Gas leakage traceability detection method with additional confidence space fault-tolerant rate

Abstract

The invention discloses a gas leakage traceability detection method with an additional confidence space error-tolerant rate. On the basis of an atmospheric diffusion model, confidence space fault-tolerant rates are added to multiple gas leakage source estimation positions detected within a small interval time, leakage point fault-tolerant spaces within continuous detection moments are established in combination with a radio positioning method, then clustering is carried out to form a fault-tolerant circle, and a fault-tolerant result is obtained. A fault-tolerant space of an initial estimation leakage point is obtained by performing intersection processing on a fault-tolerant circle formed at continuous detection moments, and finally estimation leakage point fault-tolerant space filtering is performed through environmental information to obtain a final leakage point estimation area. And the credibility and scientificity of a leakage source positioning result are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas leakage source positioning, and in particular to a gas leakage source tracing detection method with additional confidence space fault tolerance. Background Art

[0002] Rapid urbanization has driven residents' growing demand for clean air quality. The construction of air pollution monitoring networks in various regions has been continuously promoted, and the monitoring coverage has been gradually expanded. In the process of improving air quality, it has become a major responsibility to accurately identify pollution sources, timely monitor changes in pollutant concentrations, and effectively warn of air quality deterioration. This is not only a prominent challenge in addressing environmental health risks, but also the key to achieving sustainable urban development.

[0003] Gas leakage pollutants diffuse into the atmospheric environment through atmospheric diffusion, and atmospheric simulation is needed to study their diffusion. Traditional atmospheric modeling methods include Lagrangian models, computational fluid dynamics models, and statistical models. Due to its unique particle characteristics, the Lagrangian model can track turbulent vortices and a large number of particles well without distortion, which is helpful for studying concentration changes. While the CFD model has high accuracy, it affects the computational efficiency due to the complexity of parameters. On the contrary, the statistical model is relatively simple and shows computational efficiency. However, in complex cases, such as the Gaussian plume model and the Gaussian dispersion model, it is difficult to accurately describe atmospheric diffusion. The traditional static mechanism model has a trade-off between computational accuracy and computational efficiency, which limits its effectiveness in practical applications. To overcome this limitation, researchers have adopted machine learning and data mining techniques to study gas diffusion. For example, Yeganeh et al. improved the traditional support vector machine (SVM) and combined it with partial least squares (PLS) to predict carbon monoxide concentrations in Tehran, thereby obtaining more accurate predictions and improving computational efficiency. In addition, Kopbayev A et al. highlighted the various applications of artificial neural networks in atmospheric science, including pattern classification and air quality forecasting.

[0004] The existing gas leakage source positioning algorithm based on the Gaussian plume atmospheric diffusion model focuses on positioning a certain point. The positioning result has large errors, poor robustness, and does not have a regional universal relationship. Summary of the invention

[0005] In view of this, the present invention provides a gas leakage source tracing detection method with additional confidence space tolerance, which greatly improves the credibility and scientificity of the leakage source positioning result.

[0006] The gas leakage source tracing detection method with additional confidence space fault tolerance of the present invention comprises:

[0007] Step 1: Move the detection equipment to detect polluted gases in the air; select N groups of detection data at consecutive moments, and deduce the suspected leakage sources of each group of detection data based on the gas diffusion model; where N≥3;

[0008] Step 2: for each group of detection data, data expansion is performed on the detection data of the group based on the confidence interval of the detection equipment, specifically: the factory confidence interval of each sensor of the detection equipment is used as the detection error range, and the detection data containing random detection errors is generated; the suspected leakage source of the expanded data is obtained based on the gas diffusion model, so as to realize the expansion of the group of suspected leakage sources;

[0009] Step 3, clustering the expanded suspected leakage sources in each group to obtain N cluster circles; and finding the minimum circumscribed circle of all cluster circles;

[0010] Step 4, divide the minimum circumscribed circle into 2m sector areas; m is 3 to 8; count the number of suspected leakage sources in each sector area and sort them from large to small, and take the first m sector areas for the following judgment:

[0011] If the first m fan-shaped areas are discontinuous areas, the area where the minimum circumscribed circle is located is considered to be the area where the non-leakage source is located;

[0012] If the first m sector-shaped areas are continuous areas, it is considered that the first m sector-shaped areas are the areas where the leakage source is located.

[0013] Preferably, in step 1, the gas diffusion model adopts a Gaussian plume atmospheric diffusion model, a Lagrangian model or an Euler model.

[0014] Preferably, the gas diffusion model adopts a Gaussian plume atmospheric diffusion model; the methane concentration detection confidence interval of the detection equipment is ±ΔC, and the wind speed detection confidence interval is ±Δu; within these two detection confidence intervals, P methane concentration values ​​and P wind speed values ​​are randomly generated, respectively, and are recorded as C1, C2, ..., C P and u1, u2,...,u P , according to the rules of permutation and combination, they are combined in pairs to obtain P×P groups of data, and P×P suspected leakage sources are inverted and calculated according to the Gaussian plume atmospheric diffusion model to expand the group of suspected leakage sources.

[0015] Preferably, in step 3, the clustering adopts DBSCAN clustering, K-Means clustering, mean shift clustering or hierarchical clustering.

[0016] Preferably, in step 4, if the first m fan-shaped areas are continuous areas, the first m fan-shaped areas are screened according to the ambient wind direction, and the screened fan-shaped areas are the areas where the leakage source is located.

[0017] Preferably, the minimum circumscribed circle diameter perpendicular to the wind direction is used as the boundary to screen the fan-shaped area on the windward side as the area where the leakage source is located.

[0018] Beneficial effects:

[0019] (1) Based on the traditional Gaussian plume atmospheric diffusion model, the present invention adds a confidence space tolerance rate to the estimated positions of multiple gas leakage sources detected within a small interval, and combines the radio positioning method to propose a more accurate and scientific gas leakage source tracing detection model in an urban environment. According to the internal confidence interval of the gas detection equipment when it leaves the factory, the traditional Gaussian plume atmospheric diffusion model is used to establish the leakage point tolerance space within the continuous detection time, and then clustering is performed to form a tolerance circle. By performing intersection processing on the tolerance circles formed at the continuous detection time, the tolerance space of the initial estimated leakage point is obtained, and finally the estimated leakage point tolerance space is filtered through environmental information to obtain the final leakage point estimation area. The credibility and scientificity of the leakage source positioning results are greatly improved.

[0020] (2) The present invention is not limited to the Gaussian plume atmospheric diffusion model, and other gas diffusion models such as the Lagrangian model and the Euler model may also be used to deduce the suspected leakage source of each group of detection data.

[0021] (3) Further narrow the leak source location range based on environmental factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of the detection method of the present invention.

[0023] Figure 2 Flowchart for building a gas leak source tracing detection model. DETAILED DESCRIPTION

[0024] The present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0025] Based on the traditional Gaussian plume atmospheric diffusion model, the present invention adds a confidence space tolerance rate to the estimated positions of multiple gas leakage sources detected within a small interval, and combines the radio positioning method to propose a more accurate and scientific gas leakage source tracing detection method in an urban environment. The present invention first uses the traditional Gaussian plume atmospheric diffusion model to establish a leakage point tolerance space within the continuous detection time based on the internal confidence interval of the polluted gas detection equipment when it leaves the factory, and then clusters to form a tolerance circle. By performing intersection processing on the tolerance circle formed at the continuous detection time, the tolerance space of the initial estimated leakage point is obtained. Finally, the estimated leakage point tolerance space is filtered through environmental information to obtain the final leakage point estimation area, which greatly improves the credibility and scientificity of the leakage source positioning result.

[0026] The schematic diagram and flow chart of the detection method of the present invention are as follows Figure 1 and Figure 2 As shown, taking gas leak detection as an example, the details are as follows:

[0027] The Gaussian model of continuous point source diffusion in unbounded space can be used to simulate the concentration of pollutants C(x,y,z) at any downwind point:

[0028]

[0029] Among them, the leakage point is taken as the origin, x is the downwind distance, y is the horizontal crosswind distance, and z is the height of the gas sampling port of the ABB natural gas leak detection vehicle, which is set to 0.4m. Q is the methane emission intensity of the leakage source, u is the wind speed, σ y ,σ z represents the Gaussian diffusion coefficient, which is the standard deviation of the methane emission concentration of the natural gas leakage source along the y-axis and the z-axis, and obeys the Gaussian distribution. This embodiment uses a set of commonly used empirical formulas (Formula 2) to determine the Gaussian diffusion coefficient σ x ,σ y ,σ z :

[0030]

[0031] Among them, a, b, c, and d are influencing factors, which are usually determined by atmospheric stability indices such as the Pasquill-Turner stability grade, as shown in Table 1.

[0032] Table 1 Pasquale atmospheric stability classification table

[0033]

[0034]

[0035] The hardware sensor of the detection vehicle continuously collects methane concentration C (mol / m 3 ), wind speed u (m / s), wind direction θ (°). When the detection vehicle arrives at the suspected leakage point area, three sets of detection data at consecutive moments are selected and the Gaussian smoke plume diffusion model (Formula 1) is used to obtain the coordinates of the three rough positioning points A, B, and C of the leakage source, which are (x1, y1), (x2, y2), and (x3, y3), respectively.

[0036] According to the internal confidence interval of the gas detection equipment when it leaves the factory, based on the traditional Gaussian smoke plume atmospheric diffusion model, the leakage point fault tolerance space within the continuous detection time is established.

[0037] Specifically, assuming that the methane concentration detection error of the sensor is ±ΔC and the wind speed detection error is ±Δu, within the detection error of the two types of detection values, 10 groups of methane concentration values ​​and 10 groups of wind speed values ​​are randomly generated, respectively, and recorded as C1, C2, ..., C10 and u1, u2,...,u 10 , according to the rules of permutation and combination, 100 sets of data are obtained, and 100 possible leakage points are inverted and deduced according to formula (1). The above steps are performed on the rough positioning points of the three leakage sources A, B, and C respectively to expand the group of suspected leakage sources.

[0038] Then, the three clusters of suspected leakage sources are clustered to form a fault tolerance circle. Clustering algorithms such as DBSCAN clustering, K-Means clustering, mean shift clustering, and hierarchical clustering can be used for clustering.

[0039] In this embodiment, DBSCAN clustering is performed on three clusters of data points respectively. Taking 100 suspected leakage sources expanded from roughly located point A as an example, the data set format is (X a1 ,Y a1 ),(X a2 ,Y a2 ),...,(X a100 ,Y a100 ), the DBSCAN clustering algorithm steps are as follows:

[0040] (1) Distance metric

[0041] Use Euclidean distance as the metric:

[0042]

[0043] where x i =(Xa i ,Ya i ) and x j =(Xa j ,Ya j ) are two-dimensional coordinate points.

[0044] (2) Neighborhood definition

[0045] For a point xi, its ∈ domain is defined as

[0046] N ∈ (x i )={x j ∈X|d(x i ,x j )≤∈} (4)

[0047] (3) Core point determination

[0048]

[0049] (4) Cluster expansion

[0050] From a core point x iFirst, add all points in its ∈ neighborhood to the cluster, and recursively check the points in the ∈ neighborhood of these new points:

[0051] C={x j ∈X|x j For x i Density up to} (6)

[0052] Density accessibility is defined as: there exists a set of points x i ,x i+1 ,...,x j , so that each point is a core point, and

[0053] (5) Calculation of the center (center of mass) of the clustering circle

[0054] The center of the cluster can be obtained by calculating the average value of all the points in the cluster. Assume that the point set in cluster C is {(Xa1, Ya1), (Xa2, Ya2), ..., (Xa k ,Ya k )}, center x a The calculation formula is:

[0055]

[0056] Where k is the number of points in the cluster.

[0057] (6) Calculation of the radius of the clustering circle

[0058] The radius of a cluster can be obtained by calculating the maximum distance from all points in the cluster to the center of the circle. The calculation formula for radius a is:

[0059]

[0060] Where μ=(μ x ,μ y ) are the coordinates of the center of the circle

[0061] Perform the above steps for the rough positioning points B and C respectively, and get three radii a, b, c, and the center coordinates (x a ,y a ),(x b ,y b ),(x c ,y c )'s clustering circles A1, B1, C1

[0062] Then, find the minimum circumscribed circle D of the three cluster circles:

[0063] (1) Calculate the center of mass G of the minimum circumscribed circle

[0064]

[0065] (2) Calculate the distance from the center of mass G to the three circle centers

[0066]

[0067] (3) The lengths of the sides of the triangle are:

[0068]

[0069] (4) Calculate the area of ​​a triangle

[0070]

[0071] (5) Calculate the radius R of the circumscribed circle of the triangle

[0072]

[0073] (6) Calculate the center of the triangle's circumscribed circle

[0074] Using the geometric properties of the circumcircle of a triangle, the center of the circumcircle (x d ,y d ) can be calculated by the following formula:

[0075] First, calculate the coordinates of the center of the circumscribed circle D (x d ,y d ):

[0076]

[0077] Then determine the radius R of the minimum circumscribed circle d

[0078]

[0079] The minimum circumscribed circle D is denoted as (x d ,y d ,R d ), which is the error tolerance space of the initial estimated leakage source.

[0080] Next, based on the continuity principle of gas diffusion, the scope of the leakage source area is determined, specifically:

[0081] The initial estimated leakage source fault tolerance space (circumscribed circle D) obtained in the above step is evenly divided into 2m sectors, where m is 3 to 8. In this embodiment, m=6. The number of suspected leakage sources in each sector is counted and sorted from most to least. The first six sectors are selected for leakage source prediction area judgment:

[0082] If the six fan-shaped areas are non-continuous areas, this possibility is excluded, and the fault-tolerant space (circumscribed circle D) is not the area where the leakage source is located;

[0083] If the six fan-shaped areas can be connected into a semicircular area, the first m fan-shaped areas are the areas where the leakage source is located. The area where the leakage source is located can also be further narrowed down based on the environmental information (wind speed and wind direction) at the time of detection. Specifically, the fan-shaped area located on the leeward side can be deleted based on the minimum circumscribed circle diameter perpendicular to the wind direction, and the fan-shaped area located on the windward side of the six fan-shaped areas can be selected as the area where the leakage source is located.

[0084] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the protection scope of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A gas leakage source tracing detection method with additional confidence space tolerance, characterized in that: include: Step 1: Move the detection equipment to detect polluted gases in the air; Select N groups of detection data at consecutive moments, and deduce the suspected leakage source of each group of detection data based on the gas diffusion model; where N ≥ 3; Step 2: for each group of detection data, data expansion is performed on the detection data of the group based on the confidence interval of the detection equipment, specifically: the factory confidence interval of each sensor of the detection equipment is used as the detection error range, and the detection data containing random detection errors is generated; the suspected leakage source of the expanded data is obtained based on the gas diffusion model, so as to realize the expansion of the group of suspected leakage sources; Step 3, clustering the expanded suspected leakage sources in each group to obtain N cluster circles; and finding the minimum circumscribed circle of all cluster circles; Step 4, divide the minimum circumscribed circle into 2m sector areas; m is 3 to 8; count the number of suspected leakage sources in each sector area and sort them from large to small, and take the first m sector areas for the following judgment: If the first m fan-shaped areas are discontinuous areas, the area where the minimum circumscribed circle is located is considered to be the area where the non-leakage source is located; If the first m sector-shaped areas are continuous areas, it is considered that the first m sector-shaped areas are the areas where the leakage source is located.

2. The method according to claim 1, characterized in that In the step 1, the gas diffusion model adopts a Gaussian plume atmospheric diffusion model, a Lagrangian model or an Euler model.

3. The method according to claim 2, characterized in that The gas diffusion model adopts the Gaussian plume atmospheric diffusion model; the methane concentration detection confidence interval of the detection equipment is ±ΔC, and the wind speed detection confidence interval is ±Δu; within these two detection confidence intervals, P methane concentration values ​​and P wind speed values ​​are randomly generated, respectively, and are recorded as C1, C2, ..., C P and u1, u2,...,u P , according to the rules of permutation and combination, they are combined in pairs to obtain P×P groups of data, and P×P suspected leakage sources are inverted and calculated according to the Gaussian plume atmospheric diffusion model to expand the group of suspected leakage sources.

4. The method according to claim 1, characterized in that In step 3, the clustering adopts DBSCAN clustering, K-Means clustering, mean shift clustering or hierarchical clustering.

5. The method according to claim 1, characterized in that In step 4, if the first m fan-shaped areas are continuous areas, the first m fan-shaped areas are screened according to the ambient wind direction, and the screened fan-shaped areas are the areas where the leakage source is located.

6. The method according to claim 5, characterized in that Taking the minimum circumscribed circle diameter perpendicular to the wind direction as the boundary, the fan-shaped area on the windward side is screened as the area where the leakage source is located.

Images