Buried gas pipeline leakage tracing method under multi-source data driving

Abstract

The application discloses a buried gas pipeline leakage tracing method under multi-source data driving, and comprises the following steps: primary tracing screening based on tracing boundary division; secondary tracing screening based on comparison between actual monitoring concentration and theoretical maximum concentration in underground space; tertiary screening of high-leakage gas pipe section based on diffusion inversion of actual alarm maximum concentration; quaternary screening of high-leakage gas pipe section based on diffusion inversion of predicted alarm maximum concentration; investigation priority ranking based on investigation objects and result analysis; investigation priority correction based on information feedback; the application is based on dynamic tracing analysis on alarm concentration state and spatial pipeline distribution in underground space, greatly reduces the tracing range, improves the tracing accuracy, and realizes high-reliability tracing based on a small amount of data that can be actually obtained.

Description

Technical Field

[0001] This invention relates to the field of gas safety technology, specifically to a method for tracing leaks in buried gas pipelines driven by multi-source data. Background Technology

[0002] In the process of rapid urban development and construction, land resources, including underground space, are becoming increasingly scarce. The number of underground spaces and pipelines is increasing, highlighting problems such as proximity, intersections, or insufficient safety distances between these facilities and gas pipelines. Explosions caused by leaked gas due to pipeline damage are not uncommon. When gas pipeline leaks spread to independent wells such as gas wells and water supply wells, accidents can occur where manhole covers explode and injure people. Therefore, it is necessary to install combustible gas monitoring devices in underground spaces adjacent to gas pipeline networks (such as drainage wells and power supply wells) to detect the accumulation of combustible gas in surrounding underground spaces under the influence of even minor leaks in the gas pipeline network. When monitoring equipment alarms, quickly and accurately tracing the potentially leaking gas pipeline section is crucial for preventing explosions and improving maintenance and repair efficiency.

[0003] The invention patent with authorization announcement number CN111895273B discloses a method for determining the source range based on the farthest diffusion distance of leaked gas under underground space alarm. This patent mainly calculates the source range for the plane diffusion range corresponding to the known surface covering medium, without considering factors such as diffusion range calculation, concentration changes at monitoring points, and attenuation of gas diffusion into underground space. Therefore, it cannot achieve dynamic analysis and the source range is too large.

[0004] The invention patent with authorization announcement number CN110043806B discloses a method for locating leaks in buried gas pipelines based on two-point optimization and source tracing. This patent dynamically estimates the maximum leakage radius of gas based on the maximum diffusion rate of natural gas in soil, constructs a mathematical model of the leakage pipeline concentration diffusion, and combines borehole detection and optimization to trace the leaking pipeline. However, although this patent predicts the gas diffusion range by assuming a leakage time, the uncertainty of leakage time in real-world scenarios can lead to errors in the tracing range. Furthermore, parameters such as the initial leakage time and soil properties are difficult to obtain in actual leak tracing scenarios. Therefore, in underground space alarm scenarios, how to accurately trace the source based on limited, multi-source heterogeneous data such as available underground pipeline topology data, monitoring data, and underground space attribute data is a core problem that gas safety currently needs to solve. Summary of the Invention

[0005] The purpose of this invention is to provide a method for tracing leaks in buried gas pipelines driven by multi-source data, and to solve the technical problem of how to perform accurate tracing based on a small amount of heterogeneous multi-source data such as available underground pipeline topology data, monitoring data, and underground space attribute data.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A multi-source data-driven method for tracing leaks in buried gas pipelines includes the following steps:

[0008] Step 1: First source screening based on source boundary division: Taking the alarm underground space as the center and the source analysis boundary as the source radius R, the gas pipelines within the boundary are obtained as the first screening set of gas pipeline segments and points;

[0009] Step 2: Secondary source screening based on theoretical analysis of underground space concentration calculation: Based on the first screening of gas pipeline segments and points, calculate the theoretically predicted concentration that the corresponding gas pipeline can cause the alarm in the underground space, and eliminate gas pipelines whose theoretically predicted concentration is less than the maximum actual alarm concentration in the past Q1 days, thus forming a secondary screening set of gas pipeline segments and points.

[0010] Step 3: Three-stage screening of high-leakage gas pipeline segments based on diffusion inversion of the actual maximum alarm concentration: After a leak occurs in a buried gas pipeline, the gas diffusion is driven by the concentration gradient, and the concentration gradually decreases from the leak point to the far field. Therefore, the higher the monitored concentration, the closer to the leak point. Thus, based on the set of gas pipeline segments and points screened in the second stage, the source tracing radius L of the corresponding gas pipeline leak point is calculated based on the diffusion range of combustible gas. The source tracing range is defined with the alarm point as the center and the source tracing radius L. Gas pipeline segments within the source tracing range are screened as the set of gas pipeline segments and points screened in the third stage.

[0011] Step 4: Four-stage screening of high-leakage gas pipeline sections based on predicted maximum alarm concentration diffusion inversion: Based on actual concentration change trends, historical monitoring curves, and concentration diffusion decay analysis, the predicted maximum alarm concentration C in the underground space at the alarm point is determined. max And the maximum alarm concentration C is predicted by diffusion inversion calculation. max Calculate the traceability radius L1 of the corresponding gas pipeline, delineate the traceability range with the alarm point as the center and the traceability radius L1, and screen the gas pipelines within the traceability range as a set of gas pipeline segments and points for four screenings;

[0012] Step 5: System-wide layered display of screening results: Gas pipelines that have undergone secondary screening, tertiary screening, and quaternary screening are displayed using different colors or layers, not limited to different colors or layers.

[0013] Step Six: Prioritize the investigation: Based on the results of the second, third, and fourth screenings, calculate the priority of leak investigation by combining the attributes of the gas pipe sections and points, and sort the priority of the investigation according to the numerical values ​​of the calculation results.

[0014] Step 7: Priority Correction of Multi-Point Alarm Investigation Based on Information Feedback: When there are multiple alarm points, the gas pipe sections and points to be investigated are determined according to Steps 1 to 5 for different alarm points. Based on the overlapping relationship of the screened pipelines, the gas pipe sections and points are re-divided. According to Step 6, the priority correction calculation of the re-divided gas pipe sections and points is carried out. The priority of the investigation is sorted and corrected according to the value of the new calculation result.

[0015] As a further aspect of the present invention, the method for calculating the theoretically predicted concentration that can be achieved in the underground space due to the corresponding gas pipeline alarm includes, but is not limited to, theoretical derivation based on Fick's law and the mass conservation equation, numerical simulation based on fluid simulation software, or data fitting based on full-size and scale-down experiments.

[0016] As a further aspect of the present invention, the method for calculating the source radius L of the corresponding gas pipeline includes, but is not limited to, source inversion based on diffusion concentration prediction, theoretical derivation, numerical simulation, and experimental acquisition.

[0017] As a further aspect of the present invention: calculating the predicted maximum concentration C in the underground space at the alarm point. max The methods include, but are not limited to, analysis based on ventilation rate, the trend of underground space concentration curve changes, and the similarity analysis of historical underground space concentration curves in the database.

[0018] As a further aspect of the present invention: The maximum concentration of C at the alarm point in the underground space is predicted based on the ventilation rate. max The specific method involves constructing a mapping relationship between a correction coefficient a and the service life Y, structural characteristics J, and shortest distance P from each gas pipeline in the underground space, based on the service life Y, structural characteristics J, and shortest distance P from each gas pipeline in the underground space, where a = f(Y, J, P) and a ≥ 1; obtaining the highest alarm concentration C0 in the underground space over the past Q1 days, and then predicting the highest alarm concentration C. max = a × C0.

[0019] As a further aspect of the present invention: The maximum predicted alarm concentration C in the underground space is calculated based on the trend of the underground space concentration curve. max The specific method involves extracting alarm concentration data for the two days prior to the alarm point, and using a hot-winter machine learning algorithm to analyze the concentration change trend over the next three days. The maximum concentration within the previous two days (Q2 and Q3) is taken as the highest alarm concentration C. max .

[0020] As a further aspect of the present invention: Based on the similarity analysis of historical underground space concentration curves in the database, the maximum predicted alarm concentration C at the alarm point underground space is calculated. maxThe specific method involves extracting alarm concentration data from the four days preceding the alarm point to obtain its highest alarm concentration C4; based on the time-frequency characteristics of the concentration change curve period, rate of change, and peak value at the alarm point, comparing it with all historical data in the database to obtain curves with a similarity rate greater than b, where the value of b is determined based on feature analysis of past alarm data to establish a similarity threshold; obtaining the ratio coefficient c between the concentration at the corresponding point on this historical curve and the maximum concentration within the next three days, thus predicting the highest alarm concentration C at that alarm point. max =max(C4,c×C4).

[0021] As a further aspect of the present invention: The priority of investigation is calculated by combining the properties of the gas pipeline, and the priority of investigation is ranked according to the numerical results of the calculation. The specific method is as follows:

[0022] Construct a mapping relationship between the priority score A for investigation and the attributes of gas pipelines and the screening results: A = f(a) x A x ), where a x The pipeline attributes are scored, and the attribute objects include, but are not limited to, pipeline material, service life, and pipeline type; A x Score the screening results;

[0023] Sort according to the numerical value of the priority score A, with higher values ​​indicating higher priority.

[0024] As a further aspect of the present invention: the priority ranking correction of the re-divided gas pipe sections and pipe points according to step six is ​​specifically as follows: the calculation results of the investigation priority of the re-divided gas pipe sections and pipe points are accumulated, and the investigation priority is ranked according to the accumulated score data.

[0025] The beneficial effects of this invention are:

[0026] (1) Based on the dynamic source tracing analysis of the alarm concentration status and spatial pipeline distribution in underground space, this invention significantly narrows the source tracing range and improves the source tracing accuracy. Furthermore, it achieves highly reliable source tracing based on the limited amount of data available in the current situation.

[0027] (2) This invention establishes a correlation between dynamic alarm concentration and data such as gas pipeline pressure, gas pipeline network and alarm underground space topology, and soil covering medium. By analyzing the credibility of gas pipeline leakage around the alarm point causing the alarm point to reach the limit alarm concentration, it can quickly find and optimize the sorting of possible leaking pipe sections and points. Attached Figure Description

[0028] The invention will now be further described with reference to the accompanying drawings.

[0029] Figure 1 This is a schematic diagram of the tracing method of the present invention.

[0030] Figure 2 This is a schematic diagram of the three-stage screening of gas pipeline sections and points based on the traceability range of the present invention. Detailed Implementation

[0031] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figure 1 - Figure 2 As shown, this invention is a method for tracing leaks in buried gas pipelines driven by multi-source data, comprising the following steps:

[0033] Step 1: First source screening based on source boundary division: Taking the alarm underground space as the center and the source analysis boundary as the source radius R, the gas pipelines within the boundary are obtained as the first screening set of gas pipeline segments and points;

[0034] Step 2: Secondary source screening based on theoretical analysis of underground space concentration calculation: Based on the first screening of gas pipeline segments and points, calculate the theoretically predicted concentration that the corresponding gas pipeline can cause the alarm in the underground space, and eliminate gas pipelines whose theoretically predicted concentration is less than the maximum actual alarm concentration in the past Q1 days, thus forming a secondary screening set of gas pipeline segments and points.

[0035] Step 3: Three-stage screening of high-leakage gas pipeline segments based on diffusion inversion of the actual maximum alarm concentration: After a leak occurs in a buried gas pipeline, the gas diffusion is driven by the concentration gradient, and the concentration gradually decreases from the leak point to the far field. Therefore, the higher the monitored concentration, the closer to the leak point. Thus, based on the set of gas pipeline segments and points screened in the second stage, the source tracing radius L of the corresponding gas pipeline leak point is calculated based on the diffusion range of combustible gas. The source tracing range is defined with the alarm point as the center and the source tracing radius L. Gas pipeline segments within the source tracing range are screened as the set of gas pipeline segments and points screened in the third stage.

[0036] Step 4: Four-stage screening of high-leakage gas pipeline sections based on predicted maximum alarm concentration diffusion inversion: Based on actual concentration change trends, historical monitoring curves, and concentration diffusion decay analysis, the predicted maximum alarm concentration C in the underground space at the alarm point is determined. max And the maximum alarm concentration C is predicted by diffusion inversion calculation. maxThe tracing radius L1 of the corresponding gas pipeline is used as the center of the alarm point and the tracing radius L1 is used to define the tracing range. Gas pipelines within the tracing range are screened as a set of gas pipeline segments and points for four screenings.

[0037] Step 5: System-wide layered display of screening results: Gas pipelines that have undergone secondary screening, tertiary screening, and quaternary screening are displayed using different colors or layers, not limited to different colors or layers.

[0038] Step Six: Prioritize the investigation: Based on the results of the second, third, and fourth screenings, calculate the priority of leak investigation by combining the attributes of the gas pipe sections and points, and sort the priority of the investigation according to the numerical values ​​of the calculation results.

[0039] Step 7: Priority Correction of Multi-Point Alarm Investigation Based on Information Feedback: When there are multiple alarm points, the gas pipe sections and points to be investigated are determined according to Steps 1 to 5 for different alarm points. Based on the overlapping relationship of the screened pipelines, the gas pipe sections and points are re-divided. According to Step 6, the priority correction calculation of the re-divided gas pipe sections and points is carried out. The priority of the investigation is sorted and corrected according to the value of the new calculation result.

[0040] The specific steps for correcting the priority ranking of the re-divided gas pipeline segments and points according to step six are as follows: the calculation results of the investigation priority of the re-divided gas pipeline segments and points are added together, and the investigation priority is ranked according to the accumulated score data.

[0041] Methods for calculating the theoretically predicted concentration that can be achieved in the underground space due to the corresponding gas pipeline alarm include, but are not limited to, theoretical derivation based on Fick's law and the mass conservation equation, numerical simulation based on fluid simulation software, or data fitting based on full-scale and scale-down experiments.

[0042] This embodiment provides a method for calculating the theoretically predicted concentration that can be reached in an underground space due to a corresponding gas pipeline alarm, as detailed below:

[0043] The theoretical predicted concentration for underground space alarms is calculated based on parameters related to different covering media and the spatial relationship of gas pipelines, using the following formula:

[0044]

[0045] In the above formula: c in Predicted concentration for alarm space, in %VOL; e is the natural constant, d s The burial depth of the leaking gas pipeline, in meters (m); d f The depth of an independent inspection well, in meters (m). f >d s When, take d f =d s Lpm η refers to the lateral distance between the surface covering medium of the m-th type of gas pipeline and the surface covering medium of the m-1 type of gas pipeline; m The correction factor refers to the type of surface cover medium for the m-th type of gas pipeline. This correction factor can be obtained based on experimental, simulation, and theoretical analysis methods.

[0046] When the burial depth of a buried gas pipeline is greater than 2m, it is considered to be a pure soil covering medium.

[0047] Methods for calculating the source radius L of the corresponding gas pipeline include, but are not limited to, source inversion based on diffusion concentration prediction, theoretical derivation, numerical simulation, and experimental acquisition.

[0048] This embodiment provides a method for calculating the traceability radius L of a corresponding gas pipeline as follows:

[0049] The traceability radius of gas pipelines is calculated based on parameters such as different covering media and spatial relationships of gas pipelines, using the following formula:

[0050]

[0051] In the above formula: c0 is the actual maximum concentration value at the alarm point, in %VOL, η m-1 I refers to the correction factor for the surface cover medium type of the (m-1)th gas pipeline. pm-1 This refers to the lateral distance between the surface covering medium of the (m-1)th type of gas pipeline and the surface covering medium of the (m-2)th type of gas pipeline.

[0052] Calculate the predicted maximum concentration of C in the underground space at the alarm point. max The methods include, but are not limited to, analysis based on ventilation rate, the trend of underground space concentration curve changes, and the similarity analysis of historical underground space concentration curves in the database.

[0053] Based on the ventilation rate, the maximum predicted alarm concentration of C in the underground space at the alarm point is calculated. max The specific method involves constructing a mapping relationship between a correction coefficient a and the service life Y, structural characteristics J, and shortest distance P from each gas pipeline in the underground space, based on the service life Y, structural characteristics J, and shortest distance P from each gas pipeline in the underground space, where a = f(Y, J, P) and a ≥ 1; and obtaining the highest alarm concentration C in the underground space over the past Q1 days. 0， Then predict the highest alarm concentration C max = a × C0.

[0054] Based on the trend of underground space concentration curve changes, the alarm point is calculated, and the predicted maximum alarm concentration C in the underground space is calculated. maxThe specific method involves extracting alarm concentration data for the two days prior to the alarm point, and using a hot-winter machine learning algorithm to analyze the concentration change trend over the next three days. The maximum concentration within the previous two days (Q2 and Q3) is taken as the highest alarm concentration C. max .

[0055] Based on the similarity analysis of historical underground space concentration curves in the database, the predicted maximum C concentration at the alarm point in the underground space is calculated. max The specific method involves extracting alarm concentration data from the four days preceding the alarm point to obtain its highest alarm concentration C4; based on the time-frequency characteristics of the concentration change curve period, rate of change, and peak value at the alarm point, comparing it with all historical data in the database to obtain curves with a similarity rate greater than b, where the value of b is determined based on feature analysis of past alarm data to establish a similarity threshold; obtaining the ratio coefficient c between the concentration at the corresponding point on this historical curve and the maximum concentration within the next three days, thus predicting the highest alarm concentration C at that alarm point. max =max(C4,c×C4).

[0056] The method for calculating the priority of inspections based on the attributes of gas pipelines, and then ranking the inspection priorities according to the calculation results, is as follows:

[0057] Construct a mapping relationship between the priority score A for investigation and the attributes of gas pipelines and the screening results: A = f(a) x A x ), where a x The pipeline attributes are scored, and the attribute objects include, but are not limited to, pipeline material, service life, and pipeline type; A x Score the screening results;

[0058] Sort according to the numerical value of the priority score A, with higher values ​​indicating higher priority.

[0059] This embodiment provides an example of a method for calculating screening priority:

[0060] A = a1 × a2 × a3 × A1

[0061] a1 represents the score corresponding to the material. When the material is cast iron, steel, or PE, the scores are 2, 1.5, and 1, respectively.

[0062] a2 represents the score corresponding to the length of service. When the length of service is 0-10, 11-20, and 20 or more, the scores are 1, 1.2, and 1.5 respectively.

[0063] a3 represents the type score, with a score of 3 for pipe segment nodes and 1 for the pipe segment body;

[0064] A1 represents the score of the screening set sequence. Secondary screening, tertiary screening, and quaternary screening correspond to scores of 2, 3, and 4, respectively.

[0065] Examples of source tracing methods:

[0066] The model method and its feasibility are illustrated through an example: On July 20, 2022, a communication manhole DX00XXX072, 1.5m deep and 0.7m in diameter, under the sidewalk on XX Road in XX City, was found to have excessive flammable gas concentration. The concentration in the confined space of the manhole was 3.13% VOL, reaching the level two alarm limit, and a leak source tracing analysis was carried out.

[0067] Step 1: A first source tracing screening based on source tracing boundary division

[0068] Based on GIS data analysis, the source tracing range is defined with the underground space under the alarm as the center and a tracing radius of 12.5m. The set of gas pipelines within the range is obtained, as shown in Table 1, to achieve a single screening.

[0069] Table 1. Collection of gas pipeline sections and points identified in the first screening.

[0070]

[0071] Step 2: Calculation of underground space concentration based on theoretical analysis

[0072] Based on the theoretical concentration prediction formula described in step 2, the theoretical concentration of pipeline diffusion into underground space after the first source tracing screening is calculated. A second screening of the source tracing pipeline is then conducted based on the calculation results. The road surface covering medium correction coefficient η = 0.4 is used, and the results are shown in Table 2 below. TQL-34XX-1960301 and TQL-34XX-1960305 are retained as the set of gas pipeline segments and points for the second screening.

[0073] Table 2. Collection of gas pipeline sections and points identified in the secondary screening.

[0074] Gas pipeline section number Concentration (%VOL) in underground space based on theoretical analysis Should we screen out? TQL-34XX-1960301 3.72 No, keep. TQL-34XX-1960305 13.05 No, keep. TQL-34XX-1960312 0.28 yes TQL-34XX-1942301 4.4E-9 yes

[0075] Step 3: Analysis of high-leakage-probability gas pipeline sections based on diffusion inversion of actual alarm maximum concentration

[0076] Based on the results of the secondary screening, for each screened pipeline, based on data such as pipeline burial depth and alarm point spacing, a diffusion inversion is performed with the alarm point as the center, based on the maximum alarm concentration, to calculate the source tracing distance. A circle is then drawn based on this source tracing distance to perform a third screening of gas pipeline sections with high leakage probability. Figure 2 As shown, Figure 2 With L = 4.72m and l = 4.42m, the sets of gas pipe sections and points selected three times are shown in Table 3 below.

[0077] Table 3. Set of gas pipe sections and points selected three times.

[0078] Gas pipeline section number Alarm point corresponding to the source tracing range Should we screen out? TQL-34XX-1960301 4.42 No, retain TQL-34XX-1960301-1 TQL-34XX-1960305 4.72 No, retain TQL-34XX-1960305-1

[0079] Step 4: Analysis of high-leakage-probability gas pipeline sections based on the diffusion inversion of predicted alarm maximum concentration

[0080] Predicting alarm concentration in underground space based on ventilation rate: Data from the alarm well was retrieved. The well was constructed in November 2009 and is made of brick. There is a high probability of cracks within the well. A correction factor of 1.2 was calculated using the following formula, resulting in a predicted maximum alarm concentration of C. max =1.2×3.13=3.756. Then, the maximum concentration Cmax of the alarm is predicted by diffusion inversion calculation, and the source tracing radius L1 of the corresponding gas pipeline is calculated. With the alarm point as the center, the source tracing range is defined by the source tracing radius L1. Gas pipelines belonging to the source tracing range are screened as the four-stage screening set of gas pipeline segments and pipeline points. The four-stage screening set of gas pipeline segments and pipeline points is shown in Table 4 below.

[0081] Table 4. Set of gas pipe sections and points selected in four rounds of screening

[0082]

[0083] Step 5: The filtering results are displayed in a hierarchical manner.

[0084] By combining the GIS system, the gas pipeline sections that have undergone secondary screening, tertiary screening, and quaternary screening can be displayed using methods such as different colors or different layers.

[0085] Step 6: Prioritize the investigation

[0086] Based on the results of the second, third, and fourth screenings, and considering the pipeline node conditions, the priority of the investigation was calculated and ranked. According to the pipeline detection data in the database, in the second and third screening results, TQL-34XX-1960305 has a differential node RE-XX-16783 within the screening range. Using the pipe age classification standards of 0-10 years, 11-20 years, and over 20 years as the classification criteria, the priority calculation results are as follows, and the investigation ranking based on the priority is shown in Table 5 below:

[0087] Table 5. Priority Calculation of Inspection Sections

[0088] Gas pipeline section number Priority TQL-34XX-1960301 3.6 TQL-34XX-1960305 3.6 TQL-34XX-1960301-1 5.4 TQL-34XX-1960305-1 5.4 TQL-34XX-1960305-1-1 7.2 RE-XX-16783 16.2

[0089] Step 7: Prioritize multi-point alarm troubleshooting based on information feedback

[0090] After on-site investigation, no alarms were found in other underground spaces in the vicinity. The investigation was then carried out according to the priority order obtained in step 6.

[0091] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A method for leakage tracing of buried gas pipeline under multi-source data driving, characterized in that, The method comprises the following steps: Step one: primary source tracing pipeline screening based on source tracing boundary division: taking the alarm underground space as the center and the source tracing radius R as the boundary, the gas pipeline in the boundary range is obtained as the primary screening gas pipeline and pipe point set; Step two: secondary source tracing pipeline screening based on underground space theoretical concentration prediction comparison: on the basis of the primary screening gas pipeline and pipe point set, the theoretical prediction concentration of the alarm underground space caused by the corresponding gas pipeline is calculated, the gas pipeline with a theoretical prediction concentration less than the actual alarm maximum concentration within Q1 days is removed, and the secondary screening gas pipeline and pipe point set is formed; Step three: tertiary screening of high-leakage gas pipeline based on actual alarm maximum concentration diffusion inversion: on the basis of the secondary screening gas pipeline and pipe point set, the source tracing radius L of the corresponding gas pipeline leak point investigation is calculated based on the combustible gas diffusion range inversion, the alarm point is taken as the center and the source tracing radius L is taken as the boundary to divide the source tracing range, and the gas pipeline within the source tracing range is screened as the tertiary screening gas pipeline and pipe point set; Step four, four times screening of high leakage gas pipe section based on predicted alarm maximum concentration diffusion inversion: based on the actual concentration trend, historical monitoring curve, concentration diffusion attenuation analysis, predict the alarm point underground space predicted alarm maximum concentration C max , and calculate the predicted alarm maximum concentration C max Corresponding to the tracing radius L1 of the gas pipeline, the tracing range is drawn with the alarm point as the center and the tracing radius L1, and the gas pipelines belonging to the tracing range are screened as the four times screening gas pipe section, pipe point set; Step five: hierarchical display of screening results: the secondary screening, tertiary screening and quaternary screening gas pipeline and pipe point are displayed by different colors or different layers; Step six: investigation priority sorting: based on the results of secondary screening, tertiary screening and quaternary screening, the leak point investigation priority is calculated combined with the properties of the investigation gas pipeline and pipe point, and the priority of investigation is sorted according to the numerical value of the calculation result; Step seven: multi-point alarm investigation priority correction based on information feedback: when there are multiple alarm points, the gas pipeline and pipe point to be investigated are obtained according to steps one to five, the gas pipeline and pipe point are re-divided based on the screening pipeline overlap relationship, the priority of the re-divided gas pipeline and pipe point is corrected according to step six, and the priority of investigation is sorted and corrected according to the numerical value of the new calculation result.

2. The multi-source data driven buried gas pipeline leak sourcing method according to claim 1, wherein, The method for calculating the theoretical prediction concentration of the alarm underground space caused by the corresponding gas pipeline is based on Fick's law and mass conservation equation for theoretical derivation, numerical simulation based on fluid simulation software or data fitting based on full-size and reduced-size experiments.

3. The multi-source data driven buried gas pipeline leak sourcing method of claim 1, wherein, The method for calculating the source tracing radius L of the corresponding gas pipeline is based on diffusion concentration prediction for source inversion, theoretical derivation, numerical simulation and experiment acquisition.

4. The multi-source data driven buried gas pipeline leak sourcing method of claim 1, wherein, Calculate the maximum concentration predicted for alarm points in underground spaces. The method is based on the ventilation rate, the trend of underground space concentration curve changes, and the similarity analysis of historical underground space concentration curves in the database.

5. The multi-source data driven buried gas pipeline leak sourcing method of claim 4, wherein, Method for calculating alarm point underground space prediction alarm maximum concentration The method is specifically to construct a mapping relationship between a correction coefficient a and an underground space service life Y, a structure characteristic J, and a shortest distance P from each gas pipeline, based on the underground space service life Y, the structure characteristic J, and the shortest distance P from each gas pipeline , and a>1; obtaining a near Q1 daily maximum alarm concentration C0, then predicting the maximum alarm concentration .

6. The multi-source data driven buried gas pipeline leak sourcing method of claim 4, wherein, The method comprises the following steps: extracting the alarm concentration data of the previous Q2 days at the alarm point, and adopting a hot-winter machine learning algorithm to analyze the concentration change trend in the future Q3 days, and taking the maximum concentration in the previous Q2 days and the Q3 days as the highest alarm concentration. The method comprises the following steps: extracting the alarm concentration data of the previous Q2 days at the alarm point, and adopting a hot-winter machine learning algorithm to analyze the concentration change trend in the future Q3 days, and taking the maximum concentration in the previous Q2 days and the Q3 days as the highest alarm concentration. The method comprises the following steps: extracting the alarm concentration data of the previous Q2 days at the alarm point, and adopting a hot-winter machine learning algorithm to analyze the concentration change trend in the future Q3 days, and taking the maximum concentration in the previous Q2 days and the Q3 days as the highest 7. The multi-source data driven buried gas pipeline leak sourcing method of claim 4, wherein, Based on the similarity analysis of historical underground space concentration curves in the database, the predicted maximum alarm concentration of underground space at alarm points is calculated. The specific method involves extracting alarm concentration data from the Q4 days prior to the alarm point to obtain its highest alarm concentration C4; based on the time-frequency characteristics of the concentration change curve period, rate of change, and peak value at the alarm point, comparing it with all historical data in the database to obtain curves with a similarity rate > b; obtaining the ratio coefficient c between the concentration at the corresponding point on the historical underground space concentration curve and the maximum concentration within the next Q3 days; and then predicting the highest alarm concentration for that alarm point. .

8. The multi-source data driven buried gas pipeline leak sourcing method of claim 1, wherein, The gas pipeline properties include gas pipeline segment properties and pipe point properties, and the numerical value of the calculation result of the investigation priority according to the gas pipeline properties is sorted as follows: Mapping relationship between the priority score A and the gas pipeline attributes and screening results is constructed wherein is the gas pipeline attribute score, and the attribute objects include pipeline material, service life, and pipeline type; is the screening result score, and the objects include screening results; According to the numerical value of the investigation priority score A, the numerical value is larger, the investigation priority is higher.

9. The multi-source data driven buried gas pipeline leak sourcing method of claim 1, wherein, The specific method for sorting and correcting the priority of the re-divided gas pipeline and pipe point according to step six is as follows: the calculation results of the investigation priority of the re-divided gas pipeline and pipe point are added, and the investigation priority is sorted according to the added score data.

Images