Multi-Sensor Based Gas Leak Source Identification System, Method and Storage Medium

Abstract

The present invention discloses a gas leakage source identification system, method and storage medium based on multi-sensors, which relates to the technical field of probability analysis. The present invention collects and records the detection data of sensors and the effective relative diffusion coefficient of soil; constructs a gas diffusion model during underground pipeline leakage by using the collected data; calculates the optimal monitoring distance for monitoring the underground pipeline by using the historical leakage port length, and sets monitoring points on the underground pipeline within the diffusion area by using the optimal monitoring distance; calculates the hazard of each area in the diffusion area, determines the high-risk area according to the hazard, and places gas detection sensors in the high-risk area; for each set monitoring point, inputs all data into the gas diffusion model to generate an adjoint equation for each gas detection sensor; obtains the adjoint probability equation for each monitoring point to calculate the adjoint probability, and locates the gas leakage source of the underground pipeline according to the adjoint probability.

Description

Technical Field

[0001] The present invention relates to the technical field of probability analysis, and specifically to a gas leakage source identification system, method and storage medium based on multi-sensors. Background Art

[0002] In the rapid economic development and rapid urban construction, the use of land resources has become very tense. The underground space, pipeline connections, and tunnel construction have become more and more complex, and they are intricately intertwined with urban gas pipelines, resulting in insufficient safety distances, which is a very serious problem. Urban gas pipelines are very important lifelines for the entire city; due to the natural aging corrosion or human damage factors that more or less exist during the operation of natural gas pipelines, leakage accidents occur. Moreover, urban gas pipelines are mainly distributed in densely populated areas, and the surrounding environment is very complex. Once accidents such as leakage, explosion, and poisoning occur, they will cause serious casualties, property losses, and environmental damage.

[0003] Currently, the main method for detecting gas leakage in engineering is through manual detection. Workers are arranged to patrol and use some specific detectors to detect whether the gas concentration in a certain place exceeds the standard to determine whether the pipeline has gas leakage. Then, large-scale excavation is carried out within the scope of this pipeline to find the leakage point, which will cause a large amount of loss of human and material resources as well as time and economy.

[0004] Currently, the current underground pipeline leakage source tracing methods generally have the problem of insufficient quantitative calculation means. The traditional current underground pipeline leakage source tracing methods generally have the problem of insufficient quantitative calculation means. The traditional methods mainly rely on repeated trial and error, with low efficiency and time-consuming. When detecting, the placement position of the sensor is not accurate, and the leakage source is found by trying to measure continuously, which affects the efficiency; this patent innovatively proposes a high-precision quantitative calculation method based on a small amount of measurement data, which greatly improves the tracing efficiency. The traditional complex formula is applicable to the above-ground convection situation, but not applicable to the underground pure diffusion situation. In view of the underground diffusion characteristics, we propose a method to simplify the adjoint equation, which is designed specifically for underground diffusion and basically maintains the original calculation accuracy. Summary of the Invention

[0005] The purpose of the present invention is to provide a gas leakage source identification system, method and storage medium based on multi-sensors to solve the problems raised in the prior art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A gas leakage source identification method based on multi-sensors, the method includes the following steps:

[0008] S100. Collect the records of gas leakage in historical underground pipelines, extract the positions of sensors when measuring the concentration of leaked gas in the records, collect the detection data of the sensors and the effective relative diffusion coefficients of the soil in the records; construct a gas diffusion model for underground pipeline leakage using the collected data.

[0009] Further, the specific steps for constructing a gas diffusion model for underground pipeline leakage using the collected data are as follows:

[0010] S101. Collect the records of gas leakage in historical underground pipelines, extract the position of the sensor as (x, y) when measuring the gas concentration in the records, and the position of the leakage source of the underground pipeline in the records is (x1, y1); calculate the relative distance between the sensor and the leakage source of the underground pipeline using the position coordinates, and the formula is:

[0011]

[0012] In the formula, r represents the relative distance between the sensor and the leakage source of the underground pipeline, and the effective relative diffusion coefficient D of the leaked gas in the soil is obtained by looking up in the soil management knowledge according to the soil type in the record. m ;

[0013] S102. Use the relative distance between the sensor and the leakage source of the underground pipeline, combined with the effective relative diffusion coefficient, to construct a gas diffusion model for underground pipeline leakage, and the formula is:

[0014]

[0015] In the formula, C(r, t) represents the gas diffusion concentration in the soil after the underground pipeline gas leakage, q represents the detection data of the sensor in the historical record, erf represents the Gaussian error function, r represents the calculated relative distance between the sensor and the leakage source, t represents the gas leakage time when measuring with the sensor in the historical record; D m represents the effective relative diffusion coefficient of the leaked gas in the soil.

[0016] S200. Collect the length of the leakage opening when historical underground pipelines have gas leakage, calculate the optimal monitoring distance for monitoring the underground pipeline using the historical leakage opening length, when the underground pipeline has real-time gas leakage, observe and measure the diffusion area after the gas leakage, and set monitoring points on the underground pipeline within the diffusion area using the optimal monitoring distance.

[0017] Further, the specific steps for setting monitoring points on the underground pipeline within the diffusion area using the optimal monitoring distance are as follows:

[0018] S201. Collect the lengths of the pipeline leakage openings when historical underground pipelines have gas leakage as {L1, L2, L3,..., Ln}, L1, L2, L3, ..., L n represent the lengths of the pipeline leakage openings when the first, second, third, ..., nth gas leakages occur in the collected historical underground pipelines. n is a positive integer. The length of the pipeline leakage opening refers to the length of the leakage opening in the longitudinal direction of the pipeline. Calculate the average value Lp of the lengths of the leakage openings during the n times of historical underground pipeline gas leakages. Based on the average value, calculate the optimal monitoring distance for monitoring the underground pipeline. The formula is:

[0019]

[0020] In the formula, Lz represents the optimal monitoring distance for monitoring the underground pipeline, Lp represents the average value of the lengths of the leakage openings during the n times of historical underground pipeline gas leakages, and L max represents the maximum value among the lengths of the leakage openings during the n times of historical underground pipeline gas leakages, and L min represents the minimum value among the lengths of the leakage openings during the n times of historical underground pipeline gas leakages;

[0021] S202. When a gas leakage occurs in the underground pipeline in real time, observe and measure the diffusion area after the gas leakage. Extract the pipeline length S in the gas diffusion area. Set monitoring points according to the calculated optimal monitoring distance and calculate the total number of monitoring points. The formula is: Round up M; In the formula, M represents the number of monitoring points that need to be set when a gas leakage occurs in the real-time underground pipeline, Lz represents the optimal monitoring distance, and the monitoring points set in the real-time underground pipeline where a gas leakage occurs are {J1, J2, J3, ..., J M},J1, J2, J3, ..., J M represent the 1st, 2nd, 3rd, ..., Mth monitoring points set in the real-time underground pipeline where a gas leakage occurs. M is a positive integer.

[0022] Calculating the optimal monitoring distance based on the lengths of the leakage openings when gas leakages occur in the historical underground pipelines and setting monitoring points according to the optimal monitoring distance ensure the monitoring of all positions in the pipeline, not only improving the detection speed but also preventing detection omissions, and reducing false alarm situations caused by environmental interference or local gas concentration fluctuations. If the monitoring points are too dense, it may be misjudged as a pipeline leakage due to local minor gas concentration changes, resulting in false alarms. If the distance between monitoring points is too large, it may miss the real leakage signal, leading to missed alarms.

[0023] S300. Collect the number of leakage times, environmental data, and pipeline data corresponding to each area in the diffusion area during the real-time underground pipeline gas leakage that were leakage sources in history. Calculate the risk of each area in the diffusion area and determine the high-risk areas according to the risk. Place gas detection sensors in the high-risk areas;

[0024] Further, the specific steps for placing gas detection sensors in high-risk areas are as follows:

[0025] S301. Collect the number of leakage times, environmental data, and pipeline data corresponding to each area in the diffusion area during real-time pipeline gas leakage in history that were leakage sources; let the number of leakage times for each area be represented by F, form the set A of the collected different types of environmental data, and form the set B of the collected different types of pipeline data; respectively calculate the linear functions of each type of environmental data in the environmental data set A and the number of leakage times, and each type of pipeline data in the pipeline data set B and the number of leakage times according to the linear regression algorithm, and extract the independent variable coefficients of all linear functions as {a1, a2, a3,..., a u+v}, a1, a2, a3,..., a u+v represent the independent variable coefficients of the 1st, 2nd, 3rd,..., u + v linear functions calculated from the two sets of extractions, where u is the number of types of environmental data in the set A, and v is the number of types of pipeline data in the set B;

[0026] S302. Use the independent variable coefficients of each linear function as the weights of each type of environmental data and pipeline data, and calculate the risk of each area when a gas leakage occurs in the underground pipeline. The formula is:

[0027] Dan = sum{F, a×A, a×B};

[0028] In the formula, Dan represents the calculated risk of each area, sum represents summation, F represents the number of times each area was a leakage source in history, a represents the weights of each type of environmental data and pipeline data, A represents the environmental data set, and B represents the pipeline data set;

[0029] S303. Calculate the risk of each area when a gas leakage occurs in the underground pipeline, calculate the average value of all risks as Dan_p, and use the average value to judge each risk. When Dan ≥ Dan_p, judge the corresponding area as a high-risk area; when Dan < Dan_p, judge the corresponding area as a low-risk area; place gas detection sensors in the judged high-risk areas.

[0030] Judge the risk of each area in the diffusion area after a gas leakage occurs in the underground pipeline, search for high-risk areas, and place gas detection sensors in the high-risk areas. Setting sensors in high-risk areas can detect gas leakage situations more quickly. Setting sensors in these areas can avoid monitoring blind spots caused by too large a distance between monitoring points. The sensors can monitor the location and degree of gas leakage more accurately; concentrating the sensors in high-risk areas can avoid over-setting sensors in low-risk areas, thus saving sensor resources and related installation and maintenance costs.

[0031] S400. For each specified monitoring point, calculate the distance between each gas detection sensor placed and the monitoring point, collect the detection data and leakage time of each gas detection sensor, and input all the data into the gas diffusion model to generate the adjoint equation for each gas detection sensor;

[0032] Further, the specific steps of inputting all the data into the gas diffusion model to generate the adjoint equation for each gas detection sensor are as follows:

[0033] S401. For the specified monitoring point J b , take the underground pipeline as the abscissa, set the ordinate at the center point of the pipeline to draw a coordinate system, assume that all monitoring points are on the pipeline, and the coordinates are expressed as (Xg, 0); the coordinates of the gas detection sensors placed in the high-risk area are (Xc, Yc); use the distance formula in S101 to calculate the real-time distance R between each gas detection sensor placed and the monitoring point; and extract the real-time time T of gas leakage in the underground pipeline and the detection data Q of each gas detection sensor;

[0034] S402. Input all the collected real-time data into the gas diffusion model to generate the adjoint equation for each gas detection sensor, specifically:

[0035]

[0036] In the formula, represents the adjoint concentration of the i-th gas detection sensor calculated, Q i represents the detection data of the i-th gas detection sensor, R i represents the real-time distance between the i-th sensor and the monitoring point; calculate for all gas detection sensors to obtain the adjoint equations of all gas detection sensors and the monitoring point J b .

[0037] Determine the leakage location of the leakage source through gas diffusion. Overcome the problem of low efficiency existing in the existing method for determining the leakage source, simplify the calculation process compared with the existing method, improve the efficiency, thereby reducing environmental pollution and reducing the harm caused by harmful gases;

[0038] S500. For each specified monitoring point, integrate the adjoint equations of each sensor to obtain the adjoint probability equation of each monitoring point, calculate the adjoint probability of each monitoring point using the data of each gas detection sensor; find the location of the gas leakage source in the underground pipeline according to the adjoint probability.

[0039] Further, the specific steps of finding the location of the gas leakage source in the underground pipeline according to the adjoint probability are as follows:

[0040] S501. For the established monitoring point J b , integrate the adjoint equations of each sensor to obtain the adjoint probability equation for each monitoring point, specifically as follows:

[0041]

[0042] In the formula, f b represents the adjoint probability of the b-th monitoring point calculated, where b belongs to 1 to M, and G represents the total number of all gas detection sensors placed; when the adjoint probabilities of all detection points are calculated, select the monitoring point corresponding to the maximum adjoint probability as the leakage source location.

[0043] The gas leakage source identification system based on multi-sensors includes a data collection module, a gas diffusion model construction module, a monitoring point determination module, a sensor placement module, an adjoint equation calculation module, and a leakage source search module;

[0044] The data collection module is used to collect the sensor positions and detection data when gas leakage occurs in historical underground pipelines;

[0045] The gas diffusion model construction module is used to extract the effective relative diffusion coefficient of the soil according to the sensor positions and detection data when gas leakage occurs in historical underground pipelines, and construct a gas diffusion model;

[0046] The monitoring point determination module is used to calculate the leakage port length when gas leakage occurs in historical underground pipelines, obtain the optimal monitoring distance, and set monitoring points;

[0047] The sensor placement module is used to calculate the risk of each area when gas leakage occurs in the underground pipeline, determine the high-risk areas, and place gas detection sensors in the high-risk areas;

[0048] The adjoint equation calculation module is used to input the data of the gas detection sensors placed in real time into the gas diffusion model to generate the adjoint equation of each gas detection sensor;

[0049] The leakage source search module is used to integrate the adjoint equations of each sensor to obtain the adjoint probability equation of each monitoring point, calculate the adjoint probability of each monitoring point; and find the underground pipeline gas leakage source location according to the adjoint probability.

[0050] The monitoring point determination module includes an optimal monitoring distance calculation unit and a monitoring point determination unit;

[0051] The optimal monitoring distance calculation unit is used to calculate the leakage port length when gas leakage occurs in historical underground pipelines to obtain the optimal monitoring distance;

[0052] The monitoring point determination unit is used to collect the diffusion area when a gas leak occurs in the underground pipeline, obtain the pipeline length within the diffusion area, and set monitoring points according to the optimal monitoring distance.

[0053] The leak source search module includes an adjoint probability calculation unit and a leak source location search unit;

[0054] The adjoint probability calculation unit is used to integrate the adjoint equations of each sensor to obtain the adjoint probability equation of each monitoring point, and calculate the adjoint probability using the adjoint probability equation;

[0055] The leak source location search unit is used to compare the adjoint probabilities of each calculated monitoring point, and select the monitoring point with the largest adjoint probability as the leak source location.

[0056] A computer-readable storage medium stores a computer program, and the computer program is used to execute the method in the above steps.

[0057] Compared with the prior art, the beneficial effects of the present invention are:

[0058] 1. The present invention calculates the optimal monitoring distance according to the leakage orifice length when a gas leak occurs in the historical underground pipeline, and sets monitoring points according to the optimal monitoring distance, ensuring the monitoring of all positions in the pipeline, not only improving the detection speed but also avoiding detection omissions, and reducing false alarms caused by environmental interference or local gas concentration fluctuations.

[0059] 2. The present invention places gas detection sensors in high-risk areas. Setting sensors in high-risk areas can detect gas leakage situations more quickly, and setting sensors in these areas can avoid monitoring blind spots caused by too large monitoring point spacing.

[0060] 3. The present invention determines the leakage position of the leak source through gas diffusion. It overcomes the problem of low efficiency existing in the existing method for determining the leak source, simplifies the calculation process compared with the existing method, improves the efficiency, and thus can reduce environmental pollution and the harm caused by harmful gases. Description of the Drawings

[0061] Figure 1 It is a module distribution diagram of the gas leak source identification system based on multi-sensors of the present invention;

[0062] Figure 2 It is a step schematic diagram of the gas leak source identification method based on multi-sensors of the present invention;

[0063] Figure 3 It is a monitoring point adjoint probability diagram of the gas leak source identification method based on multi-sensors of the present invention. Detailed Embodiments

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] Embodiment: As Figures 1 - 3 shown, the present invention provides a technical solution

[0066] A gas leakage source identification method based on multi-sensors, the method comprising the following steps:

[0067] S100. Collect records of gas leakage occurring in historical underground pipelines, extract the positions of sensors when measuring the concentration of leaked gas in the records, collect the detection data of the sensors and the effective relative diffusion coefficient of the soil in the records; construct a gas diffusion model when the underground pipeline leaks by using the collected data;

[0068] The specific steps of constructing a gas diffusion model when the underground pipeline leaks by using the collected data are as follows:

[0069] S101. Collect records of gas leakage occurring in historical underground pipelines, extract the position of the sensor as (x, y) when measuring the gas concentration in the records, and the position of the underground pipeline leakage source in the records is (x1, y1); calculate the relative distance between the sensor and the underground pipeline leakage source in the records by using the position coordinates, and the formula is:

[0070]

[0071] In the formula, r represents the relative distance between the sensor and the underground pipeline leakage source, and the effective relative diffusion coefficient D of the leaked gas in the soil is obtained by looking up in the soil management knowledge according to the soil type in the records m ;

[0072] S102. Use the relative distance between the sensor and the underground pipeline leakage source, and combine the effective relative diffusion coefficient to construct a gas diffusion model when the underground pipeline leaks, and the formula is:

[0073]

[0074] In the formula, C(r, t) represents the gas diffusion concentration in the soil after the underground pipeline gas leaks, q represents the detection data of the sensor in the historical record, erf represents the Gaussian error function, r represents the calculated relative distance between the sensor and the leakage source, t represents the gas leakage time when measuring by using the sensor in the historical record; D m represents the effective relative diffusion coefficient of the leaked gas in the soil.

[0075] S200. Collect the lengths of the leakage openings when gas leaks occur in historical underground pipelines, calculate the optimal monitoring distance for monitoring the underground pipelines using the historical leakage opening lengths. When gas leaks occur in real-time in the underground pipelines, observe and measure the diffusion area after the gas leakage, and set monitoring points on the underground pipelines within the diffusion area using the optimal monitoring distance.

[0076] The specific steps for setting monitoring points on the underground pipelines within the diffusion area using the optimal monitoring distance are as follows:

[0077] S201. Collect the lengths of the pipeline leakage openings {L1, L2, L3,..., L n} when gas leaks occur in historical underground pipelines. L1, L2, L3,..., L n represent the lengths of the pipeline leakage openings when the 1st, 2nd, 3rd,..., nth gas leaks occur in the collected historical underground pipelines, where n is a positive integer; the pipeline leakage opening length represents the length of the leakage opening in the longitudinal direction of the pipeline. Calculate the average value Lp of the lengths of the leakage openings during the n times of historical underground pipeline gas leaks, and calculate the optimal monitoring distance for monitoring the underground pipelines based on the average value. The formula is:

[0078]

[0079] In the formula, Lz represents the calculated optimal monitoring distance for monitoring the underground pipelines, Lp represents the average value of the lengths of the leakage openings during the n times of historical underground pipeline gas leaks, L max represents the maximum value among the lengths of the leakage openings during the n times of historical underground pipeline gas leaks, and L min represents the minimum value among the lengths of the leakage openings during the n times of historical underground pipeline gas leaks.

[0080] S202. When gas leaks occur in real-time in the underground pipelines, observe and measure the diffusion area after the gas leakage, extract the pipeline length S within the gas diffusion area, set monitoring points according to the calculated optimal monitoring distance, and calculate the total number of monitoring points. The formula is: Round up M; in the formula, M represents the number of monitoring points that need to be set when gas leaks occur in real-time in the underground pipelines, Lz represents the optimal monitoring distance, and the monitoring points set on the real-time underground pipelines where gas leaks occur are {J1, J2, J3,..., J M}, J1, J2, J3,..., J M represent the 1st, 2nd, 3rd,..., Mth monitoring points set on the real-time underground pipelines where gas leaks occur, and M is a positive integer.

[0081] The optimal monitoring distance is calculated based on the length of the leak when the underground pipeline has leaked gas in the past. The monitoring points are set according to the optimal monitoring distance to ensure that all positions in the pipeline are monitored. This not only improves the detection speed without missing any detection, but also reduces false alarms caused by environmental interference or local gas concentration fluctuations. If the monitoring points are too dense, they may be misjudged as pipeline leaks due to slight changes in local gas concentration, resulting in false alarms. If the distance between the monitoring points is too large, the real leakage signal may be missed, resulting in missed alarms.

[0082] S300, collecting the number of leakages, environmental data and pipeline data corresponding to the historical leakage source in each area within the diffusion area of ​​the real-time underground pipeline gas leakage, calculating the danger of each area in the diffusion area, determining the high-risk area according to the danger, and placing gas detection sensors in the high-risk area;

[0083] The specific steps for placing gas detection sensors in high-risk areas are:

[0084] S301, collect the number of leakages, environmental data and pipeline data corresponding to the leakage source in the history of each area within the diffusion area of ​​the real-time pipeline gas leakage; let the number of leakages in each area be represented by F, and the collected different types of environmental data constitute a set A, and the collected different types of pipeline data constitute a set B; according to the linear regression algorithm, calculate the linear function of each type of environmental data and the number of leakages in the environmental data set A, and each type of pipeline data and the number of leakages in the pipeline data set B, and extract the independent variable coefficients of all linear functions as {a1, a2, a3, ..., a u+v}, a1, a2, a3, ..., a u+v represents the independent variable coefficients of the 1st, 2nd, 3rd, ...u+vth linear functions calculated on the two extracted sets, where u is the number of environmental data types in set A, and v is the number of pipeline data types in set B;

[0085] S302, using the independent variable coefficient of each extracted linear function as the weight of each environmental data and pipeline data, calculate the danger of each area when a gas leak occurs in the underground pipeline, the formula is:

[0086] Dan = sum{F, a×A, a×B};

[0087] In the formula, Dan represents the calculated danger of each area, sum represents the sum, F represents the number of times each area has been a leakage source in history, a represents the weight of each environmental data and pipeline data, A represents the environmental data set, and B represents the pipeline data set;

[0088] S303. Calculate the risk of each area when a gas leak occurs in the underground pipeline. Calculate the average value of all risks as Dan_p. Use the average value to judge each risk. When Dan≥Dan_p, judge the corresponding area as a high-risk area. When Dan<Dan_p, judge the corresponding area as a low-risk area. Place gas detection sensors in the judged high-risk areas.

[0089] Judge the risk of each area within the diffusion area after a gas leak occurs in the underground pipeline, search for high-risk areas, and place gas detection sensors in the high-risk areas. Setting sensors in high-risk areas can detect gas leak situations more quickly. Setting sensors in these areas can avoid monitoring blind spots caused by too large a distance between monitoring points. The sensors can more accurately monitor the location and degree of gas leaks; concentrating the sensors in high-risk areas can avoid over-setting sensors in low-risk areas, thus saving sensor resources and related installation and maintenance costs.

[0090] S400. For each monitoring point formulated, calculate the distance between each placed gas detection sensor and the monitoring point, collect the detection data and leakage time of each gas detection sensor, and input all the data into the gas diffusion model to generate the adjoint equation of each gas detection sensor;

[0091] The specific steps for inputting all the data into the gas diffusion model to generate the adjoint equation of each gas detection sensor are as follows:

[0092] S401. For the formulated monitoring point J b , take the underground pipeline as the abscissa and set the ordinate at the center point of the pipeline to draw a coordinate system. Assume that all monitoring points are on the pipeline, and the coordinates are expressed as (Xg, 0); the coordinates of the gas detection sensors placed in the high-risk area are (Xc, Yc); use the distance formula in S101 to calculate the real-time distance R between each placed gas detection sensor and the monitoring point; and extract the real-time time T when a gas leak occurs in the underground pipeline and the detection data Q of each gas detection sensor.

[0093] S402. Input all the collected real-time data into the gas diffusion model to generate the adjoint equation of each gas detection sensor, specifically:

[0094]

[0095] In the formula, represents the adjoint concentration of the i-th calculated gas detection sensor, Q i represents the detection data of the i-th gas detection sensor, R iRepresents the real-time distance between the i-th sensor and the monitoring point; calculations are performed for all gas detection sensors to obtain the adjoint equations for all gas detection sensors and the monitoring point J b of the adjoint equation.

[0096] Determine the leakage location of the leakage source through gas diffusion. Overcome the problem of low efficiency existing in the existing methods for determining the leakage source. Simplify the calculation process compared with the existing methods, improve the efficiency, thereby reducing environmental pollution and the harm caused by harmful gases;

[0097] S500. For each established monitoring point, integrate the adjoint equations of each sensor to obtain the adjoint probability equation for each monitoring point, and calculate the adjoint probability for each monitoring point using the data of each gas detection sensor; find the location of the underground pipeline gas leakage source according to the adjoint probability.

[0098] The specific steps to find the location of the underground pipeline gas leakage source according to the adjoint probability are as follows:

[0099] S501. For the established monitoring point J b , integrate the adjoint equations of each sensor to obtain the adjoint probability equation for each monitoring point, specifically:

[0100]

[0101] In the formula, f b represents the adjoint probability of the calculated b-th monitoring point, b belongs to 1 to M, G represents the total number of all placed gas detection sensors; when the adjoint probabilities of all detection points are calculated, select the monitoring point corresponding to the maximum adjoint probability as the leakage source location.

[0102] A gas leakage source identification system based on multiple sensors, the gas leakage source identification system includes a data collection module, a gas diffusion model construction module, a monitoring point determination module, a sensor placement module, an adjoint equation calculation module, and a leakage source search module;

[0103] The data collection module is used to collect the sensor positions and detection data when gas leakage occurs in historical underground pipelines;

[0104] The gas diffusion model construction module is used to extract the effective relative diffusion coefficient of the soil according to the sensor positions and detection data when gas leakage occurs in historical underground pipelines, and construct a gas diffusion model;

[0105] The monitoring point determination module is used to calculate the leakage port length when gas leakage occurs in historical underground pipelines to obtain the optimal monitoring distance and set monitoring points;

[0106] The sensor placement module is used to calculate the risk of each area when gas leaks occur in underground pipelines, determine high-risk areas, and place gas detection sensors in the high-risk areas;

[0107] The adjoint equation calculation module is used to input the data of the gas detection sensors placed in real time into the gas diffusion model to generate the adjoint equation of each gas detection sensor;

[0108] The leakage source search module is used to integrate the adjoint equations of each sensor to obtain the adjoint probability equation of each monitoring point, calculate the adjoint probability of each monitoring point; and find the position of the gas leakage source in the underground pipeline according to the adjoint probability.

[0109] The monitoring point determination module includes an optimal monitoring distance calculation unit and a monitoring point determination unit;

[0110] The optimal monitoring distance calculation unit is used to calculate the length of the leakage opening when gas leaks occur in historical underground pipelines to obtain the optimal monitoring distance;

[0111] The monitoring point determination unit is used to collect the diffusion area when gas leaks occur in underground pipelines, obtain the pipeline length within the diffusion area, and set monitoring points according to the optimal monitoring distance.

[0112] The leakage source search module includes an adjoint probability calculation unit and a leakage source position search unit;

[0113] The adjoint probability calculation unit is used to integrate the adjoint equations of each sensor to obtain the adjoint probability equation of each monitoring point, and calculate the adjoint probability using the adjoint probability equation;

[0114] The leakage source position search unit is used to compare the adjoint probabilities of each calculated monitoring point, and select the monitoring point with the maximum adjoint probability as the leakage source position.

[0115] A computer-readable storage medium stores a computer program, and the computer program is used to execute the method in the above steps.

[0116] Example: According to the data given by the local land bureau, the effective relative diffusion coefficient of the gas in the soil is D m = 5.2×10-6m2 / s;

[0117] Select a monitoring point with coordinates (0, 0), a constant source strength of 1 kg / s. The number of selected sensors is 2, named I1 and I2 respectively; set the coordinates of sensor I1 to (-2, 5), and set the coordinates of sensor 2 to (7, 5). And it is measured that the gas has reached stability after leaking for 20 s, and the concentrations detected by the two sensors are 100 ppm and 130 ppm respectively;

[0118] Substitute the set regional coordinates, the concentrations recorded by the sensors, and the data such as the time of gas leakage into the gas diffusion model to obtain the adjoint equations for each of the two sensors;

[0119] Integrate the adjoint equations to obtain the adjoint probability equation, and then substitute the experimental data information detected by sensors S1 and S2 into the adjoint probability of the multi-sensor combination:

[0120]

[0121] Calculate that the adjoint probability of the monitoring point (0, 0) is 0. Then calculate the remaining monitoring points, and the position of the maximum adjoint probability is at (7, 0). Determine that the leakage source position is (7, 0).

[0122] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. A method for identifying gas leakage sources based on multi-sensors, characterized in that: The method includes the following steps: S100. Collect records of gas leakage in historical underground pipelines, extract the positions of sensors when measuring the concentration of leaked gas in the records, collect the detection data of the sensors and the effective relative diffusion coefficient of the soil in the records; construct a gas diffusion model when the underground pipeline leaks by using the collected data; The specific steps for constructing a gas diffusion model when the underground pipeline leaks by using the collected data are as follows: S101. Collect records of gas leakage in historical underground pipelines, extract the position of the sensor as (x, y) when measuring the gas concentration by using the sensor in the records, and the position of the leakage source of the underground pipeline in the records is (x1, y1); calculate the relative distance between the sensor and the leakage source of the underground pipeline in the records by using the position coordinates, and the formula is: In the formula, r represents the relative distance between the sensor and the leakage source of the underground pipeline, and the effective relative diffusion coefficient D of the leakage gas in the soil is obtained by looking up in the soil management knowledge according to the soil type in the record m ; S102. Construct a gas diffusion model when the underground pipeline leaks by using the relative distance between the sensor and the leakage source of the underground pipeline and combining with the effective relative diffusion coefficient, and the formula is: In the formula, C(r, t) represents the gas diffusion concentration in the soil after the gas leakage from the underground pipeline, q represents the detection data of the sensor in the historical record, erf represents the Gaussian error function, r represents the relative distance between the calculated sensor and the leakage source, t represents the gas leakage time when the sensor is used for measurement in the historical record; D m represents the effective relative diffusion coefficient of the leaked gas in the soil; S200. Collect the length of the leakage orifice when the historical underground pipeline leaks, calculate the optimal monitoring distance for monitoring the underground pipeline by using the historical leakage orifice length, when the underground pipeline leaks in real time, observe and measure the diffusion area after the gas leakage, and set monitoring points on the underground pipeline within the diffusion area by using the optimal monitoring distance; S300. Collect the number of leakage times, environmental data and pipeline data corresponding to each area within the diffusion area in the real-time gas leakage of the underground pipeline as leakage sources in history, calculate the risk of each area within the diffusion area, determine the high-risk areas according to the risk, and place gas detection sensors in the high-risk areas; S400. For each monitoring point formulated, calculate the distance between each placed gas detection sensor and the monitoring point, collect the detection data and leakage time of each gas detection sensor, and input all the data into the gas diffusion model to generate an adjoint equation for each gas detection sensor; The specific steps for inputting all the data into the gas diffusion model to generate an adjoint equation for each gas detection sensor are as follows: S401. For the established monitoring point J b , take the underground pipeline as the abscissa, set the ordinate of the center point of the pipeline to draw a coordinate system, assume that all monitoring points are on the pipeline, and the coordinates are expressed as (Xg, 0); the coordinates of the gas detection sensors placed in the high-risk area are (Xc, Yc); use the distance formula in S101 to calculate the real-time distance R between each placed gas detection sensor and the monitoring point; and extract the real-time time T of gas leakage in the underground pipeline and the detection data Q of each gas detection sensor; S402. Input all the collected real-time data into the gas diffusion model to generate an adjoint equation for each gas detection sensor, specifically: In the formula, represents the adjoint concentration of the i-th gas detection sensor in the calculation, and Q i represents the detection data of the i-th gas detection sensor, and R i represents the real-time distance between the i-th sensor and the monitoring point; calculations are performed for all gas detection sensors to obtain the adjoint equations for all gas detection sensors and the monitoring point J b ; S500. For each monitoring point formulated, integrate the adjoint equations of each sensor to obtain an adjoint probability equation for each monitoring point, calculate the adjoint probability of each monitoring point by using the data of each gas detection sensor; find the position of the gas leakage source of the underground pipeline according to the adjoint probability; The specific steps for finding the position of the gas leakage source of the underground pipeline according to the adjoint probability are as follows: S501. For the established monitoring point J b , integrate the adjoint equations of each sensor to obtain the adjoint probability equation for each monitoring point, specifically as follows: In the formula, f b represents the adjoint probability of the b-th monitoring point calculated, where b belongs to 1 to M, and G represents the total number of all gas detection sensors placed; after calculating the adjoint probabilities of all detection points, select the monitoring point corresponding to the maximum adjoint probability as the leakage source location.

2. The multi-sensor-based gas leakage source identification method according to claim 1, characterized in that: The specific steps for setting monitoring points on the underground pipeline within the diffusion area in S200 by using the optimal monitoring distance are as follows: S201. Collect the lengths of the pipeline leakage openings {L1, L2, L3, ..., L n} when gas leaks occur in historical underground pipelines. L1, L2, L3, ..., L n represent the lengths of the pipeline leakage openings when the 1st, 2nd, 3rd, ..., nth gas leaks occur in the collected historical underground pipelines, where n is a positive integer. The length of the pipeline leakage opening refers to the length of the leakage opening in the longitudinal direction of the pipeline. Calculate the average value Lp of the lengths of the leakage openings during the n times of gas leaks in the collected historical underground pipelines, and obtain the optimal monitoring distance for monitoring the underground pipelines according to the average value. The formula is: In the formula, Lz represents the optimal monitoring distance for monitoring underground pipelines during calculation, Lp represents the average value of the leakage orifice lengths during n historical gas leakages of underground pipelines collected, L max represents the maximum value among the leakage orifice lengths during n historical gas leakages of underground pipelines collected, L min represents the minimum value among the leakage orifice lengths during n historical gas leakages of underground pipelines collected; S202. When a gas leak occurs in the underground pipeline in real time, observe and measure the diffusion area after the gas leak, extract the pipeline length S in the gas diffusion area, set monitoring points according to the calculated optimal monitoring distance, and calculate the total number of monitoring points. The formula is as follows: M is rounded up; in the formula, M represents the number of monitoring points to be set when a gas leak occurs in the underground pipeline in real time, Lz represents the optimal monitoring distance, and the monitoring points set in the underground pipeline where a gas leak occurs are {J1, J2, J3,..., J M}, J1, J2, J3,..., J M represent the 1st, 2nd, 3rd,..., Mth monitoring points set in the underground pipeline where a gas leak occurs, and M is a positive integer.

3. The multi-sensor-based gas leakage source identification method according to claim 1, wherein: The specific steps for placing gas detection sensors in the high-risk areas in S300 are as follows: S301. Collect the number of leakage times, environmental data, and pipeline data corresponding to each area in the diffusion area during real-time underground pipeline gas leakage that were leakage sources in history. Let the number of leakage times in each area be represented by F. The collected different types of environmental data form set A, and the collected different types of pipeline data form set B. According to the linear regression algorithm, calculate the linear functions of each environmental data in the environmental data set A and the number of leakage times, and each pipeline data in the pipeline data set B and the number of leakage times respectively. Extract the independent variable coefficients of all linear functions as {a1, a2, a3,..., a u+v}, a1, a2, a3,..., a u+v represent the independent variable coefficients of the 1st, 2nd, 3rd,..., u + v linear functions calculated from the two sets of extractions, where u is the number of types of environmental data in set A and v is the number of types of pipeline data in set B; S302. Use the independent variable coefficients of each extracted linear function as the weights of each environmental data and pipeline data, calculate the risk of each area when the underground pipeline leaks, and the formula is: Dan = sum{F, a×A, a×B}; In the formula, Dan represents the risk of each calculated area, sum represents summation, F represents the number of times each area was a leakage source in history, a represents the weight of each environmental data and pipeline data, A represents the set of environmental data, and B represents the set of pipeline data; S303. Calculate the risk of each area when a gas leakage occurs in the underground pipeline, calculate the average value of all risks as Dan_p, and use the average value to judge each risk. When Dan ≥ Dan_p, judge the corresponding area as a high-risk area; when Dan < Dan_p, judge the corresponding area as a low-risk area; Place gas detection sensors in the judged high-risk areas.

4. A multi-sensor-based gas leakage source identification system applying the multi-sensor-based gas leakage source identification method according to any one of claims 1-3, characterized in that: The gas leakage source identification system includes a data collection module, a gas diffusion model construction module, a monitoring point determination module, a sensor placement module, an adjoint equation calculation module, and a leakage source search module; The data collection module is used to collect the sensor positions and detection data when a gas leakage occurred in the historical underground pipeline; The gas diffusion model construction module is used to extract the effective relative diffusion coefficient of the soil based on the sensor positions and detection data when a gas leakage occurred in the historical underground pipeline, and construct a gas diffusion model; The monitoring point determination module is used to calculate the leakage port length when a gas leakage occurred in the historical underground pipeline, obtain the optimal monitoring distance, and set monitoring points; The sensor placement module is used to calculate the risk of each area when a gas leakage occurs in the underground pipeline, judge the high-risk areas, and place gas detection sensors in the high-risk areas; The adjoint equation calculation module is used to input the data of the gas detection sensors placed in real time into the gas diffusion model to generate an adjoint equation for each gas detection sensor; The leakage source search module is used to integrate the adjoint equations of each sensor to obtain an adjoint probability equation for each monitoring point, calculate the adjoint probability for each monitoring point; and find the location of the gas leakage source in the underground pipeline according to the adjoint probability.

5. The multi-sensor based gas leakage source identification system according to claim 4, wherein: The monitoring point determination module includes an optimal monitoring distance calculation unit and a monitoring point determination unit; The optimal monitoring distance calculation unit is used to calculate the leakage port length when a gas leakage occurred in the historical underground pipeline to obtain the optimal monitoring distance; The monitoring point determination unit is used to collect the diffusion area when a gas leakage occurs in the underground pipeline, obtain the pipeline length within the diffusion area, and set monitoring points according to the optimal monitoring distance.

6. The multi-sensor-based gas leakage source identification system according to claim 4, wherein: The leakage source search module includes an adjoint probability calculation unit and a leakage source location search unit; The adjoint probability calculation unit is used to integrate the adjoint equations of each sensor to obtain an adjoint probability equation for each monitoring point, and calculate the adjoint probability using the adjoint probability equation; The leakage source location search unit is used to compare the adjoint probabilities calculated for each monitoring point, and select the monitoring point with the maximum adjoint probability as the leakage source location.

7. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and the computer program is used to execute the method described in any one of claims 1 to 3 above.

Images