Natural gas pipeline leakage monitoring system and method thereof

Abstract

The invention discloses a natural gas pipeline leakage monitoring system and method, and belongs to the technical field of gas monitoring, and the system comprises a leakage monitoring data acquisition module, an air pressure and airflow correlation analysis module, a leakage condition evaluation module, a leakage period monitoring module and an early warning and response module, the leakage monitoring data acquisition module is used for acquiring leakage monitoring data of each natural gas pipeline section, the air pressure and air flow correlation analysis module is used for acquiring an air pressure-air flow correlation coefficient, and the leakage condition evaluation module is used for evaluating the leakage condition of each natural gas pipeline section; the leakage period monitoring module is used for analyzing the leakage period of each natural gas pipeline section, and the early warning and response module is used for sending out a corresponding early warning signal and taking corresponding measures for response. Real-time accurate monitoring of natural gas pipeline leakage, intelligent analysis in the leakage period and staged early warning response are achieved, and the monitoring intelligent level and safety are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas monitoring, and in particular relates to a natural gas pipeline leakage monitoring system and method. Background Art

[0002] As a vital vehicle for energy transportation, the safe operation of natural gas pipelines is crucial. Currently, traditional manual inspection methods suffer from low efficiency, a high risk of missed inspections, and delayed response times, making them inadequate for the safety requirements of modern long-distance pipelines. While real-time monitoring technologies based on parameters such as pressure and flow have advanced in recent years, single-use monitoring methods are susceptible to interference and have a high false alarm rate.

[0003] With the rapid development of intelligent sensing, the Internet of Things, and big data technologies, the natural gas industry is increasingly in need of intelligent, high-precision, and multi-parameter integrated natural gas pipeline leak monitoring systems. However, existing natural gas pipeline leak monitoring technologies still have some shortcomings: on the one hand, the monitoring of natural gas pipeline leaks is not accurate enough; on the other hand, it is difficult to distinguish the specific stage of natural gas pipeline leaks (such as early, middle, and late stages) by combining multiple factors. This leads to a lack of accurate leakage location and analysis, making targeted maintenance of natural gas pipelines difficult. Summary of the Invention

[0004] The purpose of the present invention is to provide a natural gas pipeline leakage monitoring system and method. Through multi-parameter fusion and artificial intelligence algorithms, it realizes real-time and accurate monitoring of natural gas pipeline leakage, intelligent analysis of the leakage period and phased early warning response, which significantly improves the monitoring intelligence level and safety.

[0005] To achieve the above object, the technical solution adopted by the present invention is: A natural gas pipeline leakage monitoring system includes the following modules in communication connection: The leakage monitoring data acquisition module divides the target natural gas pipeline into several natural gas pipeline sections of equal length and assigns numbers to each natural gas pipeline section to collect leakage monitoring data for each natural gas pipeline section, including natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data, and positioning data. The pressure-airflow correlation analysis module combines natural gas pipeline data with a multivariate linear regression algorithm to construct a pressure-airflow correlation model and obtain the pressure-airflow correlation coefficient. The leakage assessment module uses natural gas pipeline data to calculate the pressure drop gradient and flow balance difference in each natural gas pipeline section. It also uses natural gas pipeline environmental data to calculate the axial soil temperature difference in each natural gas pipeline section. Combining the output of the pressure-airflow correlation model with the leakage gas cloud data, it uses a convolutional neural network algorithm to construct a natural gas pipeline section leakage monitoring model, output the corresponding natural gas pipeline section leakage index, and then evaluate the leakage situation of each natural gas pipeline section. The leakage period monitoring module integrates natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data to calculate the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Combining the natural gas pipeline environmental data with a neural network algorithm, it constructs a natural gas pipeline section leakage period monitoring model, outputs the corresponding natural gas pipeline section leakage period reference index, and then analyzes the leakage period of each natural gas pipeline section. The early warning and response module matches the positioning data with the analysis results of the leakage period of each natural gas pipeline section according to the number of each natural gas pipeline section. Based on the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, it issues corresponding early warning signals and takes corresponding measures to respond.

[0006] Preferably, in the leakage monitoring data acquisition module, the process of collecting leakage monitoring data of each natural gas pipeline section includes: Deploy different types of data collection equipment to collect leakage monitoring data for each natural gas pipeline section, including pressure transmitters, ultrasonic flow meters, temperature transmitters, infrared thermometers, infrared thermal imagers, lidar, soil temperature sensors, industrial-grade acoustic sensors, hyperspectral imagers, multispectral sensors, and GPS receivers; Natural gas pipeline data includes the real-time gas pressure, real-time gas flow, and real-time gas temperature within each natural gas pipeline section, the inlet and outlet flow rates of each natural gas pipeline section, and the real-time temperature of the pipe wall of each natural gas pipeline section. Leakage gas cloud data includes the real-time temperature and real-time plane expansion radius of the leakage gas cloud. Natural gas pipeline environmental data includes the temperature of the soil within 0.5 meters of each natural gas pipeline, the real-time sound wave frequency, red edge displacement, near-infrared band reflectivity, and red light band reflectivity of each natural gas pipeline section. Positioning data includes the latitude and longitude coordinates of each natural gas pipeline section, which are mapped to the natural gas pipeline section number. The collected natural gas pipeline leakage monitoring data is cleaned and normalized, and timestamps are assigned to each natural gas pipeline leakage monitoring data. The timestamps are adjusted to synchronize the collection time of natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data and positioning data. The pre-processed natural gas pipeline leakage monitoring data are integrated to generate a natural gas pipeline leakage monitoring data set.

[0007] Preferably, in the air pressure and air flow correlation analysis module, a process of constructing an air pressure-air flow correlation model, analyzing the correlation between the real-time gas pressure and the real-time gas flow in each natural gas pipeline section, and then obtaining the air pressure-air flow correlation coefficient includes: Extracting the real-time gas pressure and real-time gas flow in each natural gas pipeline section from the natural gas pipeline leakage monitoring data set, and dividing the extracted data into a first training set and a first test set; Using a multiple linear regression algorithm, the first training set data is used as input and the pressure-airflow correlation coefficient is used as output to learn the linear relationship between the real-time gas pressure and real-time gas flow in each natural gas pipeline section and train the pressure-airflow correlation model; The first test set data is input into the pressure-airflow correlation model. The SGD optimizer is used to adjust the regression coefficient and intercept term of the pressure-airflow correlation model to optimize the performance of the pressure-airflow correlation model. The optimized pressure-airflow correlation model is then deployed in a natural gas pipeline leak monitoring system to obtain the final pressure-airflow correlation model. The real-time gas pressure and real-time gas flow in each natural gas pipeline section are input into the air pressure-air flow correlation model, the corresponding air pressure-air flow correlation coefficient is output, and the air pressure-air flow correlation coefficient is integrated into the natural gas pipeline leakage monitoring data set.

[0008] Preferably, in the leakage assessment module, the calculation process of the pressure drop gradient, flow balance difference and axial soil temperature difference of each natural gas pipeline section includes: Based on the real-time gas pressure in each natural gas pipeline section, a graph of the real-time gas pressure change over time is drawn for each natural gas pipeline section. The n1 time point and n2 time point corresponding to the real-time gas pressure drop in each natural gas pipeline section are analyzed and selected. Based on the timestamp, the gas pressure in each natural gas pipeline section corresponding to the n1 time point and the n2 time point is obtained, and the pressure drop gradient of each natural gas pipeline section is calculated. Extract the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section from time point n1 to time point n2 from the inlet flow rate and outlet flow rate of each natural gas pipeline section, and calculate the flow balance difference of each natural gas pipeline section by taking the difference between the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section; The maximum and minimum values are selected from the soil temperatures within 0.5 meters from each natural gas pipeline to obtain the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The axial soil temperature difference of each natural gas pipeline section is calculated by the difference between the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The pressure drop gradient, flow balance difference and axial soil temperature difference of each natural gas pipeline section are integrated into the natural gas pipeline leakage monitoring dataset.

[0009] Preferably, in the leakage assessment module, a process of constructing a natural gas pipeline section leakage monitoring model, outputting a corresponding natural gas pipeline section leakage index, and then assessing the leakage situation of each natural gas pipeline section includes: Extract the pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud for each natural gas pipeline section from the natural gas pipeline leakage monitoring dataset, and convert the extracted data into a second training set and a second test set. Using a convolutional neural network algorithm, the second training set data is used as input and the natural gas pipeline section leakage index is used as output. The nonlinear relationship between the pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and the real-time temperature of the leaking gas cloud in each natural gas pipeline section is learned, and a natural gas pipeline section leakage monitoring model is trained. Input the second test set data into the natural gas pipeline section leakage monitoring model. Using the Adam optimizer, adjust the parameters of the natural gas pipeline section leakage monitoring model to optimize the performance of the natural gas pipeline section leakage monitoring model. Deploy the optimized natural gas pipeline section leakage monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage monitoring model. Combined with the current pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud in each natural gas pipeline section, the corresponding natural gas pipeline section leakage index is output; When the leakage index of a natural gas pipeline section is less than 0.3, it indicates that no leakage occurs in the corresponding natural gas pipeline section; when the leakage index of a natural gas pipeline section is greater than or equal to 0.3, it indicates that leakage occurs in the corresponding natural gas pipeline section.

[0010] Preferably, in the leakage period monitoring module, the calculation process of the pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment includes: Using the real-time gas pressure in each natural gas pipeline section, calculate the gas pressure fluctuation trend rate of each natural gas pipeline section; The gas flow rate in each natural gas pipeline section is corresponding to the time points n1 and n2, and the gas flow rate change rate of each natural gas pipeline section is calculated. The calculation formula is: ,in, is the gas flow rate change rate of each natural gas pipeline section, and are the gas flow rates in each natural gas pipeline section corresponding to time n1 and n2 respectively; Extract the maximum and minimum values of the real-time wall temperature of each natural gas pipeline section from time point n1 to time point n2, obtain the maximum and minimum wall temperature of each natural gas pipeline section, and calculate the wall temperature extreme value of each natural gas pipeline section by taking the difference between the maximum and minimum wall temperature of each natural gas pipeline section; The normalized vegetation index and ratio vegetation index of each natural gas pipeline section are calculated based on the near-infrared band reflectance and red light band reflectance of the environment. Select the time points n1 and n2 that correspond to the plane expansion radius of the leaking gas cloud, and then calculate the plane diffusion speed of the leaking gas cloud; The pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment are integrated into the natural gas pipeline leakage monitoring dataset.

[0011] Preferably, in the leakage period monitoring module, the process of constructing a natural gas pipeline section leakage period monitoring model and then outputting a corresponding natural gas pipeline section leakage period reference index includes: Extract the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes of each natural gas pipeline section from the natural gas pipeline leakage monitoring dataset; the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement, and the planar diffusion velocity of the leaking gas cloud of each natural gas pipeline section; and divide the extracted data into a third training set and a third test set; Using a neural network algorithm, the third training set data is used as input and the natural gas pipeline section leakage period reference index is used as output. The nonlinear relationship between the pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud and the natural gas pipeline section leakage period reference index is learned to train a natural gas pipeline section leakage period monitoring model. Inputting the third test set data into the natural gas pipeline section leakage period monitoring model, adjusting the parameters of the natural gas pipeline section leakage period monitoring model using the Adam optimizer to optimize the performance of the natural gas pipeline section leakage period monitoring model, and deploying the optimized natural gas pipeline section leakage period monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage period monitoring model; The current pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud are input into the natural gas pipeline section leakage period monitoring model to obtain the corresponding natural gas pipeline section leakage period reference index.

[0012] Preferably, in the leakage period monitoring module, the analysis process of the leakage period of each natural gas pipeline section includes: Based on the output of the natural gas pipeline section leakage period monitoring model, the leakage period of each natural gas pipeline section is analyzed accordingly; When the reference index of the natural gas pipeline section leakage period is greater than or equal to 0 and less than 0.2, the corresponding natural gas pipeline section is in the early stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than or equal to 0.2 and less than or equal to 0.6, the corresponding natural gas pipeline section is in the middle stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than 0.6 and less than or equal to 1, the corresponding natural gas pipeline section is in the late stage of leakage; The numbers of the natural gas pipeline sections are matched with the analysis results of the leakage periods of the natural gas pipeline sections.

[0013] Preferably, in the early warning and response module, the process of matching the positioning data with the analysis results of the leakage period of each natural gas pipeline section, issuing corresponding early warning signals based on the evaluation results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, and taking corresponding response measures includes: According to the corresponding relationship between each natural gas pipeline section number and the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period, the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period are matched; When no leakage occurs in each natural gas pipeline section, no early warning signal will be issued and no intervention measures will be taken; alarms will be installed in each natural gas pipeline section. When a leakage occurs in each natural gas pipeline section, if each natural gas pipeline section corresponds to an early leakage, a low-frequency alarm will be issued, the monitoring frequency of each natural gas pipeline leakage monitoring data will be increased, the upstream and downstream valves of the corresponding natural gas pipeline section will be remotely closed, and the pressure of the corresponding natural gas pipeline section will be reduced to a safe value; if each natural gas pipeline section corresponds to a mid-term leakage, a medium-frequency alarm will be issued, and a liquid nitrogen tanker will be called to perform local freezing and solidification on the corresponding natural gas pipeline section, and temporary leakage control will be carried out by combining rapid plugging glue technology and airbag sealing technology, and adsorption materials will be laid in the corresponding natural gas pipeline section; if each natural gas pipeline section corresponds to a late-stage leakage, a high-frequency alarm will be issued, the emergency evacuation plan will be activated, the explosion suppression water curtain system will be used to block the spread of the gas cloud, the leaking gas cloud will be controlled and burned, and an environmental monitoring vehicle will be deployed to monitor the air quality in real time.

[0014] A natural gas pipeline leakage monitoring method, based on the above system, includes the following steps: Divide the target natural gas pipeline into several natural gas pipeline sections of equal length, assign a number to each natural gas pipeline section, and collect leakage monitoring data for each natural gas pipeline section, including natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data, and positioning data; Using natural gas pipeline data, calculate the pressure drop gradient and flow balance difference of each natural gas pipeline section; using natural gas pipeline environmental data, calculate the axial soil temperature difference of each natural gas pipeline section; Combining natural gas pipeline data with a multivariate linear regression algorithm, a pressure-airflow correlation model was constructed to analyze the correlation between real-time gas pressure and real-time gas flow within each natural gas pipeline section, thereby obtaining the pressure-airflow correlation coefficient. Based on the output of the pressure-airflow correlation model, combined with leaking gas cloud data and the pressure drop gradient, flow balance difference, and axial soil temperature difference of each natural gas pipeline section, a convolutional neural network algorithm was used to construct a natural gas pipeline section leakage monitoring model. The model outputs the corresponding natural gas pipeline section leakage index, which is then used to evaluate the leakage situation of each natural gas pipeline section. By integrating natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data, the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section are calculated, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Based on the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaking gas cloud in each natural gas pipeline section environment, combined with natural gas pipeline environmental data and a neural network algorithm, a natural gas pipeline section leakage period monitoring model was constructed, and the corresponding natural gas pipeline section leakage period reference index was output. Analyze the leakage period of each natural gas pipeline section based on the output results of the natural gas pipeline section leakage period monitoring model; According to the number of each natural gas pipeline section, the positioning data is matched with the analysis results of the leakage period of each natural gas pipeline section. According to the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, corresponding early warning signals are issued and corresponding measures are taken.

[0015] The beneficial effects of the present invention are: The present invention can comprehensively and accurately monitor the leakage of natural gas pipelines through the collaborative work of multiple modules, including data acquisition, air pressure and airflow analysis, leakage assessment, period monitoring and early warning response; by using advanced algorithms such as multivariate linear regression and convolutional neural networks, combined with data collected by multiple sensors, it realizes real-time assessment and precise positioning of pipeline leaks, solves the problem of insufficient accuracy of traditional monitoring technology, effectively improves the intelligence level and reliability of monitoring, and provides a strong guarantee for the safe operation of natural gas pipelines.

[0016] This invention accurately analyzes the stage of a pipeline leak and implements targeted early warning and response measures for the early, mid, and late stages of a leak. This phased early warning mechanism not only improves the timeliness and effectiveness of responses, but also reduces accident risks, energy waste, and environmental pollution. Through intelligent monitoring and early warning, the system significantly enhances the intelligence level of natural gas pipeline leak monitoring, providing an efficient and reliable solution for modern pipeline safety management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a principle block diagram of the system of the present invention; Figure 2 Schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention are described clearly and completely below with reference to the accompanying drawings.

[0019] Example 1, as Figure 1 As shown, a natural gas pipeline leakage monitoring system includes a leakage monitoring data acquisition module, an air pressure and airflow correlation analysis module, a leakage situation assessment module, a leakage period monitoring module and an early warning and response module, wherein each module is communicatively connected; The leakage monitoring data acquisition module divides the target natural gas pipeline into several natural gas pipeline sections of equal length and assigns a number to each natural gas pipeline section to collect leakage monitoring data for each natural gas pipeline section. The leakage monitoring data for each natural gas pipeline section includes natural gas pipeline data, leaking gas cloud data, natural gas pipeline environmental data, and positioning data, providing a data basis for subsequent monitoring of the leakage status and leakage period of the natural gas pipeline section. The pressure-airflow correlation analysis module combines natural gas pipeline data with a multivariate linear regression algorithm to construct a pressure-airflow correlation model and obtain the pressure-airflow correlation coefficient. This provides a basis for the subsequent construction of a natural gas pipeline section leakage monitoring model and improves the accuracy of leakage assessment for each natural gas pipeline section. The leakage assessment module uses natural gas pipeline data to calculate the pressure drop gradient and flow balance difference in each natural gas pipeline section. It also uses natural gas pipeline environmental data to calculate the axial soil temperature difference in each natural gas pipeline section. Combining the output of the pressure-airflow correlation model with the leakage gas cloud data and natural gas pipeline environmental data, it uses a convolutional neural network algorithm to construct a natural gas pipeline section leakage monitoring model, output the corresponding natural gas pipeline section leakage index, and then evaluate the leakage situation of each natural gas pipeline section. The leakage period monitoring module integrates natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data to calculate the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Combining the natural gas pipeline environmental data with a neural network algorithm, it constructs a natural gas pipeline section leakage period monitoring model, outputs the corresponding natural gas pipeline section leakage period reference index, and then analyzes the leakage period of each natural gas pipeline section, thereby achieving accurate analysis of the leakage period of each natural gas pipeline section. The early warning and response module matches the positioning data with the analysis results of the leakage period of each natural gas pipeline section according to the number of each natural gas pipeline section. Based on the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, it issues corresponding early warning signals and takes corresponding measures to respond.

[0020] In the leakage monitoring data acquisition module, the collection process of leakage monitoring data for each natural gas pipeline section includes: Deploy different types of data collection equipment to collect leakage monitoring data for each natural gas pipeline section, including pressure transmitters, ultrasonic flow meters, temperature transmitters, infrared thermometers, infrared thermal imagers, lidar, soil temperature sensors, industrial-grade acoustic sensors, hyperspectral imagers, multispectral sensors, and GPS receivers; Natural gas pipeline data includes the real-time gas pressure, real-time gas flow, and real-time gas temperature within each natural gas pipeline section, the inlet and outlet flow rates of each natural gas pipeline section, and the real-time temperature of the pipe wall of each natural gas pipeline section. Leakage gas cloud data includes the real-time temperature and real-time plane expansion radius of the leakage gas cloud. Natural gas pipeline environmental data includes the temperature of the soil within 0.5 meters of each natural gas pipeline, the real-time sound wave frequency, red edge displacement, near-infrared band reflectivity, and red light band reflectivity of each natural gas pipeline section. Positioning data includes the latitude and longitude coordinates of each natural gas pipeline section, which are mapped to the natural gas pipeline section number. Pressure transmitters, temperature transmitters, strain gauge sensors, vibration monitors, and infrared thermometers are used to collect real-time gas pressure and temperature, real-time metal stress, and real-time vibration frequency and amplitude within each natural gas pipeline section. Ultrasonic flow meters are used to collect real-time gas flow within each natural gas pipeline section, as well as the inlet and outlet flows of each natural gas pipeline section. Infrared thermal imagers and lidar are used to collect the real-time temperature and plane expansion radius of the leaking gas cloud. Soil temperature sensors, industrial-grade acoustic sensors, hyperspectral imagers, and multispectral sensors are used to collect the soil temperature within 0.5 meters of each natural gas pipeline, the real-time sound wave frequency, red edge displacement, near-infrared band reflectivity, and red light band reflectivity of each natural gas pipeline section. GPS receivers are used to collect the longitude and latitude coordinates of each natural gas pipeline section. The collected natural gas pipeline leakage monitoring data is cleaned and normalized, and timestamps are assigned to each natural gas pipeline leakage monitoring data. The timestamps are adjusted to synchronize the collection time of natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data and positioning data. The pre-processed natural gas pipeline leakage monitoring data are integrated to generate a natural gas pipeline leakage monitoring data set.

[0021] In the pressure-airflow correlation analysis module, a pressure-airflow correlation model is constructed to analyze the correlation between the real-time gas pressure and real-time gas flow in each natural gas pipeline section. The process of obtaining the pressure-airflow correlation coefficient includes the following: Extract the real-time gas pressure and real-time gas flow in each natural gas pipeline section from the natural gas pipeline leakage monitoring dataset, and divide the extracted data into a first training set and a first test set in a ratio of 8:2; Using a multiple linear regression algorithm, the first training set data is used as input and the pressure-airflow correlation coefficient is used as output to learn the linear relationship between the real-time gas pressure and real-time gas flow in each natural gas pipeline section and train the pressure-airflow correlation model. The pressure-airflow correlation coefficient describes the linear relationship between the real-time gas pressure and real-time gas flow in each natural gas pipeline section. The first test set data is input into the pressure-airflow correlation model. The SGD optimizer (stochastic gradient descent optimizer) is used to adjust the regression coefficient and intercept term of the pressure-airflow correlation model to optimize the performance of the pressure-airflow correlation model. The optimized pressure-airflow correlation model is then deployed in a natural gas pipeline leak monitoring system to obtain the final pressure-airflow correlation model. Input the current real-time gas pressure and real-time gas flow in each natural gas pipeline section into the gas pressure-airflow correlation model, output the corresponding gas pressure-airflow correlation coefficient, and integrate the gas pressure-airflow correlation coefficient into the natural gas pipeline leakage monitoring data set; The expression of the pressure-airflow correlation model is: ; Where T is the pressure-airflow correlation coefficient, and They are the real-time gas pressure and real-time gas flow in each natural gas pipeline section, and are the regression coefficients of the real-time gas pressure and real-time gas flow in each natural gas pipeline section, and are the intercept term and error term of the pressure-airflow correlation model, respectively.

[0022] In the leakage assessment module, the calculation process of the pressure drop gradient, flow balance difference, and axial soil temperature difference for each natural gas pipeline section includes: Based on the real-time gas pressure in each natural gas pipeline section, a graph of the real-time gas pressure change over time is drawn. The n1 time point and n2 time point corresponding to the real-time gas pressure drop in each natural gas pipeline section are analyzed and selected (the n1 time point when the real-time gas pressure starts to drop and the n2 time point when the drop ends). Based on the timestamp, the gas pressure in each natural gas pipeline section corresponding to the n1 time point and the n2 time point is obtained, and then the pressure drop gradient of each natural gas pipeline section is calculated. The calculation process is as follows: ; in, The pressure drop gradient of each natural gas pipeline section describes the gas pressure drop rate; and The time points n1 and n2 correspond to the gas pressure in each natural gas pipeline section respectively; and n2 are the selected time points; Extract the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section from time point n1 to time point n2 from the inlet flow rate and outlet flow rate of each natural gas pipeline section, and calculate the flow balance difference of each natural gas pipeline section by taking the difference between the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section; The maximum and minimum values are selected from the soil temperatures within 0.5 meters from each natural gas pipeline to obtain the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The axial soil temperature difference of each natural gas pipeline section is calculated by the difference between the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The pressure drop gradient, flow balance difference and axial soil temperature difference of each natural gas pipeline section are integrated into the natural gas pipeline leakage monitoring dataset.

[0023] In the leakage assessment module, a natural gas pipeline section leakage monitoring model is constructed, and the corresponding natural gas pipeline section leakage index is output. The process of evaluating the leakage situation of each natural gas pipeline section includes the following: The pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud were extracted from the natural gas pipeline leakage monitoring dataset for each natural gas pipeline section. The extracted data were converted into a second training set and a second test set, with a data ratio of 7:3 between the second training set and the second test set. Using a convolutional neural network algorithm, the second training set data is used as input and the natural gas pipeline section leakage index is used as output. The nonlinear relationship between the pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and the real-time temperature of the leaking gas cloud in each natural gas pipeline section is learned, and a natural gas pipeline section leakage monitoring model is trained. Input the second test set data into the natural gas pipeline section leakage monitoring model. Using the Adam optimizer, adjust the parameters of the natural gas pipeline section leakage monitoring model to optimize the performance of the natural gas pipeline section leakage monitoring model. Deploy the optimized natural gas pipeline section leakage monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage monitoring model. Combined with the current pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud in each natural gas pipeline section, the corresponding natural gas pipeline section leakage index is output; When the leakage index of a natural gas pipeline section is less than 0.3, it indicates that no leakage occurs in the corresponding natural gas pipeline section; when the leakage index of a natural gas pipeline section is greater than or equal to 0.3, it indicates that leakage occurs in the corresponding natural gas pipeline section.

[0024] In the leakage monitoring module, the calculation process of the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extreme value of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment includes: The real-time gas pressure in each natural gas pipeline section is used to calculate the pressure fluctuation trend rate of each natural gas pipeline section. The calculation formula is: ; in, is the pressure fluctuation trend rate of each natural gas pipeline section, describing the pressure fluctuation trend rate of each natural gas pipeline section within 24 hours a day; m is an integer between 0 and 24, representing 24 hours a day; is the gas pressure of each natural gas pipeline section corresponding to the mth hour; The gas flow rate in each natural gas pipeline section is corresponding to the time points n1 and n2, and the gas flow rate change rate of each natural gas pipeline section is calculated. The calculation formula is: ,in, is the gas flow rate change rate of each natural gas pipeline section, and are the gas flow rates in each natural gas pipeline section corresponding to time n1 and n2 respectively; Extract the maximum and minimum values of the real-time wall temperature of each natural gas pipeline section from time point n1 to time point n2, obtain the maximum and minimum wall temperature of each natural gas pipeline section, and calculate the wall temperature extreme value of each natural gas pipeline section by taking the difference between the maximum and minimum wall temperature of each natural gas pipeline section; The normalized vegetation index and ratio vegetation index of each natural gas pipeline section are calculated based on the near-infrared band reflectance and red light band reflectance of the environment. The calculation process is as follows: ; ; in, and They are the normalized vegetation index and ratio vegetation index of each natural gas pipeline section environment; and They are the near-infrared band reflectivity and red light band reflectivity of the environment of each natural gas pipeline section; The plane expansion radius of the leaked gas cloud corresponding to the time points n1 and n2 is selected, and the plane diffusion velocity of the leaked gas cloud is calculated. The calculation formula is as follows: ; in, is the plane diffusion velocity of the leaking gas cloud, and are the plane expansion radii of the leaked gas cloud corresponding to time points n1 and n2, respectively; The pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment are integrated into the natural gas pipeline leakage monitoring dataset.

[0025] In the leakage period monitoring module, the process of constructing a natural gas pipeline section leakage period monitoring model and then outputting the corresponding natural gas pipeline section leakage period reference index includes: From the natural gas pipeline leakage monitoring data set, we extracted the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement, and the planar diffusion velocity of the leaking gas cloud for each natural gas pipeline section. The extracted data were then divided into a third training set and a third test set, with the ratio of the third training set data to the third test set data being 8:2. Using a neural network algorithm, the third training set data is used as input and the natural gas pipeline section leakage period reference index is used as output. The nonlinear relationship between the pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud and the natural gas pipeline section leakage period reference index is learned to train a natural gas pipeline section leakage period monitoring model. Inputting the third test set data into the natural gas pipeline section leakage period monitoring model, adjusting the parameters of the natural gas pipeline section leakage period monitoring model using the Adam optimizer to optimize the performance of the natural gas pipeline section leakage period monitoring model, and deploying the optimized natural gas pipeline section leakage period monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage period monitoring model; The current pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud are input into the natural gas pipeline section leakage period monitoring model to obtain the corresponding natural gas pipeline section leakage period reference index.

[0026] In the leakage period monitoring module, the analysis process of the leakage period of each natural gas pipeline section includes: Based on the output of the natural gas pipeline section leakage period monitoring model, the leakage period of each natural gas pipeline section is analyzed accordingly; When the reference index of the natural gas pipeline section leakage period is greater than or equal to 0 and less than 0.2, the corresponding natural gas pipeline section is in the early stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than or equal to 0.2 and less than or equal to 0.6, the corresponding natural gas pipeline section is in the middle stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than 0.6 and less than or equal to 1, the corresponding natural gas pipeline section is in the late stage of leakage; The numbers of the natural gas pipeline sections are matched with the analysis results of the leakage periods of the natural gas pipeline sections.

[0027] In the early warning and response module, the positioning data is matched with the analysis results of the leakage period of each natural gas pipeline section. Based on the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, corresponding early warning signals are issued and corresponding measures are taken. The process includes: According to the corresponding relationship between each natural gas pipeline section number and the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period, the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period are matched; When no leak occurs in a natural gas pipeline section, no warning signal is issued and no intervention measures are taken. Alarms are installed in each natural gas pipeline section. When a leak occurs in each natural gas pipeline section, if the natural gas pipeline section corresponds to an early leak, a low-frequency alarm sounds, specifically once every 10 seconds. The monitoring frequency of each natural gas pipeline leak monitoring data is increased, and the upstream and downstream valves of the corresponding natural gas pipeline section are remotely closed to reduce the pressure of the corresponding natural gas pipeline section to a safe value. If the natural gas pipeline section corresponds to a mid-term leak, a medium-frequency alarm sounds, specifically once every 3 seconds. Liquid nitrogen tankers are deployed to locally freeze and solidify the corresponding natural gas pipeline section. A combination of rapid plugging glue technology and airbag sealing technology is used for temporary leak control. Adsorbent materials are laid in the corresponding natural gas pipeline section to prevent the spread of pollution. If the natural gas pipeline section corresponds to a late leak, a high-frequency alarm sounds, specifically once every 1 second, and the emergency evacuation plan is activated. An explosion suppression water curtain system is used to block the spread of the gas cloud. The leaking gas cloud is controlled and burned to prevent violent explosions. Environmental monitoring vehicles are deployed to monitor air quality in real time.

[0028] Example 2, as Figure 2 As shown, a natural gas pipeline leakage monitoring method, based on the system in Example 1, includes the following steps: Divide the target natural gas pipeline into several natural gas pipeline sections of equal length, assign a number to each natural gas pipeline section, and collect leakage monitoring data for each natural gas pipeline section, including natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data, and positioning data; Using natural gas pipeline data, calculate the pressure drop gradient and flow balance difference of each natural gas pipeline section; using natural gas pipeline environmental data, calculate the axial soil temperature difference of each natural gas pipeline section; Combining natural gas pipeline data with a multivariate linear regression algorithm, a pressure-airflow correlation model was constructed to analyze the correlation between real-time gas pressure and real-time gas flow within each natural gas pipeline section, thereby obtaining the pressure-airflow correlation coefficient. Based on the output of the pressure-airflow correlation model, combined with leaking gas cloud data and the pressure drop gradient, flow balance difference, and axial soil temperature difference of each natural gas pipeline section, a convolutional neural network algorithm was used to construct a natural gas pipeline section leakage monitoring model. The model outputs the corresponding natural gas pipeline section leakage index, which is then used to evaluate the leakage situation of each natural gas pipeline section. By integrating natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data, the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section are calculated, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Based on the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaking gas cloud in each natural gas pipeline section environment, combined with natural gas pipeline environmental data and a neural network algorithm, a natural gas pipeline section leakage period monitoring model was constructed, and the corresponding natural gas pipeline section leakage period reference index was output. Analyze the leakage period of each natural gas pipeline section based on the output results of the natural gas pipeline section leakage period monitoring model; According to the number of each natural gas pipeline section, the positioning data is matched with the analysis results of the leakage period of each natural gas pipeline section. According to the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, corresponding early warning signals are issued and corresponding measures are taken.

Claims

1. A natural gas pipeline leakage monitoring system, characterized in that: Includes communication connections for the following modules: The leakage monitoring data acquisition module divides the target natural gas pipeline into several natural gas pipeline sections of equal length and assigns numbers to each natural gas pipeline section to collect leakage monitoring data for each natural gas pipeline section, including natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data, and positioning data. The pressure-airflow correlation analysis module combines natural gas pipeline data with a multivariate linear regression algorithm to construct a pressure-airflow correlation model and obtain the pressure-airflow correlation coefficient. The leakage assessment module uses natural gas pipeline data to calculate the pressure drop gradient and flow balance difference in each natural gas pipeline section. It also uses natural gas pipeline environmental data to calculate the axial soil temperature difference in each natural gas pipeline section. Combining the output of the pressure-airflow correlation model with the leakage gas cloud data, it uses a convolutional neural network algorithm to construct a natural gas pipeline section leakage monitoring model, output the corresponding natural gas pipeline section leakage index, and then evaluate the leakage situation of each natural gas pipeline section. The leakage period monitoring module integrates natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data to calculate the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Combining the natural gas pipeline environmental data with a neural network algorithm, it constructs a natural gas pipeline section leakage period monitoring model, outputs the corresponding natural gas pipeline section leakage period reference index, and then analyzes the leakage period of each natural gas pipeline section. The early warning and response module matches the positioning data with the analysis results of the leakage period of each natural gas pipeline section according to the number of each natural gas pipeline section. Based on the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, it issues corresponding early warning signals and takes corresponding measures to respond.

2. A natural gas pipeline leakage monitoring system according to claim 1, characterized in that: In the leakage monitoring data acquisition module, the process of collecting leakage monitoring data of each natural gas pipeline section includes: Deploy different types of data collection equipment to collect leakage monitoring data for each natural gas pipeline section, including pressure transmitters, ultrasonic flow meters, temperature transmitters, infrared thermometers, infrared thermal imagers, lidar, soil temperature sensors, industrial-grade acoustic sensors, hyperspectral imagers, multispectral sensors, and GPS receivers; Natural gas pipeline data includes the real-time gas pressure, real-time gas flow, and real-time gas temperature within each natural gas pipeline section, the inlet and outlet flow rates of each natural gas pipeline section, and the real-time temperature of the pipe wall of each natural gas pipeline section. Leakage gas cloud data includes the real-time temperature and real-time plane expansion radius of the leakage gas cloud. Natural gas pipeline environmental data includes the temperature of the soil within 0.5 meters of each natural gas pipeline, the real-time sound wave frequency, red edge displacement, near-infrared band reflectivity, and red light band reflectivity of each natural gas pipeline section. Positioning data includes the latitude and longitude coordinates of each natural gas pipeline section, which are mapped to the natural gas pipeline section number. The collected natural gas pipeline leakage monitoring data is cleaned and normalized, and timestamps are assigned to each natural gas pipeline leakage monitoring data. The timestamps are adjusted to synchronize the collection time of natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data and positioning data. The pre-processed natural gas pipeline leakage monitoring data are integrated to generate a natural gas pipeline leakage monitoring data set.

3. A natural gas pipeline leakage monitoring system according to claim 2, characterized in that: In the air pressure and air flow correlation analysis module, a process of constructing an air pressure-air flow correlation model, analyzing the correlation between the real-time gas pressure and the real-time gas flow in each natural gas pipeline section, and then obtaining the air pressure-air flow correlation coefficient includes: Extracting the real-time gas pressure and real-time gas flow in each natural gas pipeline section from the natural gas pipeline leakage monitoring data set, and dividing the extracted data into a first training set and a first test set; Using a multiple linear regression algorithm, the first training set data is used as input and the pressure-airflow correlation coefficient is used as output to learn the linear relationship between the real-time gas pressure and real-time gas flow in each natural gas pipeline section and train the pressure-airflow correlation model; The first test set data is input into the pressure-airflow correlation model. The SGD optimizer is used to adjust the regression coefficient and intercept term of the pressure-airflow correlation model to optimize the performance of the pressure-airflow correlation model. The optimized pressure-airflow correlation model is then deployed in a natural gas pipeline leak monitoring system to obtain the final pressure-airflow correlation model. The real-time gas pressure and real-time gas flow in each natural gas pipeline section are input into the air pressure-air flow correlation model, the corresponding air pressure-air flow correlation coefficient is output, and the air pressure-air flow correlation coefficient is integrated into the natural gas pipeline leakage monitoring data set.

4. A natural gas pipeline leakage monitoring system according to claim 3, characterized in that: In the leakage assessment module, the calculation process of the pressure drop gradient, flow balance difference and axial soil temperature difference of each natural gas pipeline section includes: Based on the real-time gas pressure in each natural gas pipeline section, a graph of the real-time gas pressure change over time is drawn for each natural gas pipeline section. The n1 time point and n2 time point corresponding to the real-time gas pressure drop in each natural gas pipeline section are analyzed and selected. Based on the timestamp, the gas pressure in each natural gas pipeline section corresponding to the n1 time point and the n2 time point is obtained, and the pressure drop gradient of each natural gas pipeline section is calculated. Extract the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section from time point n1 to time point n2 from the inlet flow rate and outlet flow rate of each natural gas pipeline section, and calculate the flow balance difference of each natural gas pipeline section by taking the difference between the maximum inlet flow rate and the maximum outlet flow rate of each natural gas pipeline section; The maximum and minimum values are selected from the soil temperatures within 0.5 meters from each natural gas pipeline to obtain the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The axial soil temperature difference of each natural gas pipeline section is calculated by the difference between the maximum and minimum temperatures of the soil within 0.5 meters from each natural gas pipeline. The pressure drop gradient, flow balance difference and axial soil temperature difference of each natural gas pipeline section are integrated into the natural gas pipeline leakage monitoring dataset.

5. A natural gas pipeline leakage monitoring system according to claim 4, characterized in that: In the leakage assessment module, a natural gas pipeline section leakage monitoring model is constructed, corresponding natural gas pipeline section leakage index is output, and the process of evaluating the leakage situation of each natural gas pipeline section includes: Extract the pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud for each natural gas pipeline section from the natural gas pipeline leakage monitoring dataset, and convert the extracted data into a second training set and a second test set. Using a convolutional neural network algorithm, the second training set data is used as input and the natural gas pipeline section leakage index is used as output. The nonlinear relationship between the pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and the real-time temperature of the leaking gas cloud in each natural gas pipeline section is learned, and a natural gas pipeline section leakage monitoring model is trained. Input the second test set data into the natural gas pipeline section leakage monitoring model. Using the Adam optimizer, adjust the parameters of the natural gas pipeline section leakage monitoring model to optimize the performance of the natural gas pipeline section leakage monitoring model. Deploy the optimized natural gas pipeline section leakage monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage monitoring model. Combined with the current pressure drop gradient, flow balance difference, axial soil temperature difference, pressure-airflow correlation coefficient, and real-time temperature of the leaking gas cloud in each natural gas pipeline section, the corresponding natural gas pipeline section leakage index is output; When the leakage index of a natural gas pipeline section is less than 0.3, it indicates that no leakage occurs in the corresponding natural gas pipeline section; when the leakage index of a natural gas pipeline section is greater than or equal to 0.3, it indicates that leakage occurs in the corresponding natural gas pipeline section.

6. A natural gas pipeline leakage monitoring system according to claim 5, characterized in that: In the leakage period monitoring module, the calculation process of the pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment includes: Using the real-time gas pressure in each natural gas pipeline section, calculate the gas pressure fluctuation trend rate of each natural gas pipeline section; The gas flow rate in each natural gas pipeline section is respectively selected at the time points n1 and n2, and the gas flow rate change rate of each natural gas pipeline section is calculated. The calculation formula is: ,in, is the gas flow rate change rate of each natural gas pipeline section, and are the gas flow rates in each natural gas pipeline section corresponding to time n1 and n2 respectively; Extract the maximum and minimum values of the real-time wall temperature of each natural gas pipeline section from time point n1 to time point n2, obtain the maximum and minimum wall temperature of each natural gas pipeline section, and calculate the wall temperature extreme value of each natural gas pipeline section by taking the difference between the maximum and minimum wall temperature of each natural gas pipeline section; The normalized vegetation index and ratio vegetation index of each natural gas pipeline section are calculated based on the near-infrared band reflectance and red light band reflectance of the environment. Select the time points n1 and n2 that correspond to the plane expansion radius of the leaking gas cloud, and then calculate the plane diffusion speed of the leaking gas cloud; The pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index and plane diffusion velocity of the leaking gas cloud of each natural gas pipeline section environment are integrated into the natural gas pipeline leakage monitoring dataset.

7. A natural gas pipeline leakage monitoring system according to claim 6, characterized in that: In the leakage period monitoring module, the process of constructing a natural gas pipeline section leakage period monitoring model and then outputting a corresponding natural gas pipeline section leakage period reference index includes: Extract the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes of each natural gas pipeline section from the natural gas pipeline leakage monitoring dataset; the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement, and the planar diffusion velocity of the leaking gas cloud of each natural gas pipeline section; and divide the extracted data into a third training set and a third test set; Using a neural network algorithm, the third training set data is used as input and the natural gas pipeline section leakage period reference index is used as output. The nonlinear relationship between the pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud and the natural gas pipeline section leakage period reference index is learned to train a natural gas pipeline section leakage period monitoring model. Inputting the third test set data into the natural gas pipeline section leakage period monitoring model, adjusting the parameters of the natural gas pipeline section leakage period monitoring model using the Adam optimizer to optimize the performance of the natural gas pipeline section leakage period monitoring model, and deploying the optimized natural gas pipeline section leakage period monitoring model into a natural gas pipeline leakage monitoring system to obtain the final natural gas pipeline section leakage period monitoring model; The current pressure fluctuation trend rate, gas flow rate change rate and pipe wall temperature extreme value of each natural gas pipeline section, the normalized vegetation index, ratio vegetation index, real-time acoustic wave frequency and red edge displacement of each natural gas pipeline section environment, and the plane diffusion velocity of the leaking gas cloud are input into the natural gas pipeline section leakage period monitoring model to obtain the corresponding natural gas pipeline section leakage period reference index.

8. A natural gas pipeline leakage monitoring system according to claim 7, characterized in that: In the leakage period monitoring module, the analysis process of the leakage period of each natural gas pipeline section includes: Based on the output of the natural gas pipeline section leakage period monitoring model, the leakage period of each natural gas pipeline section is analyzed accordingly; When the reference index of the natural gas pipeline section leakage period is greater than or equal to 0 and less than 0.2, the corresponding natural gas pipeline section is in the early stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than or equal to 0.2 and less than or equal to 0.6, the corresponding natural gas pipeline section is in the middle stage of leakage; when the reference index of the natural gas pipeline section leakage period is greater than 0.6 and less than or equal to 1, the corresponding natural gas pipeline section is in the late stage of leakage; The numbers of the natural gas pipeline sections are matched with the analysis results of the leakage periods of the natural gas pipeline sections.

9. A natural gas pipeline leakage monitoring system according to claim 8, characterized in that: In the early warning and response module, the positioning data is matched with the analysis results of the leakage period of each natural gas pipeline section, and corresponding early warning signals are issued based on the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, and corresponding response measures are taken. The process includes: According to the corresponding relationship between each natural gas pipeline section number and the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period, the longitude and latitude coordinates of each natural gas pipeline section and the analysis results of each natural gas pipeline section leakage period are matched; When no leakage occurs in each natural gas pipeline section, no early warning signal will be issued and no intervention measures will be taken; alarms will be installed in each natural gas pipeline section. When a leakage occurs in each natural gas pipeline section, if each natural gas pipeline section corresponds to an early leakage, a low-frequency alarm will be issued, the monitoring frequency of each natural gas pipeline leakage monitoring data will be increased, the upstream and downstream valves of the corresponding natural gas pipeline section will be remotely closed, and the pressure of the corresponding natural gas pipeline section will be reduced to a safe value; if each natural gas pipeline section corresponds to a mid-term leakage, a medium-frequency alarm will be issued, and a liquid nitrogen tanker will be called to perform local freezing and solidification on the corresponding natural gas pipeline section, and temporary leakage control will be carried out by combining rapid plugging glue technology and airbag sealing technology, and adsorption materials will be laid in the corresponding natural gas pipeline section; if each natural gas pipeline section corresponds to a late-stage leakage, a high-frequency alarm will be issued, the emergency evacuation plan will be activated, the explosion suppression water curtain system will be used to block the spread of the gas cloud, the leaking gas cloud will be controlled and burned, and an environmental monitoring vehicle will be deployed to monitor the air quality in real time.

10. A natural gas pipeline leakage monitoring method, based on the system according to any one of claims 1 to 9, characterized in that: The following steps are involved: Divide the target natural gas pipeline into several natural gas pipeline sections of equal length, assign a number to each natural gas pipeline section, and collect leakage monitoring data for each natural gas pipeline section, including natural gas pipeline data, leakage gas cloud data, natural gas pipeline environmental data, and positioning data; Using natural gas pipeline data, calculate the pressure drop gradient and flow balance difference of each natural gas pipeline section; using natural gas pipeline environmental data, calculate the axial soil temperature difference of each natural gas pipeline section; Combining natural gas pipeline data with a multivariate linear regression algorithm, a pressure-airflow correlation model was constructed to analyze the correlation between real-time gas pressure and real-time gas flow within each natural gas pipeline section, thereby obtaining the pressure-airflow correlation coefficient. Based on the output of the pressure-airflow correlation model, combined with leaking gas cloud data and the pressure drop gradient, flow balance difference, and axial soil temperature difference of each natural gas pipeline section, a convolutional neural network algorithm was used to construct a natural gas pipeline section leakage monitoring model. The model outputs the corresponding natural gas pipeline section leakage index, which is then used to evaluate the leakage situation of each natural gas pipeline section. By integrating natural gas pipeline data, natural gas pipeline environmental data, and leaked gas cloud data, the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes for each natural gas pipeline section are calculated, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaked gas cloud for each natural gas pipeline section. Based on the pressure fluctuation trend rate, gas flow rate change rate, and pipe wall temperature extremes of each natural gas pipeline section, as well as the normalized vegetation index, ratio vegetation index, and planar diffusion velocity of the leaking gas cloud in each natural gas pipeline section environment, combined with natural gas pipeline environmental data and a neural network algorithm, a natural gas pipeline section leakage period monitoring model was constructed, and the corresponding natural gas pipeline section leakage period reference index was output. Analyze the leakage period of each natural gas pipeline section based on the output results of the natural gas pipeline section leakage period monitoring model; According to the number of each natural gas pipeline section, the positioning data is matched with the analysis results of the leakage period of each natural gas pipeline section. According to the assessment results of the leakage situation of each natural gas pipeline section and the analysis results of different leakage periods, corresponding early warning signals are issued and corresponding measures are taken.

Images