Figure 4 This is a schematic diagram showing the initial distribution of leakage coordinates in different scenarios for the leakage source inversion model provided according to some embodiments of this application;

[0054] Figure 5 This is a schematic diagram illustrating the convergence of leakage coordinates in different scenarios for the leakage source inversion model provided according to some embodiments of this application;

[0055] Figure 6 This is a histogram of the posterior probability distribution of the abscissa of the leakage source in an obstacle-free scenario according to some embodiments of this application;

[0056] Figure 7 This is a histogram of the posterior probability distribution of the abscissa of the leakage source in an obstacle-filled scenario, according to some embodiments of this application.

[0057] Figure 8 This is a histogram of the posterior probability distribution of the vertical coordinate of a leakage source in an obstacle-free scenario, according to some embodiments of this application.

[0058] Figure 9 This is a histogram of the posterior probability distribution of the vertical coordinate of a leakage source in an obstacle-filled scenario, according to some embodiments of this application.

[0059] Figure 10 This is a posterior probability distribution histogram of leakage intensity (leakage natural gas concentration) in an obstacle-free scenario according to some embodiments of this application;

[0060] Figure 11 This is a posterior probability distribution histogram of leakage intensity (leakage natural gas concentration) in an obstacle-filled scenario according to some embodiments of this application;

[0061] Figure 12 This is a schematic diagram of the structure of a pipeline natural gas leak source inversion system provided according to some embodiments of this application. Detailed Implementation

[0062] The present application will now be described in detail with reference to the accompanying drawings and embodiments. Various examples are provided by way of explanation and not by way of limitation. In fact, those skilled in the art will understand that modifications and variations can be made to the present application without departing from the scope or spirit of the present application. For example, a feature shown or described as part of one embodiment may be used in another embodiment to produce yet another embodiment. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention should fall within the scope of protection of the embodiments of the present invention.

[0063] In the detection of leaks in natural gas pipelines, existing gas leak detection methods mostly utilize Bayesian methods or swarm intelligence algorithms. Leak detection using Bayesian methods requires a large number of samples and involves a huge amount of computation, making it difficult to adapt to complex situations. Leak detection using swarm intelligence algorithms cannot consider model and instrument errors, and the positioning accuracy is affected by the search strategy.

[0064] To promote the automation and intelligence of pipeline leak identification and overcome the shortcomings of existing technologies based on traditional algorithms, this application proposes a pipeline natural gas leak source parameter inversion method. Through rapid and forward diffusion simulation using a surrogate model, high-precision CFD calculation results are approximately obtained to meet the needs of emergency response and provide support for promoting the automation and intelligence of emergency response to leak accidents.

[0065] like Figures 1 to 11 As shown, the method for inverting the parameters of the pipeline natural gas leak source includes:

[0066] Step S101: Obtain natural gas concentration data at time t at different monitoring points when the pipeline natural gas leaks.

[0067] In this application, a grid layout method is used to deploy sensors in the pipeline natural gas area. Therefore, when a pipeline natural gas leak occurs, the sensors at different monitoring points can achieve real-time and uniform monitoring of the natural gas concentration within the leak area, obtaining natural gas concentration data at the time of the leak. Alternatively, a drone equipped with a mid-infrared sensor can scan the leak area to obtain natural gas concentration data. Then, the obtained natural gas concentration data is preprocessed, including checking for and handling missing values ​​and normalizing the acquired data.

[0068] Step S102: Based on the pipeline natural gas leakage diffusion proxy model constructed by intrinsic orthogonal decomposition and artificial neural network, obtain the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks, according to the natural gas concentration data.

[0069] By preprocessing the natural gas concentration data of the leak area obtained by sensors, a natural gas concentration dataset C0 is generated within the leak field during a pipeline natural gas leak. Then, based on intrinsic orthogonal decomposition, the natural gas concentration dataset C0 is dimensionality-reduced to obtain a dimensionality-reduced dataset C′ for the pipeline natural gas leak. Specifically, through singular value decomposition, the natural gas concentration dataset C0 is transformed into a left singular matrix, a diagonal matrix, and a right singular matrix; where the diagonal matrix is ​​a diagonal matrix composed of the singular values ​​of the natural gas concentration dataset C0. Then, the singular values ​​in the diagonal matrix obtained by decomposing the natural gas concentration dataset C0 are sorted in descending order, and the first r singular values ​​greater than or equal to a preset threshold are extracted to form a new diagonal matrix. Furthermore, the elements corresponding to the first r singular values ​​in the left and right singular matrices are extracted to form new left and right singular matrices. Finally, the new left singular matrix, the new diagonal matrix, and the new right singular matrix are combined to obtain the dimensionality-reduced dataset C′. Therefore, by performing singular value decomposition on the natural gas concentration dataset C0 in the leakage field during pipeline natural gas leakage, the complex high-dimensional natural gas concentration dataset C0 is reduced in dimensionality. At the same time, features such as the spatial distribution of the concentration field, gas diffusion path and shape, as well as the relationship between the natural gas leakage intensity and the concentration field and the influence of the leakage source location on the concentration distribution are extracted during pipeline natural gas leakage.

[0070] In this application, low-dimensional features are extracted from complex high-dimensional data through singular value decomposition (SVD). Then, the data reduced in dimensionality by SVD is used to train an artificial neural network (ANN), which then performs predictions. The ANN comprises three mutually learning layers: an input layer, a hidden layer, and an output layer, for predicting pipeline natural gas leakage and diffusion. First, source terms (low-dimensional features extracted from complex high-dimensional data) are input into the hidden layer. Then, in the hidden layer, a nonlinear hyperbolic tangent function is used to transform the nonlinear features into linear features. Finally, the output features of the hidden layer are linearly summed to obtain the predicted natural gas concentration (i.e., the predicted natural gas concentration).

[0071] In other words, the dimensionality-reduced dataset obtained by dimensionality reduction of the natural gas concentration dataset C0 is input into an artificial neural network for training. The trained artificial neural network is then used for prediction to obtain the predicted natural gas concentration dataset C at time t+Δt when a pipeline natural gas leak occurs.

[0072]

[0073] In the formula, The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x jThe predicted concentration of natural gas at time t+Δt is given by i = {1,…,m} and j = {1,…,n}, where m and n are both positive integers.

[0074] Therefore, dimensionality reduction of the natural gas concentration data set is achieved through intrinsic orthogonal decomposition, resulting in a modal energy intensity matrix (dimensionality-reduced dataset). This matrix preserves the main features of the natural gas concentration field and has a strong mapping relationship with the source terms of natural gas leakage (leakage sources). The modal energy intensity matrix (dimensionality-reduced dataset) is used for training an artificial neural network, and the natural gas concentration distribution is recovered using orthogonal bases. In this process, intrinsic orthogonal decomposition reduces the number of output features of the artificial neural network from 1861 to 77, significantly reducing the output dimensionality of the neural network, effectively shrinking the model size and computational cost, and improving computational efficiency.

[0075] Step S103: Based on the Bayesian network and the improved particle filter algorithm, the leakage source inversion model of pipeline natural gas is constructed. Based on the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks, the parameters of the pipeline natural gas leakage source are estimated.

[0076] In this application, based on the leak source inversion model, the leak coordinates within the leak field at the time of the pipeline natural gas leak are cyclically resampled according to the predicted natural gas concentration dataset C at time t+Δt during the pipeline natural gas leak and the natural gas concentration data at time t at different monitoring points. In this process, firstly, based on the Bayesian network in the leak source inversion model, the leak coordinates (s) within the leak field at the time of the pipeline natural gas leak are cyclically resampled. i ,x j Predicted concentration of natural gas at time t+Δt Using the natural gas concentration data at time t, calculate the leakage coordinates (s) within the leakage field at time t+Δt when a natural gas leak occurs in the pipeline. i ,x j Position weights

[0077] Specifically, according to the formula:

[0078]

[0079] The coordinates of the leak at time t+Δt within the leak field during a pipeline natural gas leak (s) i ,x j Position weights After normalization, the weight values ​​w′ are obtained. ij In the formula, The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j The concentration of natural gas at time t. The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j The predicted concentration of natural gas at time t+Δt is σ, where σ is the leakage coordinate (s) within the leakage field when the pipeline natural gas leaks. i ,x j The measurement error of natural gas concentration at time t at point i is given by: i = {1, ..., m}, j = {1, ..., n}, where m and n are both positive integers.

[0080] In other words, by combining natural gas concentration data obtained from sensors with a Bayesian network, prior information about the natural gas leak source is transformed into posterior information; that is, according to the formula:

[0081]

[0082] Determine the likelihood probability P(N|M) of the natural gas concentration collected by the sensor when the pipeline natural gas leaks; where M is the leakage source parameters (including location coordinates and leakage amount) when the pipeline natural gas leaks, and N is the natural gas concentration data at the corresponding location coordinates collected when the pipeline natural gas leaks.

[0083] Then, based on the leakage coordinates (s) at time t+Δt within the leakage field during the pipeline natural gas leak. i ,x j The location weights are used to determine the effective leakage coordinates at time t+Δt within the leakage field during a natural gas pipeline leak. That is, according to the formula:

[0084]

[0085] The effective number of leakage coordinates, Neff, at time t+Δt within the leakage field during a natural gas leak is calculated. By defining the effective number of leakage coordinates at time t+Δt, particle degradation is assessed during particle filtering. When the effective number of leakage coordinates at time t+Δt is less than a preset threshold, the diversity of predicted leakage coordinates decreases, leading to reduced accuracy in leak source prediction. Therefore, when Neff is less than the preset threshold, the leakage coordinates within the leakage field are cyclically resampled until Neff is greater than or equal to the preset threshold.

[0086] Specifically, when a natural gas leak occurs in the pipeline, if the number of valid leak coordinates (Neff) at time t+Δt within the leak field is less than a preset threshold, the leak coordinates at time t+Δt within the leak field are filtered, and only the leak coordinates at time t+Δt are retained. i ,x jLeakage coordinates (s) whose position weight is greater than or equal to a preset weight threshold i ,x j ), and the leakage coordinates (s) at time t+Δt within the leakage field when the retained pipeline natural gas leaks. i ,x j Add noise information.

[0087] Based on the leakage diffusion proxy model, the leakage coordinates (s) after adding noise information are calculated. i+Δi ,x j+Δj Predicted concentration of natural gas at ) Where Δi and Δj are the leakage coordinates (s) after adding noise information, respectively. i ,x j The coordinate offset of ). Furthermore, the leakage coordinates (s) to which noise information is added. i+Δi ,x j+Δj Predicted concentration of natural gas at ) By comparing the natural gas concentration data at time t, the leakage coordinates (s) within the leakage field at time t+Δt after adding noise information are determined when a pipeline natural gas leak occurs. i+Δi ,x j+Δj The position weight of ).

[0088] The leakage coordinates (s) after adding noise information at time t+Δt are obtained. i+Δi ,x j+Δj After assigning positional weights, the number of valid leak coordinates at time t+Δt within the leak field during a pipeline natural gas leak, after incorporating noise information, is calculated. It is then determined whether the number of valid leak coordinates, Neff, after incorporating noise information is greater than or equal to a preset threshold. If the number of valid leak coordinates, Neff, after incorporating noise information is still less than the preset threshold, noise information is continuously added to the valid leak coordinates for resampling and valid coordinate count determination until the number of valid leak coordinates, Neff, is greater than or equal to the preset threshold.

[0089] In this application, multiple particles are generated using an improved particle filtering algorithm and assigned weights. The particle states are updated and heights are evaluated by combining the output of a leakage diffusion proxy model. Finally, the particle with the highest weight is selected as the estimated location and concentration of the final leakage source. After resampling, particle weights are repeatedly copied, leading to particle concentration in a few regions. To effectively avoid particle depletion in determining the location and concentration of the leakage source, a random perturbation (sampled from a Gaussian kernel and incorporating the corresponding leakage coordinates) is added to each resampled particle, causing the particle to diffuse into the surrounding space. Thus, by providing a random perturbation sample to each particle through the Gaussian kernel, particle concentration in a few regions is effectively avoided, causing particle positions to shift and redistribute in space, enhancing the particle's exploration capability.

[0090] Specifically, after determining that the effective number of leakage coordinates, Neff, is greater than or equal to a preset threshold, for the particles corresponding to the effective leakage coordinates, based on Gaussian perturbation and Markov transition matrix, the leakage source parameters corresponding to the cyclically resampled leakage coordinates are... The system is updated to allow for parameter estimation of pipeline natural gas leakage sources.

[0091] Specifically, according to the formula:

[0092]

[0093] Leak source parameters corresponding to the leak coordinates after cyclic resampling An update will be performed. Among other things,

[0094]

[0095] In the formula, The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j Predicted natural gas concentration at time t+Δt; The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j The leakage source parameters at time t. Leakage source parameters Leakage source parameters after adding Gaussian perturbation;

[0096] The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j Predicted natural gas concentration at time t; The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i+Δi ,x j+Δj Predicted natural gas concentration at time t+Δt;

[0097] The coordinates of the leak at time t (s) during a pipeline natural gas leak. i ,x j Leakage coordinates (s) at time t+Δt i+Δi ,x j+Δj The Markov transition matrix of ); S is the leakage source parameter. The covariance matrix constructed after Cholesky decomposition;

[0098] The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i+Δi ,x j+Δj Likelihood estimate at time t+Δt; The coordinates of the leak within the leak site when a natural gas pipeline leaks (s) i ,x j Likelihood estimate at time t.

[0099] Here, For the updated leak source parameters With leak source parameters The offset distance between them, i.e., the leakage source parameters Leakage coordinates and leakage source parameters The distance between the leak coordinates. And according to the formula:

[0100]

[0101] Determine the covariance matrix S. In the formula, A is the intermediate variable, h is the scaling factor during Cholesky decomposition, blkdiag is the block diagonal matrix constructor, u is the number of leakage source parameters, mn represents the total number of leakage coordinates, and ∑pos is the leakage coordinate (s i ,x j The state covariance matrix at position ) Leakage coordinates (s) i ,x j The state covariance matrix of the predicted natural gas concentration at point ( ). Let mn be the weighted average coordinates of the leaking coordinates. Let m be the weighted average concentration of the predicted natural gas concentration at the mn leakage coordinates.

[0102] Therefore, a natural gas predicted concentration dataset is obtained through a leakage diffusion proxy model. Compared with the classic Gaussian plume model, it does not need to consider the influence of obstacles in the diffusion area, has a fast calculation speed, accurate feature extraction, and high computational efficiency. Source term inversion is performed through a leakage source inversion model, which effectively solves the problems of pipeline natural gas leakage source location and parameters, and provides technical support for ensuring pipeline safety. It has a fast processing speed, simple structure, and small error, which can effectively enhance the physical meaning of pipeline leakage identification and overcome the problems of long calculation time and low accuracy of traditional models.

[0103] like Figure 12 As shown in the figure, this application embodiment also provides a pipeline natural gas leak source parameter inversion system, including:

[0104] Data acquisition unit 1201 is configured to acquire natural gas concentration data at time t at different monitoring points when a pipeline natural gas leak occurs;

[0105] The simulation prediction unit 1202 is configured as a pipeline natural gas leakage diffusion proxy model based on intrinsic orthogonal decomposition and artificial neural network. Based on the natural gas concentration data, it obtains the natural gas predicted concentration dataset C at time t+Δt when the pipeline natural gas leaks.

[0106] The inversion estimation unit 1203 is configured to construct an inversion model of pipeline natural gas leakage source based on Bayesian network and improved particle filter algorithm. It estimates the parameters of pipeline natural gas leakage source based on the natural gas predicted concentration dataset C at time t+Δt when pipeline natural gas leaks.

[0107] The pipeline natural gas leakage source parameter inversion system provided in this application can realize the steps and processes of the pipeline natural gas leakage source parameter inversion method of any of the above embodiments and achieve the same technical effect, which will not be described in detail here.

[0108] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing this invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0109] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0110] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0111] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can mean that the first feature is in direct contact with the second feature, or that the first feature is in indirect contact with the second feature through an intermediate medium. Furthermore, "above," "over," and "on top" of the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0112] In this invention, the terms "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0113] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for inverting parameters of a pipeline natural gas leakage source, characterized in that, include: Step S101: Obtain natural gas concentration data at time t at different monitoring points during pipeline natural gas leakage; Step S102: Based on the intrinsic orthogonal decomposition and artificial neural network, a pipeline natural gas leakage diffusion proxy model is constructed. Based on the natural gas concentration data, the leakage time of the pipeline natural gas is determined. The dataset C containing the predicted concentration of natural gas at time 1; Step S103: The pipeline natural gas leakage source inversion model is constructed based on Bayesian networks and an improved particle filter algorithm, according to the pipeline natural gas leakage... Using the natural gas predicted concentration dataset C at a given time, parameters are estimated for the source of the pipeline natural gas leak.

2. The method for inverting pipeline natural gas leakage source parameters according to claim 1, characterized in that, In step S102, The natural gas concentration data is dimensionality-reduced based on intrinsic orthogonal decomposition, and the dimensionality-reduced dataset of the pipeline natural gas leak is then input into the artificial neural network to obtain the pipeline natural gas leak data. The dataset C contains the predicted concentration of natural gas at time t; where, In the formula, The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Predicted natural gas concentration at any given time; , , All are positive integers.

3. The method for inverting parameters of pipeline natural gas leakage sources according to claim 2, characterized in that, In step S103, Based on the aforementioned leak source inversion model, and according to the pipeline natural gas leak... The predicted natural gas concentration dataset C at time t and the natural gas concentration data at different monitoring points at time t are used to cyclically resample the leakage coordinates in the leakage field when the pipeline natural gas leaks. Based on Gaussian perturbation and Markov transition matrix, the leakage source parameters corresponding to the leakage coordinates after cyclic resampling are... Update; among them, The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Predicted concentration of natural gas at any given time.

4. The method for inverting pipeline natural gas leakage source parameters according to claim 3, characterized in that, The inversion model based on the leak source is used to determine the timing of the pipeline natural gas leak. The predicted natural gas concentration dataset C at time t and the natural gas concentration data at different monitoring points at time t are used to cyclically resample the leakage coordinates within the leakage field during the pipeline natural gas leak, including: Based on the Bayesian network in the leak source inversion model, and according to the leak coordinates within the leak field when the pipeline natural gas leaks... Place Predicted natural gas concentration at time Using the natural gas concentration data at time t, calculate the leakage field when the pipeline natural gas leaks. Leakage coordinates at the specified time Position weight ;in, , , All are positive integers; According to the pipeline natural gas leak site Leakage coordinates at the specified time The position weights are determined according to the formula: Calculate the leakage field when the pipeline natural gas leaks. Effective number of leakage coordinates at time points ;in, For the description of the pipeline natural gas leak site Leakage coordinates at the specified time Position weight The weight values ​​after normalization; In response to the pipeline natural gas leak within the leak site Effective number of leakage coordinates at time points If the number of leak coordinates in the leak field is less than a preset threshold, the leak coordinates are cyclically resampled until the effective number of leak coordinates in the leak field is reached. Greater than or equal to the preset threshold.

5. The method for inverting pipeline natural gas leakage source parameters according to claim 4, characterized in that, The response to the pipeline natural gas leak within the leak area Effective number of leakage coordinates at time points If the number of leak coordinates in the leak field is less than a preset threshold, the leak coordinates are cyclically resampled until the effective number of leak coordinates in the leak field is reached. Greater than or equal to a preset threshold, including: In response to the pipeline natural gas leak within the leak site Effective number of leakage coordinates at time points When the pipeline natural gas leaks, the leakage area is less than a preset threshold. Filter and retain the leak coordinates at different times. Leakage coordinates at the specified time Leakage coordinates with position weights greater than or equal to a preset weight threshold and in the event of a gas leak in the pipeline, within the leak site. Leakage coordinates at time Add noise information; According to the pipeline natural gas leak site The positional weights of the leak coordinates after adding noise information at each step are applied to the leak field during a natural gas pipeline leak. The effective number of leakage coordinates after adding noise information at each step. The loop continues until the effective number of leakage coordinates within the leakage field at the time of the pipeline natural gas leak is reached. Greater than or equal to the preset threshold.

6. The method for inverting parameters of pipeline natural gas leakage sources according to claim 5, characterized in that, Also includes: Based on the aforementioned leakage diffusion proxy model, the leakage coordinates with added noise information are calculated. Predicted concentration of natural gas at the location ; Leakage coordinates with added noise information Predicted concentration of natural gas at the location The data was compared with the natural gas concentration data at time t to determine the leakage field at the time of the pipeline natural gas leak. Leakage coordinates after adding noise information at all times Position weights.

7. The method for inverting pipeline natural gas leakage source parameters according to claim 4, characterized in that, According to the formula: When the pipeline natural gas leaks, the leak site Leakage coordinates at the specified time Position weight Normalization is performed to obtain the weight values. ; In the formula, The coordinates of the leak within the leak site when the pipeline natural gas leaks. The concentration of natural gas at time t. The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Predicted natural gas concentration at any given time; The coordinates of the leak within the leak site when the pipeline natural gas leaks. The measurement error of natural gas concentration at time t; , , All are positive integers.

8. The method for inverting parameters of pipeline natural gas leakage sources according to claim 3, characterized in that, The leakage source parameters corresponding to the cyclically resampled leakage coordinates are obtained based on Gaussian perturbation and Markov transition matrix. Updates will be made, including: According to the formula: Leak source parameters corresponding to the leak coordinates after cyclic resampling Update; among them, In the formula, The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Predicted natural gas concentration at any given time; The coordinates of the leak within the leak site when the pipeline natural gas leaks. Leakage source parameters at time t For the parameters of the leakage source Leakage source parameters after adding Gaussian perturbation; The coordinates of the leak within the leak site when the pipeline natural gas leaks. Predicted natural gas concentration at time t; The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Predicted natural gas concentration at any given time; When the pipeline natural gas leaks, the leak site is composed of... Coordinates leaked at all times arrive Coordinates leaked at all times Markov transition matrix; S is the parameter of the leakage source. The covariance matrix constructed after Cholesky decomposition; The coordinates of the leak within the leak site when the pipeline natural gas leaks. Place Time-likelihood estimation; The coordinates of the leak within the leak site when the pipeline natural gas leaks. Likelihood estimation at time t; , is a random function; , , All are positive integers.

9. A pipeline natural gas leak source parameter inversion system, characterized in that, include: The data acquisition unit is configured to acquire natural gas concentration data at time t at different monitoring points when a pipeline natural gas leak occurs; The simulation prediction unit is configured as a leak propagation proxy model of the pipeline natural gas based on intrinsic orthogonal decomposition and artificial neural networks. Based on the natural gas concentration data, it obtains the leakage time of the pipeline natural gas. The dataset C containing the predicted concentration of natural gas at time 1; The inversion estimation unit is configured to construct an inversion model of the pipeline natural gas leakage source based on a Bayesian network and an improved particle filter algorithm, based on the pipeline natural gas leakage time. Using the natural gas predicted concentration dataset C at a given time, parameters are estimated for the source of the pipeline natural gas leak.

Images