A vehicle-mounted gas leak tracing detection system and method based on multi-source data fusion

Abstract

The present invention discloses a vehicle-mounted gas leak tracing detection system and method with multi-source data fusion, and relates to the technical field of gas leak tracing detection. The vehicle-mounted gas leak tracing detection method with multi-source data fusion obtains the gas concentration and environmental data of the vehicle in real time, dynamically corrects the diffusion coefficient, and reversely calculates the leakage source position and leakage rate in combination with an adaptive particle swarm optimization algorithm. The present invention can dynamically correct the diffusion coefficient according to the gas concentration and environmental data obtained in real time when the vehicle is driving, and accurately reversely calculate the leakage source position and leakage rate through an adaptive particle swarm optimization algorithm. Traditional detection methods usually rely on static parameters and preset models, and lack the ability to cope with rapidly changing environments. The present method adjusts the model in real time through multi-dimensional environmental data obtained in real time, such as wind speed, temperature and humidity, air pressure and other information, to ensure that the leakage source can still be quickly and accurately located in a dynamic environment.

Description

Technical Field

[0001] The present invention relates to the technical field of gas leakage tracing detection, and in particular to a vehicle-mounted gas leakage tracing detection system and method with multi-source data fusion. Background Art

[0002] Traditional gas leak detection mostly uses fixed sensors or handheld devices, which have the defects of small coverage, low detection accuracy, slow response speed, and inability to locate the source of the leak in real time. Although on-board detection improves efficiency, it is limited by environmental interference (such as wind speed, terrain) and multi-gas cross-interference (such as natural gas and biogas, gas vehicle exhaust emissions), making it difficult to accurately trace the leak point and identify the gas source type. In the existing technology, gas diffusion models are mostly based on static parameters and are not integrated with on-board dynamic data. There is a lack of correlation analysis between methane (CH4) and ethane (C2H6) and carbon dioxide (CO2) to distinguish the gas source type and locate the leak point. It is often difficult to locate the leak source and leak type, and multiple manual reviews are required. Therefore, there is still a pain point in the rapid detection and positioning of gas leaks.

[0003] The limitations of the existing technology include at least the following problems. First, most of the existing gas leakage source positioning methods rely on a fixed diffusion coefficient or a single data source, ignoring the dynamic changes of environmental factors such as wind speed, air pressure, temperature and humidity, which can easily lead to inaccurate gas diffusion predictions in complex and changeable environments, thereby affecting the positioning accuracy of the leakage source. Secondly, it is difficult for traditional methods to adjust the diffusion model in real time, and it is difficult to make full use of dynamic environmental data for refined corrections, which can easily lead to large estimation errors in the gas leakage source and make it difficult to provide high-precision source positioning results. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a vehicle-mounted gas leak tracing detection system and method with multi-source data fusion, which solves the problem of insufficient gas leak source positioning accuracy in the prior art.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a vehicle-mounted gas leak tracing detection system with multi-source data fusion, comprising:

[0006] The data acquisition unit is used to acquire in real time the position coordinates of several measuring points within a set range when the vehicle is traveling, as well as gas concentration data and environmental data at each measuring point, wherein the gas concentration data includes methane concentration value, ethane concentration value, and carbon dioxide concentration value, and the environmental data includes measured wind speed value, measured air pressure value, measured turbulence intensity value, measured temperature value, and measured humidity value; the diffusion correction unit is used to acquire the initial diffusion coefficient within the set range when the vehicle is traveling, and perform real-time correction analysis to obtain the real-time diffusion correction coefficient within the set range when the vehicle is traveling; the leakage source reverse positioning unit is used to invert the leakage source position coordinates and leakage rate based on an adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the gas concentration data at each measuring point; the gas type identification unit is used to identify the type of leaking gas at the leakage source position based on a preset discrimination rule.

[0007] Furthermore, the specific steps for obtaining the real-time diffusion correction coefficient within the set range when the vehicle is traveling are as follows: read the gas concentration data and environmental data at each measuring point within the set range when the vehicle is traveling in real time, and perform comprehensive analysis respectively to obtain the real-time gas concentration influence factor and the real-time environmental influence factor within the set range when the vehicle is traveling; read the initial diffusion coefficient within the set range when the vehicle is traveling, and perform comprehensive analysis in combination with the real-time gas concentration influence factor and the real-time environmental influence factor within the set range when the vehicle is traveling to obtain the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the calculation formula is as follows: ;in, is the real-time diffusion correction coefficient within the set range when the vehicle is driving, is the initial diffusion coefficient within the set range when the vehicle is traveling, is the real-time gas concentration influencing factor within the set range when the vehicle is driving, It is the real-time environmental impact factor within the set range when the vehicle is driving.

[0008] Furthermore, the specific steps for obtaining the real-time gas concentration influencing factor within the set range when the vehicle is traveling are as follows: obtaining gas concentration parameter data at each measuring point within the set range when the vehicle is traveling, the gas concentration parameter data including methane concentration parameter value, ethane concentration parameter value, and carbon dioxide concentration parameter value; comprehensively analyzing the methane concentration value, ethane concentration value, and carbon dioxide concentration value at each measuring point within the set range when the vehicle is traveling in combination with the gas concentration parameter data at the corresponding measuring points, to obtain the real-time gas concentration influencing factor within the set range when the vehicle is traveling.

[0009] Furthermore, the specific formula for calculating the real-time gas concentration influence factor within the set range when the vehicle is traveling is as follows: ; in, is the real-time gas concentration influencing factor within the set range when the vehicle is driving, The first The methane concentration value at each measuring point is The first The methane concentration parameter at each measuring point is: is the methane concentration adjustment factor stored in the database, is the methane concentration influence coefficient stored in the database, The first The ethane concentration value at each measuring point is The first The ethane concentration parameter at each measuring point is: is the ethane concentration adjustment factor stored in the database, is the ethane concentration influence coefficient stored in the database, The first The carbon dioxide concentration value at each measuring point is The first The carbon dioxide concentration parameter value at each measuring point is: is the carbon dioxide concentration adjustment factor stored in the database, is the carbon dioxide concentration influence coefficient stored in the database, , is the number of measurement points.

[0010] Furthermore, the specific steps for obtaining the real-time environmental impact factor within the set range when the vehicle is traveling are as follows: obtain the air pressure parameter value at each measuring point within the set range when the vehicle is traveling, and perform a comprehensive analysis in combination with the measured air pressure value at the corresponding measuring point to obtain the air pressure impact index within the set range when the vehicle is traveling; pre-process the measured wind speed value, measured turbulence intensity value, measured temperature value, and measured humidity value at each measuring point within the set range when the vehicle is traveling, respectively, and perform a comprehensive analysis after the pre-processing to obtain the environmental fluctuation impact index within the set range when the vehicle is traveling; perform a comprehensive analysis on the air pressure impact index and the environmental fluctuation impact index within the set range when the vehicle is traveling to obtain the real-time environmental impact factor within the set range when the vehicle is traveling.

[0011] Furthermore, the specific formulas for calculating the air pressure impact index, the environmental fluctuation impact index, and the real-time environmental impact factor within a set range when the vehicle is traveling are as follows: ; in, It is the air pressure impact index within the set range when the vehicle is driving. The first The measured air pressure value at each measuring point, The first The air pressure parameter value at each measuring point is is the air pressure influence coefficient stored in the database, It is the environmental fluctuation impact index within the set range when the vehicle is driving. The first The measured wind speed value at each measuring point is is the wind speed influence coefficient stored in the database, The first The measured turbulence intensity value at each measuring point is is the turbulence intensity influence coefficient stored in the database, The first The measured temperature value at each measuring point, is the temperature influence coefficient stored in the database, The first The measured humidity value at each measuring point, is the humidity influence coefficient stored in the database, is the real-time environmental impact factor within the set range when the vehicle is driving. , is the number of measurement points.

[0012] Furthermore, based on the adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the gas concentration data at each measuring point, the specific steps for inverting the leakage source position coordinates and the leakage rate are as follows: determine the concentration peak point on the vehicle's moving path as the initial candidate source position; obtain the wind speed direction of the measured wind speed value at each measuring point within the set range when the vehicle is traveling, and construct a wind field backtracking trajectory probability cloud map based on the corresponding measured wind speed value; generate several groups of leakage source parameters based on the Monte Carlo method; analyze the concentration distribution of each group of leakage source parameters based on the preset diffusion model, and evaluate the fit with the measured data; select the top 10% parameter sets with the highest fit for Gaussian kernel density estimation, and output the 95% confidence ellipse area of ​​the leakage source coordinates.

[0013] Furthermore, the specific formula for analyzing the initial candidate source position is as follows: ;in, is the initial candidate source location, The first The methane concentration value at each measuring point is The first The ethane concentration value at each measuring point is The first The carbon dioxide concentration value at each measuring point is , is the number of measurement points.

[0014] Furthermore, the specific steps for identifying the type of leaked gas at the leakage source location based on the preset discrimination rules are as follows: when the methane concentration value in the leaked gas at the leakage source location is higher than or equal to the preset first discrimination threshold, and the ratio of methane concentration to ethane concentration is higher than or equal to the preset second discrimination threshold, the leaked gas is regarded as pipeline natural gas; when the ethane concentration value in the leaked gas at the leakage source location is lower than or equal to the preset third discrimination threshold, and the ratio of methane concentration to ethane concentration is higher than or equal to the preset fourth discrimination threshold, the leaked gas is regarded as biogas; when the ethane concentration value in the leaked gas at the leakage source location is higher than or equal to the preset first discrimination threshold, and the ratio of methane concentration to ethane concentration is higher than or equal to the preset second discrimination threshold, and the carbon dioxide concentration value is higher than the preset fifth discrimination threshold, the leaked gas is regarded as gas vehicle emission interference, and the interference gas alarm is triggered.

[0015] A vehicle-mounted gas leak tracing detection method with multi-source data fusion includes the following steps: in a data acquisition unit, the position coordinates of several measurement points within a set range when the vehicle is traveling, as well as gas concentration data and environmental data at each measurement point are obtained in real time, the gas concentration data including methane concentration value, ethane concentration value, carbon dioxide concentration value, and the environmental data including measured wind speed value, measured air pressure value, measured turbulence intensity value, measured temperature value, and measured humidity value; in a diffusion correction unit, the initial diffusion coefficient within the set range when the vehicle is traveling is obtained, and real-time correction analysis is performed to obtain the real-time diffusion correction coefficient within the set range when the vehicle is traveling; in a leakage source reverse positioning unit, based on an adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the gas concentration data at each measurement point, the leakage source position coordinates and the leakage rate are inverted; in a gas type identification unit, the leaking gas type at the leakage source position is identified based on a preset discrimination rule.

[0016] The present invention has the following beneficial effects:

[0017] (1) The multi-source data fusion vehicle-mounted gas leak source tracing detection system effectively overcomes the problem of insufficient gas leak source positioning accuracy in the existing technology by introducing a real-time diffusion correction coefficient. The system dynamically adjusts the diffusion coefficient based on real-time environmental data such as gas concentration, wind speed, air pressure, and turbulence intensity, thereby accurately simulating the gas diffusion process. Traditional methods often rely on fixed diffusion coefficients and are difficult to adapt to environmental changes in real time. The system dynamically corrects the diffusion coefficient through multi-source data fusion to ensure the consistency of the gas diffusion model with the actual environmental conditions. This real-time correction mechanism enables the system to continuously provide high-precision leakage source positioning results in complex environments, especially when wind speed and meteorological conditions change greatly, thereby improving detection efficiency and accuracy.

[0018] (2) The multi-source data fusion vehicle-mounted gas leak source tracing detection system uses an adaptive particle swarm optimization algorithm combined with a real-time corrected diffusion model to accurately calculate the leak source location and leak rate through the optimization algorithm. Compared with the traditional single data source method, the adaptive particle swarm optimization algorithm can adaptively adjust the search range according to different environmental factors and gas concentration data, effectively improving the accuracy of source positioning. Especially in the face of dynamically changing environments, such as wind speed, temperature and humidity, air pressure and other factors, the system can adjust the diffusion model according to real-time data and quickly converge to the optimal solution through particle swarm optimization iterative calculation. This not only optimizes the leak source positioning process, but also greatly reduces the errors caused by environmental changes in traditional methods, thereby improving emergency response capabilities and positioning accuracy.

[0019] (3) The multi-source data fusion vehicle-mounted gas leak source tracing detection system can accurately identify the type of leaking gas based on the concentration data and relative proportions of gases such as methane, ethane, and carbon dioxide according to preset discrimination rules. This function is crucial for tracing the sources of different gases, such as identifying pipeline natural gas, biogas, or interfering gases emitted by gas vehicles. By comparing gas concentrations and ratios in real time, the system can automatically trigger an alarm and take appropriate measures when different types of gas leaks are detected. This not only helps to quickly determine the source of the gas, but also effectively eliminates other external interference sources, providing an accurate basis for gas leak tracing and improving the comprehensiveness and reliability of the detection system.

[0020] (4) The multi-source data fusion vehicle-mounted gas leak source tracing detection method can dynamically correct the diffusion coefficient according to the gas concentration and environmental data obtained in real time when the vehicle is driving, and accurately reversely infer the leakage source location and leakage rate through an adaptive particle swarm optimization algorithm. Traditional detection methods usually rely on static parameters and preset models and lack the ability to cope with rapidly changing environments. This method adjusts the model in real time through multi-dimensional environmental data obtained in real time, such as wind speed, temperature and humidity, air pressure and other information, to ensure that the leakage source can still be quickly and accurately located in a dynamic environment.

[0021] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention is a block diagram of a vehicle-mounted gas leak tracing detection system with multi-source data fusion.

[0023] Figure 2 The present invention is a flowchart of the specific steps of obtaining the real-time gas concentration influencing factors within a set range when the vehicle is traveling in a multi-source data fusion vehicle-mounted gas leakage tracing detection system.

[0024] Figure 3 The present invention is a flow chart of a vehicle-mounted gas leak source tracing detection method based on multi-source data fusion. DETAILED DESCRIPTION

[0025] See also Figure 1 The embodiment of the present invention provides a technical solution: a multi-source data fusion vehicle-mounted gas leak tracing detection system, including a data acquisition unit, which is used to acquire in real time the position coordinates of several measurement points within a set range when the vehicle is traveling, as well as gas concentration data and environmental data at each measurement point, wherein the gas concentration data includes a methane concentration value, an ethane concentration value, and a carbon dioxide concentration value, and the environmental data includes a measured wind speed value, a measured air pressure value, a measured turbulence intensity value, a measured temperature value, and a measured humidity value; the diffusion correction unit is used to acquire an initial diffusion coefficient within a set range when the vehicle is traveling, and perform real-time correction analysis to obtain a real-time diffusion correction coefficient within the set range when the vehicle is traveling; the leakage source reverse positioning unit is used to invert the leakage source position coordinates and the leakage rate based on an adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the gas concentration data at each measurement point; the gas type identification unit is used to identify the type of leaking gas at the leakage source position based on a preset discrimination rule.

[0026] Among them, the data acquisition unit consists of a mid-infrared laser spectrum detection module and a multi-source environment perception module.

[0027] The mid-infrared laser spectrum detection module includes a tunable semiconductor laser, a methane / ethane two-component gas absorption cell, a carbon dioxide absorption cell, a photoelectric detector and a signal processing unit, which is used to detect the concentrations of methane, ethane and carbon dioxide in the ambient air in real time, where the detection sensitivity of methane and ethane is ≤1ppb, and the detection sensitivity of CO2 is ≤1ppm; the flow rate of the gas sampling pump needs to be greater than 20L / min to speed up the gas exchange speed and improve the detection sensitivity.

[0028] The multi-source environmental perception module integrates a Beidou / GNSS high-precision positioning unit (positioning accuracy ≤ 10 cm), a three-dimensional ultrasonic anemometer (measuring range 0-30m / s, resolution 0.1m / s) and a five-parameter meteorological sensor (temperature, humidity, air pressure, wind speed, wind direction), which is used to synchronously obtain the geographic coordinates of the detection point, the three-dimensional wind speed vector and meteorological data.

[0029] Specifically, the specific steps for obtaining the real-time diffusion correction coefficient within the set range when the vehicle is traveling are as follows: read the gas concentration data and environmental data at each measurement point within the set range when the vehicle is traveling in real time, and perform comprehensive analysis respectively to obtain the real-time gas concentration influence factor and the real-time environmental influence factor within the set range when the vehicle is traveling; read the initial diffusion coefficient within the set range when the vehicle is traveling, and perform comprehensive analysis in combination with the real-time gas concentration influence factor and the real-time environmental influence factor within the set range when the vehicle is traveling to obtain the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the calculation formula is as follows: ;in, is the real-time diffusion correction coefficient within the set range when the vehicle is driving, is the initial diffusion coefficient within the set range when the vehicle is traveling, is the real-time gas concentration influencing factor within the set range when the vehicle is driving, It is the real-time environmental impact factor within the set range when the vehicle is driving.

[0030] In this implementation scheme, by introducing a real-time diffusion correction coefficient, the accuracy and adaptability of the gas leak detection system can be significantly improved. Traditional diffusion models mostly rely on fixed diffusion coefficients. Such models cannot respond to environmental changes in a timely manner, resulting in inaccurate predictions of gas diffusion paths, which in turn affects the location of the leakage source. By reading the gas concentration data and environmental data of each measurement point within the set range when the vehicle is driving in real time, the system can dynamically adjust the diffusion coefficient so that the model can more accurately reflect the actual gas diffusion process. This process is combined with real-time gas concentration influencing factors (such as methane, ethane, carbon dioxide and other concentration data) and real-time environmental influencing factors (such as wind speed, temperature and humidity, air pressure and other environmental data) for comprehensive analysis, thereby refining the gas diffusion model. Through this dynamic correction mechanism, the system can adapt to changing environmental factors, such as changes in wind speed and air pressure, so as to accurately predict the diffusion path of the gas and ensure the accuracy of leakage source location. This not only improves the accuracy and reliability of detection, but also can effectively cope with various complex and dynamic environmental conditions, optimize the leakage source location process, reduce errors, and improve the overall performance and responsiveness of the system.

[0031] Specifically, Figure 2 As shown, the specific steps for obtaining the real-time gas concentration influencing factor within the set range when the vehicle is traveling are as follows: obtaining gas concentration parameter data at each measuring point within the set range when the vehicle is traveling, the gas concentration parameter data including methane concentration parameter value, ethane concentration parameter value, and carbon dioxide concentration parameter value; comprehensively analyzing the methane concentration value, ethane concentration value, and carbon dioxide concentration value at each measuring point within the set range when the vehicle is traveling in combination with the gas concentration parameter data at the corresponding measuring points, to obtain the real-time gas concentration influencing factor within the set range when the vehicle is traveling.

[0032] The specific formula for calculating the real-time gas concentration impact factor within the set range when the vehicle is driving is as follows: ; in, is the real-time gas concentration influencing factor within the set range when the vehicle is driving, The first The methane concentration value at each measuring point is The first The methane concentration parameter at each measuring point is: is the methane concentration adjustment factor stored in the database, is the methane concentration influence coefficient stored in the database, The first The ethane concentration value at each measuring point is The first The ethane concentration parameter at each measuring point is: is the ethane concentration adjustment factor stored in the database, is the ethane concentration influence coefficient stored in the database, The first The carbon dioxide concentration value at each measuring point is The first The carbon dioxide concentration parameter value at each measuring point is: is the carbon dioxide concentration adjustment factor stored in the database, is the carbon dioxide concentration influence coefficient stored in the database, , is the number of measurement points.

[0033] It should be explained that the methane concentration adjustment factor stored in the database , Ethane concentration adjustment factor , Carbon dioxide concentration adjustment factor The specific acquisition steps are: it is obtained after preprocessing the historical data of gas concentration and different environmental conditions (such as air pressure, temperature, humidity, etc.) in the database. The calculation of these adjustment coefficients depends on the concentration data collected over a long period of time and is obtained through data fitting and regression analysis. The concentration adjustment coefficient of each gas reflects the concentration change law of the gas under different environmental conditions. Therefore, it is necessary to adjust it according to the difference between the actual monitored concentration and the standard concentration. These coefficients are obtained through statistical analysis of historical data, model calibration and verification, etc., to ensure that accurate concentration correction can be provided in different measurement environments.

[0034] Methane concentration influence coefficient stored in the database , Ethane concentration influence coefficient , Carbon dioxide concentration influence coefficient The specific steps for obtaining are: by analyzing the correlation between gas concentration data and environmental variables (such as wind speed, temperature, humidity, air pressure, etc.), according to different measurement conditions, the sensitivity of each gas concentration to environmental factors is calculated through statistical models (such as multiple regression analysis). These influence coefficients reflect the degree of influence of environmental factors on the gas concentration measurement results, and can reasonably correct the gas concentration according to real-time environmental data. The influence coefficient is obtained through long-term experimental data and environmental factor analysis, combined with the relationship between gas concentration and these factors, to obtain accurate values.

[0035] In this implementation scheme, more accurate and dynamic gas leak detection is achieved through the calculation of real-time gas concentration influencing factors. Traditional gas leak detection methods often ignore the impact of concentration changes of different gases on the diffusion process. The system can dynamically calculate the real-time gas concentration influencing factors of each measuring point by combining the real-time acquired methane, ethane, and carbon dioxide concentration data, as well as the gas concentration parameter data of each measuring point, thereby accurately reflecting the impact of gas concentration on the diffusion process. This method fully considers the type and concentration differences of the gas, making the correction of the diffusion model under different concentration conditions more flexible and accurate. In addition, through the adjustment coefficients and influence coefficients stored in the database, the system can optimize the calculation process in combination with historical data and real-time data to ensure that the calculation of concentration influencing factors is more accurate. Ultimately, these calculations help to more accurately predict the diffusion path of the gas, thereby improving the positioning accuracy of the leakage source, ensuring the reliability and adaptability of the system in complex and dynamic environments, reducing positioning errors, and improving detection efficiency and accuracy.

[0036] Specifically, the specific steps for obtaining the real-time environmental impact factor within the set range when the vehicle is traveling are as follows: obtain the air pressure parameter value at each measuring point within the set range when the vehicle is traveling, and perform a comprehensive analysis in combination with the measured air pressure value at the corresponding measuring point to obtain the air pressure impact index within the set range when the vehicle is traveling; pre-process the measured wind speed value, measured turbulence intensity value, measured temperature value, and measured humidity value at each measuring point within the set range when the vehicle is traveling, respectively, and perform a comprehensive analysis after the pre-processing to obtain the environmental fluctuation impact index within the set range when the vehicle is traveling; perform a comprehensive analysis of the air pressure impact index and the environmental fluctuation impact index within the set range when the vehicle is traveling to obtain the real-time environmental impact factor within the set range when the vehicle is traveling.

[0037] The specific formulas for calculating the air pressure impact index, environmental fluctuation impact index, and real-time environmental impact factor within the set range when the vehicle is driving are as follows: ; in, It is the air pressure impact index within the set range when the vehicle is driving. The first The measured air pressure value at each measuring point, The first The air pressure parameter value at each measuring point is is the air pressure influence coefficient stored in the database, It is the environmental fluctuation impact index within the set range when the vehicle is driving. The first The measured wind speed value at each measuring point is is the wind speed influence coefficient stored in the database, The first The measured turbulence intensity value at each measuring point is is the turbulence intensity influence coefficient stored in the database, The first The measured temperature value at each measuring point, is the temperature influence coefficient stored in the database, The first The measured humidity value at each measuring point, is the humidity influence coefficient stored in the database, is the real-time environmental impact factor within the set range when the vehicle is driving. , is the number of measurement points.

[0038] It needs to be explained that the air pressure influence coefficient stored in the database The specific steps of obtaining are: it is obtained by analyzing the correlation between air pressure data and gas leakage diffusion, by collecting historical air pressure data and corresponding gas concentration changes, combining the air pressure changes under different environmental conditions, and using regression analysis or other statistical methods to establish the relationship between air pressure and gas diffusion. The coefficient is stored in the database and can dynamically adjust the calculated value of gas concentration when the air pressure changes, ensuring the accuracy of the model under different air pressure conditions.

[0039] Wind speed influence coefficient stored in the database The specific steps of obtaining are as follows: It is determined by analyzing the relationship between wind speed and gas diffusion path, by collecting gas leakage data under different wind speed conditions, and combining it with the gas diffusion model, performing regression analysis, and calculating the degree of influence of wind speed changes on gas diffusion. This coefficient reflects the influence of wind speed on the speed and direction of gas diffusion, and can be dynamically corrected at different wind speeds to ensure the accuracy of gas concentration simulation.

[0040] Turbulence intensity influence coefficients stored in the database The specific acquisition steps are as follows: obtained through long-term experimental data or field data analysis. The system records the relationship between turbulence intensity and gas concentration distribution, and uses statistical methods (such as least squares regression) to fit it to obtain the influence coefficient of turbulence intensity on gas diffusion. This coefficient is stored in the database and is used to correct the gas diffusion path in the model, especially in areas with high turbulence intensity, to better simulate the diffusion behavior of gas.

[0041] Temperature influence coefficients stored in the database , Humidity influence coefficient The specific steps of obtaining are: by analyzing the influence of temperature and humidity changes on gas diffusion, it is obtained by collecting gas concentration data under different temperature and humidity conditions, and analyzing it in combination with the gas diffusion model, and calculating the influence of temperature and humidity on gas diffusion rate and concentration. The temperature and humidity coefficients are obtained through multiple experiments or historical data analysis, which reflects the correction effect of these environmental factors in gas leakage simulation, ensuring that the diffusion model under different temperature and humidity conditions is more accurate.

[0042] In this implementation scheme, by calculating the real-time environmental impact factors, the system can adapt to dynamic environmental changes more accurately and improve the positioning accuracy of the gas leakage source. Traditional methods usually ignore the influence of environmental factors such as wind speed, air pressure, temperature and humidity, and turbulence intensity on the gas diffusion process. The system obtains the environmental data of each measurement point in real time, calculates the air pressure impact index and the environmental fluctuation impact index, and can dynamically correct the gas diffusion model to reflect the actual diffusion of gas under different environmental conditions. After preprocessing and comprehensive analysis, these environmental impact factors can provide the real-time impact of environmental factors in the gas diffusion process in a timely manner, avoiding the deficiency of static models that cannot cope with environmental changes. Through the influence coefficients and adjustment coefficients stored in the database, the system can be optimized according to the combination of historical data and real-time data to ensure that the calculation of real-time environmental impact factors is more accurate, thereby improving the reliability and accuracy of leakage source positioning. This mechanism based on dynamic correction of environmental factors enables the system to perform efficient and accurate gas leakage detection in various complex and changing environments, greatly improving the adaptability and practicality of the detection system.

[0043] Specifically, based on the adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is driving, and the gas concentration data at each measuring point, the specific steps of inverting the leakage source position coordinates and the leakage rate are as follows: the concentration peak point on the vehicle's moving path is determined as the initial candidate source position, and the specific steps are: recording the methane, ethane and carbon dioxide concentrations at each measuring point, and analyzing the maximum value of the gas concentration at these points as the initial candidate source position.

[0044] The specific formula for analyzing the initial candidate source location is as follows: ;in, is the initial candidate source location, The first The methane concentration value at each measuring point is The first The ethane concentration value at each measuring point is The first The carbon dioxide concentration value at each measuring point is , is the number of measurement points.

[0045] The wind speed direction of the measured wind speed value at each measuring point within the set range when the vehicle is driving is obtained, and the wind field backtracking trajectory probability cloud map is constructed in combination with the corresponding measured wind speed value. The diffusion path of the gas can be regarded as the path of the gas molecules starting from the source point over time. The wind speed vector is used to simulate the diffusion process of the gas starting from a certain moment. Based on the wind speed and wind direction, the movement of the gas can be described by the diffusion equation: ; in, is the gas concentration (such as methane and ethane concentration) at position (x, y, z) and time t, is the wind speed vector at position (x, y, z) and time t, describing the direction and magnitude of the wind speed, is the real-time diffusion correction coefficient within the set range when the vehicle is driving, is the gas leakage source term at position (x, y, z) and time t, which represents the leakage rate of gas at the source position and is used to describe the intensity and temporal and spatial distribution of the leakage source;

[0046] The specific steps to construct the wind field backtracking trajectory probability cloud map are as follows:

[0047] The construction of the backtracking trajectory is carried out through the wind speed vector, in order to determine the diffusion path of the gas and find the possible leakage source area;

[0048] Select an initial point (the concentration peak point on the vehicle's moving path);

[0049] Starting from the location of the leak source, wind speed and direction data are used to estimate the diffusion path of the gas in different time periods.

[0050] And, the specific formula of the probability of backtracking trajectory is as follows: ;in, is the probability of retracing the gas trajectory from time t0 to time t, is the initial candidate source location, is the standard deviation of the diffusion range. As time changes, the range of gas diffusion will increase, so Increases over time.

[0051] Based on the Monte Carlo method, several sets of leakage source parameters are generated, including randomly generating multiple leakage source parameters, evaluating the location and leakage rate of each possible leakage source, and comparing these parameters with the concentration distribution of the model to find the most likely leakage source location. The specific steps are as follows: generate several sets of leakage source parameters, respectively recorded as location =(x,y) and leakage rate Q;

[0052] The concentration distribution of each set of leakage source parameters is analyzed based on the preset diffusion model, and the fit is evaluated with the measured data, including using the real-time diffusion correction coefficient within the set range when the vehicle is driving to calculate the concentration distribution of each set of parameters, and comparing the difference between the concentration calculated by the model and the measured concentration; the specific formula of the preset diffusion model is as follows: ;in, represents the gas concentration calculated by the diffusion model, and the position is , time is t, is the real-time diffusion correction coefficient within the set range when the vehicle is driving, Indicates The location coordinates of the group leakage source parameters, Indicates the location of the target point, that is, the point where the concentration is to be calculated. Through this point, the concentration at a specific time t can be obtained. , is the number of leakage source parameter groups.

[0053] Among them, the specific formula of the objective function (fit evaluation) is as follows: ;in, is the objective function, which is used to evaluate the fit between the model predicted concentration and the actual observed concentration. By minimizing the objective function, the most suitable leakage source location and leakage rate can be found. The objective function is optimized based on minimizing the square of the concentration difference at each measurement point. The goal is to minimize the sum of squares of concentration errors at all points. In this way, the most suitable model parameters (such as leakage source location and leakage rate) can be found. For the Measured gas concentration data of the leak source parameters (e.g. methane, ethane or carbon dioxide concentration), For the The concentration of the group leakage source parameters calculated by the diffusion model is based on the prediction results of the back-tracking trajectory. The first The weight coefficient of the group leakage source parameter.

[0054] The top 10% parameter sets with the highest fitting degree are selected for Gaussian kernel density estimation, and the 95% confidence ellipse area of ​​the leakage source coordinates is output.

[0055] In this implementation, more efficient and accurate gas leakage source positioning is achieved through the combination of precise algorithms and multi-dimensional data. First, the initial candidate source position selection method based on the concentration peak point can quickly locate the potential leakage source area, reducing the uncertainty caused by the overly dispersed or inaccurate detection points in the traditional method. By acquiring gas concentration data (such as methane, ethane, carbon dioxide) and environmental data (such as wind speed, air pressure, temperature and humidity) in real time, the system can dynamically update and correct the diffusion coefficient, so that the gas diffusion simulation is more in line with the changes in the actual environment, thereby improving the accuracy of the diffusion path prediction. Secondly, the probability cloud map of the wind field backtracking trajectory utilizes wind speed vector information, which can infer the gas diffusion trajectory based on real-time wind speed and wind direction data, and accurately determine the source and path of gas diffusion. This method greatly improves the prediction ability of gas diffusion paths under complex environmental conditions and solves the problem of lack of traditional models. In addition, several groups of leakage source parameters are generated by the Monte Carlo method, which can evaluate multiple possible leakage source locations and leakage rates, and fit them with the measured concentration data to ensure the accuracy of the positioning results. In the fitting evaluation, the system adopts the method of minimizing the objective function to automatically optimize the parameters of the leakage source location and leakage rate. This process can effectively eliminate the errors caused by changes in environmental factors, so as to find the most suitable leakage source model, select the top 10% parameter sets with the highest fitting degree and perform Gaussian kernel density estimation, and output the 95% confidence ellipse area of ​​the leakage source location, which further improves the reliability and accuracy of source positioning. Through the combination of this series of steps, the system can accurately predict the location and leakage rate of the gas leakage source, provide efficient emergency response capabilities and positioning accuracy in complex environments, thereby improving the overall performance and application value of the vehicle-mounted gas leak detection system.

[0056] Specifically, the specific steps for identifying the type of leaked gas at the leakage source location based on the preset discrimination rules are as follows: when the methane concentration value in the leaked gas at the leakage source location is higher than or equal to the preset first discrimination threshold value (i.e., CH4 ≥ 20ppb), and the ratio of the methane concentration to the ethane concentration is higher than or equal to the preset second discrimination threshold value (i.e., CH4 / C2H6 ≥ 50), the leaked gas is regarded as pipeline natural gas; when the ethane concentration value in the leaked gas at the leakage source location is lower than or equal to the preset third discrimination threshold value (i.e., C2H6 ≤ 15ppb), and the ratio of the methane concentration to the ethane concentration is higher than or equal to the preset second discrimination threshold value (i.e., CH4 / C2H6 ≥ 50), the leaked gas is regarded as pipeline natural gas; When the ethane concentration ratio result is higher than or equal to the preset fourth discrimination threshold (i.e. CH4 / C2H6≥100), the leaked gas is regarded as biogas; when the ethane concentration value in the leaked gas at the identified leakage source is higher than or equal to the preset first discrimination threshold (i.e. C2H6≥20ppb), and the methane concentration to ethane concentration ratio result is higher than or equal to the preset second discrimination threshold (i.e. CH4 / C2H6≥50), and the carbon dioxide concentration value is higher than the preset fifth discrimination threshold (i.e. CO2>5%), the leaked gas is regarded as gas vehicle emission interference and the interference gas alarm is triggered.

[0057] In this embodiment, through accurate gas type identification, different types of leaked gases can be effectively distinguished, thereby improving the accuracy and emergency response capabilities of the gas leak detection system. Traditional leak detection systems usually only focus on the existence of gas leaks and ignore the type of leaked gas, which may lead to false alarms or failure to identify interfering gases in a timely manner. By setting the concentration ratio of methane to ethane and a specific concentration threshold, the system can accurately determine the type of gas at the leak source. For example, when the methane concentration is high and the ratio to ethane meets specific standards, the system can identify it as pipeline natural gas, and when the ratio of methane concentration to ethane changes, it can also be distinguished as biogas. In addition, when the ethane concentration is high and the carbon dioxide concentration exceeds the set threshold, the system can determine it as gas vehicle emission interference and trigger an alarm to prevent misjudgment of other gas leaks as interference sources. Through this multi-level discrimination rule, the system can not only improve the positioning accuracy of the gas leakage source, but also reduce false alarms caused by interfering gases, thereby enhancing the reliability and practicality of the system in practical applications.

[0058] See also Figure 3The embodiment of the present invention provides a technical solution: a multi-source data fusion vehicle-mounted gas leakage tracing detection method, comprising the following steps: in a data acquisition unit, the position coordinates of several measurement points within a set range when the vehicle is traveling, and the gas concentration data and environmental data at each measurement point are obtained in real time, the gas concentration data includes a methane concentration value, an ethane concentration value, and a carbon dioxide concentration value, and the environmental data includes a measured wind speed value, a measured air pressure value, a measured turbulence intensity value, a measured temperature value, and a measured humidity value; in a diffusion correction unit, the initial diffusion coefficient within the set range when the vehicle is traveling is obtained, and a real-time correction analysis is performed to obtain a real-time diffusion correction coefficient within the set range when the vehicle is traveling; in a leakage source reverse positioning unit, based on an adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is traveling, and the gas concentration data at each measurement point, the leakage source position coordinates and the leakage rate are inverted; in a gas type identification unit, the leaking gas type at the leakage source position is identified based on a preset discrimination rule.

[0059] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0060] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A vehicle-mounted gas leak tracing detection system with multi-source data fusion, characterized in that: include: A data acquisition unit, used to acquire in real time the position coordinates of a number of measurement points within a set range when the vehicle is traveling, as well as gas concentration data and environmental data at each measurement point; The diffusion correction unit is used to obtain the initial diffusion coefficient within a set range when the vehicle is traveling, and perform real-time correction analysis, as follows: Based on the gas concentration data and environmental data at each measuring point within the set range when the vehicle is driving, the real-time gas concentration influencing factor and the real-time environmental influencing factor within the set range when the vehicle is driving are analyzed, and combined with the initial diffusion coefficient within the set range when the vehicle is driving, a comprehensive analysis is performed to obtain the real-time diffusion correction coefficient within the set range when the vehicle is driving. The calculation formula is as follows: ； in, , , , They are, in order, a real-time diffusion correction coefficient, an initial diffusion coefficient, a real-time gas concentration influence factor, and a real-time environment influence factor within a set range when the vehicle is traveling; The leakage source reverse positioning unit is used to invert the leakage source position coordinates and leakage rate based on the adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is driving, and the gas concentration data at each measuring point.

2. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 1 is characterized in that: The gas concentration data includes methane concentration value, ethane concentration value, and carbon dioxide concentration value. The specific steps for obtaining the real-time gas concentration influencing factor within the set range when the vehicle is traveling are as follows: Acquire gas concentration parameter data at each measuring point within a set range when the vehicle is traveling, wherein the gas concentration parameter data includes a methane concentration parameter value, an ethane concentration parameter value, and a carbon dioxide concentration parameter value; The methane concentration value, ethane concentration value, and carbon dioxide concentration value at each measuring point within the set range when the vehicle is traveling are combined with the gas concentration parameter data at the corresponding measuring point for comprehensive analysis to obtain the real-time gas concentration influencing factor within the set range when the vehicle is traveling.

3. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 2 is characterized in that: The specific formula for calculating the real-time gas concentration impact factor within the set range when the vehicle is driving is as follows: ； in, is the real-time gas concentration influencing factor within the set range when the vehicle is driving, , , , , , The first The methane concentration value at each measuring point, the methane concentration parameter value, the ethane concentration value, the ethane concentration parameter value, the carbon dioxide concentration value, the carbon dioxide concentration parameter value, , , They are the methane concentration adjustment coefficient, ethane concentration adjustment coefficient, and carbon dioxide concentration adjustment coefficient stored in the database, respectively. , , They are the methane concentration influence coefficient, ethane concentration influence coefficient, and carbon dioxide concentration influence coefficient stored in the database, respectively. , is the number of measurement points.

4. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 1 is characterized in that: The environmental data includes measured wind speed value, measured air pressure value, measured turbulence intensity value, measured temperature value, and measured humidity value. The specific steps for obtaining the real-time environmental impact factor within the set range when the vehicle is driving are as follows: Obtain the air pressure parameter value at each measuring point within the set range when the vehicle is traveling, and perform a comprehensive analysis in combination with the measured air pressure value at the corresponding measuring point to obtain the air pressure influence index within the set range when the vehicle is traveling; The measured wind speed value, the measured turbulence intensity value, the measured temperature value, and the measured humidity value at each measuring point within the set range when the vehicle is driving are preprocessed respectively, and a comprehensive analysis is performed after the preprocessing to obtain the environmental fluctuation impact index within the set range when the vehicle is driving; A comprehensive analysis is performed on the air pressure impact index and the environmental fluctuation impact index within the set range when the vehicle is traveling, and a real-time environmental impact factor within the set range when the vehicle is traveling is obtained.

5. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 4 is characterized in that: The specific formulas for calculating the air pressure impact index, environmental fluctuation impact index, and real-time environmental impact factor within the set range when the vehicle is driving are as follows: ； in, , , They are the air pressure impact index within the set range when the vehicle is driving, the environmental fluctuation impact index, and the real-time environmental impact factor. , The first The measured air pressure value and air pressure parameter value at each measuring point, , , , The first The measured wind speed value, turbulence intensity value, temperature value and humidity value at each measuring point are , , , , They are the air pressure influence coefficient, wind speed influence coefficient, turbulence intensity influence coefficient, temperature influence coefficient, and humidity influence coefficient stored in the database. , is the number of measurement points.

6. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 1 is characterized in that: Based on the adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is driving, and the gas concentration data at each measurement point, the specific steps for inverting the leak source location coordinates and the leak rate are as follows: Identify the concentration peak point on the vehicle moving path as the initial candidate source position; Obtain the wind speed direction of the measured wind speed value at each measuring point within the set range when the vehicle is traveling, and construct a wind field backtracking trajectory probability cloud map in combination with the corresponding measured wind speed value; Generate several groups of leakage source parameters based on Monte Carlo method; Analyze the concentration distribution of each set of leakage source parameters based on the preset diffusion model, and evaluate the fit with the measured data; The top 10% parameter sets with the highest fitting degree are selected for Gaussian kernel density estimation, and the 95% confidence ellipse area of ​​the leakage source coordinates is output.

7. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 6 is characterized in that: The specific formula for analyzing the initial candidate source position is as follows: ； in, is the initial candidate source location, , , The first The methane concentration value, ethane concentration value, and carbon dioxide concentration value at each measuring point are , is the number of measurement points.

8. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 1 is characterized in that: Also includes: The gas type identification unit is used to identify the leaking gas type at the leakage source location based on a preset discrimination rule.

9. The vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to claim 8 is characterized in that: The specific steps for identifying the leaking gas type at the leak source location based on the preset discrimination rules are as follows: When the methane concentration value of the leaked gas at the identified leak source is higher than or equal to the preset first discrimination threshold, and the ratio of the methane concentration to the ethane concentration is higher than or equal to the preset second discrimination threshold, the leaked gas is regarded as pipeline natural gas; When the ethane concentration value of the leaked gas at the identified leakage source position is lower than or equal to the preset third discrimination threshold, and the ratio of the methane concentration to the ethane concentration is higher than or equal to the preset fourth discrimination threshold, the leaked gas is regarded as biogas; When the ethane concentration value in the leaked gas at the identified leakage source location is higher than or equal to the preset first discrimination threshold, and the ratio of methane concentration to ethane concentration is higher than or equal to the preset second discrimination threshold, and the carbon dioxide concentration value is higher than the preset fifth discrimination threshold, the leaked gas is regarded as gas vehicle emission interference and the interference gas alarm is triggered.

10. A vehicle-mounted gas leak tracing detection method based on multi-source data fusion, using the vehicle-mounted gas leak tracing detection system based on multi-source data fusion according to any one of claims 1 to 9, characterized in that: The following steps are involved: In the data acquisition unit, the position coordinates of several measurement points within a set range when the vehicle is traveling, as well as the gas concentration data and environmental data at each measurement point are acquired in real time, wherein the gas concentration data includes a methane concentration value, an ethane concentration value, and a carbon dioxide concentration value, and the environmental data includes a measured wind speed value, a measured air pressure value, a measured turbulence intensity value, a measured temperature value, and a measured humidity value; In the diffusion correction unit, an initial diffusion coefficient within a set range when the vehicle is traveling is obtained, and a real-time correction analysis is performed to obtain a real-time diffusion correction coefficient within the set range when the vehicle is traveling; In the leakage source reverse positioning unit, based on the adaptive particle swarm optimization algorithm, combined with the real-time diffusion correction coefficient within the set range when the vehicle is driving, and the gas concentration data at each measuring point, the leakage source position coordinates and leakage rate are inverted; In the gas type identification unit, the leaking gas type at the leakage source position is identified based on a preset discrimination rule.

Images