Hydrogen-doped natural gas pipeline leakage monitoring method, device, equipment, medium and product

Abstract

The invention provides a hydrogen-doped natural gas pipeline leakage monitoring method and device, equipment, a medium and a product, and relates to the technical field of energy monitoring, and the method comprises the steps: obtaining a to-be-detected infrared image; the to-be-detected infrared image is obtained by shooting the to-be-monitored hydrogen-doped natural gas pipeline with gas leakage; and inputting the to-be-detected infrared image into the gas leakage monitoring model to obtain a leakage risk level of the to-be-monitored hydrogen-doped natural gas pipeline output by the gas leakage monitoring model. By means of the mode, reliable leakage monitoring of the hydrogen-doped natural gas pipeline under different conditions is achieved, meanwhile, accurate division of the leakage risk levels is achieved, workers can conveniently conduct corresponding emergency management according to the leakage risk levels, and the requirements of leakage control and accident site emergency management can be met.

Description

Technical Field

[0001] This invention relates to the field of energy monitoring technology, and in particular to a method, apparatus, equipment, medium, and product for monitoring leaks in hydrogen-blended natural gas pipelines. Background Technology

[0002] Hydrogen energy has attracted widespread attention due to its cleanliness, environmental friendliness, high efficiency, and wide applicability, and is considered an ideal alternative to fossil fuels. Natural gas's main component is methane. Because hydrogen and methane have significant differences in their physicochemical properties, transporting hydrogen through existing natural gas pipelines can adversely affect pipeline metal materials and high-strength steel infrastructure. In particular, the weld seams of natural gas pipelines are inherently weaker than the surrounding base material, making them prone to defects and residual stress. This makes the impact of hydrogen on the weld area particularly pronounced; the higher the strength of the pipeline steel, the more susceptible the natural gas pipeline is to hydrogen embrittlement. Therefore, the addition of hydrogen makes existing natural gas pipelines more prone to leaks. Furthermore, when transporting a mixture of hydrogen and methane (i.e., hydrogen-blended natural gas) through natural gas pipelines, the lower ignition energy, faster combustion speed, and higher flame velocity of the mixture compared to a single-component gas increase the likelihood of explosions. Given the high risk and severe consequences of hydrogen-blended natural gas leaks, studying the leakage patterns of hydrogen-blended natural gas and monitoring leaks in hydrogen-blended natural gas pipelines is essential.

[0003] Because gas leak monitoring involves the intersection of multiple disciplines and technologies, different leak monitoring technologies for hydrogen-blended natural gas pipelines vary considerably. For example, sensors used for leak monitoring include catalytic combustion sensors, electrochemical sensors, and semiconductor sensors. The essence of each leak monitoring technology is almost the same: to identify leak signals, make corresponding conversions and reactions, and send abnormal signals to the outside world.

[0004] However, due to the high complexity of leak monitoring in hydrogen-blended natural gas pipelines, such as the complexity of the environment in which the facility is located and the diversity of leak forms and degrees, leak monitoring by simply identifying leak signals is unreliable. This leak monitoring method is difficult to meet the needs of leak control and emergency management at accident sites.

[0005] Currently, there is no relatively reliable leak detection method that can detect leaks in hydrogen-blended natural gas pipelines under different conditions. Summary of the Invention

[0006] This invention provides a method, apparatus, equipment, medium, and product for monitoring leaks in hydrogen-blended natural gas pipelines, thereby addressing the deficiency in the prior art where there is no relatively reliable leak monitoring method that can monitor leaks in hydrogen-blended natural gas pipelines under different conditions.

[0007] This invention provides a method for monitoring leaks in hydrogen-blended natural gas pipelines, comprising: acquiring an infrared image to be detected; the infrared image to be detected is obtained by photographing the hydrogen-blended natural gas pipeline to be monitored where a gas leak exists; inputting the infrared image to be detected into a gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored output by the gas leak monitoring model; wherein, the gas leak monitoring model is trained based on sample infrared images, sample gas type labels corresponding to the sample infrared images, and sample leakage risk level labels corresponding to the sample infrared images.

[0008] According to the present invention, a method for monitoring leaks in hydrogen-blended natural gas pipelines is provided. The gas leak monitoring model is trained through the following steps: acquiring a multi-condition sample infrared image dataset; the multi-condition sample infrared image dataset includes sample infrared images of sample hydrogen-blended natural gas pipelines under multiple leak conditions, where a leak condition is determined based on the sample gas type, sample leak hole size, and sample pipeline operating pressure; determining the sample gas type label corresponding to each sample infrared image; determining the sample leak risk level label corresponding to each sample infrared image based on the leak condition corresponding to each sample infrared image; and pre-training the initial model based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leak risk level label corresponding to each sample infrared image to obtain the gas leak monitoring model.

[0009] According to the method for monitoring leaks in hydrogen-blended natural gas pipelines provided by the present invention, the sample leakage risk level label corresponding to a sample infrared image is determined by the following steps: determining the sample gas type and sample gas leakage rate under the corresponding leakage conditions; determining the lower explosive limit corresponding to the sample gas type based on Le Chatelier's law; and determining the sample leakage risk level label corresponding to the sample infrared image based on the sample gas leakage rate and the lower explosive limit corresponding to the sample gas type.

[0010] According to the present invention, a method for monitoring leaks in hydrogen-blended natural gas pipelines is provided. Based on each sample infrared image, the corresponding sample gas type label, and the corresponding sample leak risk level label, an initial model is pre-trained to obtain a gas leak monitoring model. The method includes: randomly sampling based on the sample gas type label corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set; pre-training the initial model based on the training set; and testing the initial model based on the test set to obtain the gas leak monitoring model.

[0011] According to the present invention, a method for monitoring leaks in hydrogen-blended natural gas pipelines is provided. Based on each sample infrared image, the corresponding sample gas type label, and the corresponding sample leak risk level label, an initial model is pre-trained to obtain a gas leak monitoring model. The method includes: performing stratified sampling based on the sample gas leak rate corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set; pre-training the initial model based on the training set; and testing the initial model based on the test set to obtain the gas leak monitoring model.

[0012] According to the present invention, a method for monitoring leaks in a hydrogen-blended natural gas pipeline acquires a multi-condition sample infrared image dataset, comprising: acquiring gas plume motion videos of a sample hydrogen-blended natural gas pipeline under multiple leak conditions based on a micro-leakage monitoring system for hydrogen-blended natural gas pipelines; performing frame-by-frame processing on each gas plume motion video to obtain sample infrared images of the sample hydrogen-blended natural gas pipeline under multiple leak conditions, thereby constituting a multi-condition sample infrared image dataset.

[0013] The present invention also provides a hydrogen-blended natural gas pipeline leakage monitoring device, comprising: an acquisition module for acquiring an infrared image to be detected; the infrared image to be detected is obtained by taking a picture of the hydrogen-blended natural gas pipeline to be monitored where a gas leak exists; and a monitoring module for inputting the infrared image to be detected into a gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored output by the gas leak monitoring model; wherein, the gas leak monitoring model is trained based on a sample infrared image, a sample gas type label corresponding to the sample infrared image, and a sample leakage risk level label corresponding to the sample infrared image.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any of the above-described methods for monitoring leaks in hydrogen-blended natural gas pipelines.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described methods for monitoring leaks in hydrogen-blended natural gas pipelines.

[0016] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements any of the above-described methods for monitoring leaks in hydrogen-blended natural gas pipelines.

[0017] This invention provides a method, apparatus, equipment, medium, and product for monitoring leaks in hydrogen-blended natural gas pipelines. The method involves acquiring an infrared image to be detected. This infrared image is obtained by photographing the hydrogen-blended natural gas pipeline where a gas leak is suspected. The image is then input into a gas leak monitoring model to obtain the leak risk level of the pipeline. The gas leak monitoring model is trained based on sample infrared images, corresponding gas type labels, and corresponding leak risk level labels. By introducing artificial intelligence technology into the monitoring of hydrogen-blended natural gas pipeline leaks, the gas leak monitoring model is first trained using sample infrared images, corresponding gas type labels, and corresponding leak risk level labels. This model then monitors for gas leaks based on the infrared image of the hydrogen-blended natural gas pipeline and outputs the leak risk level. This approach achieves reliable leak monitoring of hydrogen-blended natural gas pipelines under different conditions and also enables precise classification of leak risk levels. This allows personnel to conduct appropriate emergency management based on the leak risk level, meeting the needs of leak control and on-site emergency management. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating the hydrogen-blended natural gas pipeline leakage monitoring method provided by the present invention.

[0020] Figure 2 This is a schematic diagram of the structure of the hydrogen-blended natural gas pipeline micro-leakage monitoring system based on an infrared imager provided by the present invention.

[0021] Figure 3 This is one of the schematic diagrams of infrared images of a sample of a hydrogen-blended natural gas pipeline during a leak, provided by the present invention.

[0022] Figure 4 This is the second schematic diagram of a sample infrared image of a hydrogen-blended natural gas pipeline leak, provided by the present invention.

[0023] Figure 5 This is a schematic diagram of the structure of the hydrogen-blended natural gas pipeline leakage monitoring device provided by the present invention.

[0024] Figure 6This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] Please see Figure 1 , Figure 1 This is a schematic flowchart of the hydrogen-blended natural gas pipeline leakage monitoring method provided by the present invention. Figure 1 As shown, in this embodiment, the method for monitoring leaks in hydrogen-blended natural gas pipelines includes steps S110 to S120, each of which is detailed below: S110: Acquire the infrared image to be detected.

[0027] The infrared image to be detected is obtained by taking pictures of the hydrogen-blended natural gas pipeline to be monitored where there is a gas leak.

[0028] S120: Input the infrared image to be detected into the gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored, as output by the gas leak monitoring model.

[0029] The gas leak monitoring model is trained based on sample infrared images, sample gas type labels corresponding to the sample infrared images, and sample leak risk level labels corresponding to the sample infrared images.

[0030] Among various leak detection methods, infrared imaging technology has significant advantages. Compared with traditional methods that use sensors to identify leak signals, infrared imaging technology can identify leaks without direct contact with the gas in hydrogen-blended natural gas pipelines, enabling long-distance image acquisition and leak monitoring. It also boasts high sensitivity and provides a very intuitive way to identify abnormalities in the leaking gas during the monitoring process.

[0031] Based on this, for each hydrogen-blended natural gas pipeline with gas leakage to be monitored, infrared imaging equipment such as infrared imagers and infrared cameras can be used to capture images of the pipeline to be monitored, thus obtaining the infrared image of the pipeline to be monitored.

[0032] Furthermore, the infrared image to be detected is input into the gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored, as output by the gas leak monitoring model.

[0033] The hydrogen-blended natural gas pipeline leakage monitoring method provided in this embodiment acquires an infrared image to be detected. This infrared image is obtained by photographing the hydrogen-blended natural gas pipeline where a gas leak is suspected. The image is then input into a gas leak monitoring model to obtain the leakage risk level of the pipeline. The gas leak monitoring model is trained based on sample infrared images, corresponding gas type labels, and leakage risk level labels. This method introduces artificial intelligence into the hydrogen-blended natural gas pipeline leakage monitoring process. First, a gas leak monitoring model is trained using sample infrared images, corresponding gas type labels, and leakage risk level labels. Then, the model is used to monitor gas leaks based on the infrared image of the hydrogen-blended natural gas pipeline and outputs the leakage risk level. This method achieves reliable leakage monitoring of hydrogen-blended natural gas pipelines under different conditions and also accurately classifies leakage risk levels, enabling personnel to conduct appropriate emergency management based on the leakage risk level. This meets the needs of leakage control and on-site emergency management.

[0034] In some embodiments, the gas leak monitoring model is trained through the following steps: acquiring a multi-condition sample infrared image dataset; the multi-condition sample infrared image dataset includes sample infrared images of hydrogen-blended natural gas pipelines under multiple leak conditions, where a leak condition is determined based on the sample gas type, leak hole size, and pipeline operating pressure; determining the sample gas type label corresponding to each sample infrared image; determining the sample leak risk level label corresponding to each sample infrared image based on the leak condition corresponding to each sample infrared image; and pre-training the initial model based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leak risk level label corresponding to each sample infrared image to obtain the gas leak monitoring model.

[0035] Understandably, before using a gas leak monitoring model to monitor leaks in hydrogen-blended natural gas pipelines, a large dataset is needed for model training. Since most existing datasets are small or consist of leak images from non-hydrogen-blended natural gas pipelines, a comprehensive gas leak dataset covering multiple operating conditions for hydrogen-blended natural gas pipelines needs to be established to facilitate model training.

[0036] Please see Figure 2 , Figure 2 This is a schematic diagram of the structure of the hydrogen-blended natural gas pipeline micro-leakage monitoring system based on an infrared imager provided by the present invention.

[0037] Considering that the number of gas leak images collected under actual operating conditions is small and insufficient to form a multi-condition gas leak dataset, a leak experiment can be conducted using a hydrogen-blended natural gas pipeline micro-leakage monitoring system based on an infrared imager to simulate leak monitoring during the actual transportation of hydrogen-blended natural gas pipelines and obtain gas leak data under multiple operating conditions.

[0038] like Figure 2 As shown, the hydrogen-blended natural gas pipeline micro-leakage monitoring system based on infrared imager includes a gas source component, a gas processing component, a pipeline component, a monitoring component, and auxiliary components.

[0039] The gas source assembly includes multiple gas cylinders; the gas source assembly can serve as a gas source to provide different gases such as methane and hydrogen, and is the starting point for gas supply in the monitoring system for minor leaks in hydrogen-blended natural gas pipelines.

[0040] The gas handling assembly includes a gas mixing device, a first pressure regulating valve, a second pressure regulating valve, and a flow meter.

[0041] In this gas source assembly, multiple gas cylinders are connected to a gas mixing device via a first pressure regulating valve. The gas mixing device is used to mix different gases from multiple gas cylinders in a certain proportion to simulate the actual working condition of natural gas blending with hydrogen.

[0042] The outlets of multiple gas cylinders, a first pressure regulating valve, a gas mixing device, and a second pressure regulating valve are connected in sequence. The first pressure regulating valve is used to adjust the pressure of the gas output from different gas cylinders to stabilize it within a suitable range. The second pressure regulating valve is used to further regulate the pressure of the mixed gas (i.e., hydrogen-blended natural gas) output from the gas mixing device to ensure that the gas pressure entering the subsequent pipeline meets the experimental requirements.

[0043] The flow meter is installed after the second pressure regulating valve to accurately measure the flow rate of hydrogen-blended natural gas, facilitating control of the gas delivery volume.

[0044] The piping assembly includes the main pipe, pipe branch interfaces, leak zone branches, and end vent valves.

[0045] The mixed gas (i.e., hydrogen-blended natural gas) output from the gas mixing device enters the main pipeline for transportation after passing through the second pressure regulating valve and flow meter; the main pipeline is equipped with pressure gauges and thermometers, which are used to monitor the pressure and temperature of the mixed gas in the pipeline in real time; the main pipeline is also equipped with open and close valves, which can control the flow of the mixed gas in the pipeline.

[0046] The branch line interface is located on the main pipeline and is used to connect to the branch line in the leak area. Through this branch line interface, the mixed gas in the main pipeline can be introduced into the branch line in the leak area to conduct simulated leak experiments.

[0047] The leak zone branch line includes a pipe interface and a flow meter, and an experimental pipe is connected to the leak zone branch line. The pipe interface is used to connect the main pipe and the leak zone branch line. The flow meter is installed after the pipe interface and is used to measure the gas flow rate entering the leak zone branch line. During the experiment, different leakage conditions (i.e., different degrees of micro-leakage) can be simulated by replacing the experimental pipe with different sized leak holes.

[0048] The tail-end vent valve is located at the tail end of the main pipeline and is used to safely release the mixed gas when the experiment is completed or when it is necessary to empty the mixed gas in the pipeline.

[0049] The monitoring components include an infrared imager, which is positioned near a replaceable experimental pipe to capture changes in infrared radiation caused by gas leakage at the leak point in the experimental pipe, thereby monitoring minute leaks in the pipe and generating sample infrared images.

[0050] The auxiliary components include a blackbody standard source, which is used to calibrate the infrared imager and ensure the accuracy of its measurements.

[0051] Optionally, a small oil and gas gathering and transportation station can be built according to the actual station process pipeline transportation mode: First, a sample hydrogen-blended natural gas pipeline gathering and transportation system is established according to the equipment and facilities required for the actual operation of oil and gas pipelines. Then, a pipeline leakage area is set as a leakage area branch on the sample hydrogen-blended natural gas pipeline gathering and transportation system. Further, appropriate infrared imagers and specific models of gas leakage rate measurement equipment are selected to collect gas leakage videos and leakage rate data under different operating conditions.

[0052] Optionally, the sample hydrogen-blended natural gas pipeline gathering and transportation system includes an upstream gas source (corresponding to the gas source component), a gas mixing device (corresponding to the gas mixing device in the gas processing component), a gas compressor (corresponding to the second pressure regulating valve), a sample hydrogen-blended natural gas pipeline (corresponding to the replaceable experimental pipeline), a tail-end vent valve, pressure gauges, flow meters, and other instruments.

[0053] Optionally, the upstream gas source consists of multiple high-pressure gas cylinders, providing different gases such as methane and hydrogen. These gases are then transported to a gas mixing device for thorough mixing. The mixing device outputs mixed gases of different types (i.e., different hydrogen blending ratios). The mixed gas is pressurized by a small gas compressor and transported within the pipeline as hydrogen-blended natural gas with higher pressure and different hydrogen blending ratios. After pressure regulation by the pipeline valve assembly, the hydrogen-blended natural gas continues to be transported to the pipeline leak area. When the hydrogen-blended natural gas flows stably within the pipeline, the relevant devices in the pipeline leak area are activated, causing the hydrogen-blended natural gas in the sample hydrogen-blended natural gas pipeline to leak under specified hydrogen blending ratio, pressure, and leak hole size conditions. An infrared imager records the gas plume movement during the leak. Simultaneously, a flow meter is connected between the main pipeline and the sample hydrogen-blended natural gas pipeline to record the gas leakage rate, and a concentration monitoring sensor is installed directly above the leak hole to record the methane and hydrogen concentrations at different heights. After the experiment, the state of the tail-end vent valve can be controlled to achieve continuous leakage under specific pipeline operating conditions.

[0054] It should be noted that the principle of leakage experiments is to ensure the safety of the experimental process. Therefore, this can be achieved by designing appropriate anti-aircraft devices at the pipeline end to vent combustible gases, installing static electricity removal devices to eliminate static electricity that may be generated by the human body, and setting up a certain number of combustible gas alarm devices in the experimental area to control the gas emission in a timely manner. Secondly, it is essential to ensure that the experimental objective is clear. The objective of this experiment is to collect infrared images of leaks and their corresponding gas flow rates of mixed gases with different hydrogen doping ratios under different leak hole sizes and different pipeline operating pressures. The key lies in how to achieve micro-leakage and how to complete the complete and accurate acquisition of the above data. Thirdly, the experimental system should be complete. The gas supply, pipeline system construction, leakage process, and final data acquisition should all be rigorous and complete. Finally, the experimental design should be as simple as possible, and the process should be simplified as much as possible to reduce the possibility of errors and ensure the reliability of data acquisition.

[0055] Optionally, a preliminary experiment should be conducted before the formal leakage experiment.

[0056] Optionally, since it was found during the preliminary experiment that the gas leakage rate of a 2mm leak hole was far beyond the maximum range of the leakage rate measuring device, it can be determined that it does not belong to the micro-leakage under the experimental framework, and therefore, this size is excluded.

[0057] Optionally, since the hydrogen blending ratio is a key issue in the application of hydrogen-blended natural gas, and the hydrogen blending ratio of most hydrogen-blended natural gas is generally in the range of 2%-20%, this embodiment selects to set the hydrogen blending ratio to four groups of 0%, 10%, 20% and 30% respectively.

[0058] Optionally, based on the pipeline operating pressure of a typical station, the pressure required for minor leaks, and the pressure level that the test bench can withstand, this embodiment sets the maximum pipeline pressure to 2 MPa and the minimum pressure to 0.1 MPa to form 288 sets of leakage conditions as shown in Table 1.

[0059] Table 1

[0060] It should be noted that the purpose of this embodiment is to achieve real-time monitoring of hydrogen-blended natural gas leaks through a model based on clarifying the leakage patterns. The model built on modern machine learning algorithms is based on high-quality and large-scale data. Therefore, it is necessary to create a large dataset of infrared images of hydrogen-blended natural gas leaks under multiple operating conditions. The leakage conditions need to be as comprehensive as possible, reaching 288 sets.

[0061] Optionally, the leakage test procedure under multiple operating conditions is as follows: (1) Open the valves of different gas cylinders so that the gas from different gas cylinders enters the gas mixing device through the pressure regulating valve. After the different gases are mixed evenly in the gas mixing device, they are transported through the pipeline.

[0062] (2) Observe the pressure gauge and temperature gauge, and adjust the pressure regulating valve and other devices to stabilize the gas pressure and temperature in the pipeline within the range of parameters required for the experiment. After the sample hydrogen-doped natural gas pipeline gathering and transportation system has been running stably for a period of time, ensure that the gas state is stable.

[0063] (3) Close the opening and closing valve, install the experimental pipeline with a leak hole of a specific size at the branch interface of the leak area, and then open the opening and closing valve to simulate gas leakage.

[0064] (4) Immediately use an infrared thermal imager to image and photograph the leak area, record the video of the gas plume movement of the leaked gas, and analyze the temperature distribution and leak range of the leaked gas; at the same time, observe the flow meter and record the flow rate data of the leaked gas.

[0065] (5) Record data such as flow rate, pressure, and temperature at preset time intervals for a period of time so as to analyze the change of leakage rate over time.

[0066] (6) Close the valve, replace the experimental pipe with a leak hole of different size, repeat the above steps, test the gas leakage infrared imaging and leakage rate under different leak hole diameters, and obtain multiple sets of data under different leakage conditions.

[0067] Specifically, based on the micro-leakage monitoring system for hydrogen-blended natural gas pipelines, after acquiring gas plume motion videos of sample hydrogen-blended natural gas pipelines under multiple leakage conditions, the gas plume motion videos under each leakage condition can be processed into frames to obtain sample infrared images of sample hydrogen-blended natural gas pipelines under multiple leakage conditions, thus forming a multi-condition sample infrared image dataset.

[0068] The multi-condition sample infrared image dataset includes sample infrared images of hydrogen-blended natural gas pipelines under multiple leakage conditions. A leakage condition is determined based on the sample gas type (i.e., sample hydrogen blending ratio), sample leak hole size, and sample pipeline operating pressure.

[0069] Different types of sample gases, sample leak hole sizes, and sample pipeline operating pressures can create different leakage conditions, as shown in Table 1.

[0070] Optionally, for each leakage condition, the number of sample infrared images collected under that leakage condition shall be at least 172,800.

[0071] Please see Figure 3 and Figure 4 , Figure 3 This is one of the schematic diagrams of infrared images of samples from a hydrogen-blended natural gas pipeline leak, provided by the present invention. Figure 4 This is the second schematic diagram of a sample infrared image of a hydrogen-blended natural gas pipeline leak, provided by the present invention.

[0072] like Figure 3 and Figure 4 As shown, the infrared images of the sample can intuitively demonstrate the leakage situation of the hydrogen-blended natural gas pipeline.

[0073] Furthermore, the sample gas type label and the sample gas leakage rate corresponding to each sample infrared image are determined respectively.

[0074] Optionally, for each sample infrared image, the gas leakage rate of the gas plume motion video under the corresponding leakage condition collected during the leakage experiment can be used as the sample gas leakage rate of that sample infrared image.

[0075] Similarly, for each sample infrared image, the hydrogen doping ratio of the sample in the gas plume motion video under the corresponding leakage condition set during the leakage experiment can be used as the sample gas type label for that sample infrared image.

[0076] Furthermore, based on the leakage conditions corresponding to each sample infrared image, a sample leakage risk level label is determined for each sample infrared image.

[0077] Furthermore, based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image, the initial model is pre-trained to obtain a gas leakage monitoring model.

[0078] In some embodiments, the sample leakage risk level label corresponding to a sample infrared image is determined by the following steps: determining the sample gas type and sample gas leakage rate under the corresponding leakage conditions; determining the lower explosive limit corresponding to the sample gas type based on Le Chatelier's law; and determining the sample leakage risk level label corresponding to the sample infrared image based on the sample gas leakage rate and the lower explosive limit corresponding to the sample gas type.

[0079] Understandably, simply identifying the presence of a gas leak is insufficient for leak control and emergency management at the accident site. Further classification of the severity of gas leak risks is needed to achieve a comprehensive system for tiered leak detection, quantification, and selection of tiered remediation solutions. Therefore, it is also necessary to introduce sample leak risk level labels into the training samples of the gas leak monitoring model.

[0080] Specifically, after completing the leakage monitoring experiment of hydrogen-blended natural gas pipelines, a multi-condition sample infrared image dataset and the gas leakage rate corresponding to each sample infrared image in the multi-condition sample infrared image dataset can be obtained. In order to achieve accurate classification of the risk level of hydrogen-blended natural gas leakage, a classification standard for the risk level of hydrogen-blended natural gas leakage can be further established, and a complete multi-condition sample infrared image dataset of hydrogen-blended natural gas can be constructed.

[0081] This embodiment proposes a strategy for classifying leakage risk levels, using the time interval between leaked gas mixtures reaching their lower explosive limit within a certain space at a specific leakage rate as the basis for classification. During the operation of oil and gas gathering and transportation pipeline systems, various types of enclosed spaces exist, such as valve chambers and pump rooms. These enclosed spaces vary in size, and multiple pipelines converge within oil and gas gathering and transportation stations, creating equivalent enclosed spaces. Based on field surveys and calculations, the volume of a small valve chamber may be around 10-30 cubic meters, while the enclosed space within a compressor room can reach 10-50 cubic meters. Adhering to the principle that gas leakage risk level classification should be more stringent than actual conditions, this embodiment selects 10 cubic meters as the standard size for enclosed spaces for subsequent calculations. Furthermore, based on the gas leakage rate corresponding to the gas plume movement videos collected from the experiment under each leakage condition and the emergency response time of the oil and gas pipeline system, this embodiment selects 10 minutes and 50 minutes as two key time indicators for leakage risk level classification.

[0082] Specifically, for each sample infrared image, after determining the sample gas type and sample gas leakage rate under the corresponding leakage conditions, the lower explosive limit corresponding to the sample gas type can be determined based on Le Chatelier's Law.

[0083] The expression for Le Chatelier's rule is as follows: ; in, Indicates the lower (or upper) explosive limit of a gas mixture; Indicates the first gas in the mixture The lower (or upper) explosive limit of each component; Indicates the first gas in the mixture Mole fraction or volume fraction of each component; It represents the sum of all the types of components in a gas mixture.

[0084] Considering that hydrogen-blended natural gas mainly includes methane ( ) and hydrogen ( Therefore, in this embodiment, Le Chatelier's law can be further rewritten as: (Two components) ; in, Indicates the mole fraction or volume fraction of methane; This indicates the mole fraction or volume fraction of hydrogen. This indicates the lower explosive limit of methane. This indicates the lower explosive limit of hydrogen.

[0085] For example, according to the revised Le Chatelier's rule, the lower explosive limits of hydrogen-blended natural gas with hydrogen blending ratios of 10%, 20%, and 30% can be calculated to be approximately 4.88% (i.e., ), 4.76% (i.e. ), 4.65% (i.e. ).

[0086] Furthermore, based on the sample gas leakage rate and the lower explosion limit corresponding to the sample gas type, the sample leakage risk level label corresponding to the sample infrared image is determined.

[0087] Specifically, for each sample infrared image, after determining the lower explosive limit of the sample gas type corresponding to the sample infrared image, the sample leakage risk level label corresponding to the sample infrared image can be determined by combining the time required for the sample infrared image to reach the lower explosive limit at the corresponding sample gas leakage rate; wherein, the classification of sample leakage risk levels is shown in Table 2.

[0088] Table 2

[0089] As shown in Table 2, when the sample gas leakage rate is greater than or equal to 50 L / min and the time to reach the lower explosive limit is less than or equal to 10 min, the sample leakage risk level label is high risk; when the sample gas leakage rate is greater than or equal to 10 L / min and less than 50 L / min, and the time to reach the lower explosive limit is greater than 10 min and less than or equal to 50 min, the sample leakage risk level label is medium risk; when the sample gas leakage rate is less than 10 L / min and the time to reach the lower explosive limit is greater than 50 min, the sample leakage risk level label is low risk.

[0090] After adding corresponding sample gas type labels and sample leakage risk level labels to all sample infrared images, all sample infrared images can be used as training samples to train a gas leakage monitoring model.

[0091] In the application of the gas leak monitoring model, when the model identifies a leak rate level of Level 1 (i.e., high-speed leakage) in the monitored hydrogen-blended natural gas pipeline, an early warning message will be immediately issued. The warning message states: "In this leak situation, immediate action should be taken. Emergency response should be carried out according to the emergency response plan corresponding to Level 1 leaks. A comprehensive repair plan should be determined based on the site conditions and leak status of the gathering and transmission station. This may include pipeline clamp sealing, plugging, pipeline shutdown and replacement, etc. Detailed records should be kept, and continuous key monitoring should be conducted for similar situations and areas." When the leakage rate level of a hydrogen-blended natural gas pipeline is Level II (i.e., medium-speed leakage), the warning message that can be issued is "Repair work should be carried out in an orderly manner according to the maintenance plan. The repair time limit should be determined based on the overall situation on site. Generally, it is not necessary to stop the transmission. The repair should be completed within the repair time limit, and a re-inspection should be carried out in a timely manner." When the gas leakage monitoring model identifies that the leakage rate level of the hydrogen-blended natural gas pipeline to be monitored is Level III (i.e., low-speed leakage), the warning message that can be issued is "The pipeline should be continuously monitored, the leak should be accurately located, the cause of the leak should be identified, and repairs should be carried out within the specified time."

[0092] In some embodiments, the initial model is pre-trained based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image to obtain a gas leakage monitoring model. This includes: randomly sampling based on the sample gas type label corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set respectively; pre-training the initial model based on the training set, and testing the initial model based on the test set to obtain a gas leakage monitoring model.

[0093] Understandably, after adding corresponding sample gas type labels and sample leakage risk level labels to all sample infrared images, all sample infrared images can be used as training samples, and the sample gas type labels can be used as the classification criteria for the training samples.

[0094] Since the sample infrared images classified by the sample gas type label are uniformly distributed, random sampling can be performed based on the sample gas type label corresponding to each sample infrared image, and all sample infrared images can be divided into training set and test set respectively. For example, 70% of the sample infrared images of each hydrogen doping ratio (sample gas type label) can be used as training set and 30% as test set.

[0095] Furthermore, the initial model is pre-trained based on the training set and tested based on the test set to obtain a gas leak monitoring model.

[0096] In some embodiments, a gas leak monitoring model is obtained by pre-training an initial model based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image. This includes: performing stratified sampling based on the sample gas leakage rate corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set respectively; pre-training the initial model based on the training set, and testing the initial model based on the test set to obtain the gas leak monitoring model.

[0097] Understandably, after adding corresponding sample gas type labels and sample leakage risk level labels to all sample infrared images, all sample infrared images can be used as training samples, and the sample gas leakage rate can be used as the classification standard for the training samples.

[0098] Because the distribution of sample infrared images classified by the sample gas leakage rate is uneven (there are more sample infrared images of low-speed and medium-speed leakage, while there are relatively fewer sample infrared images of high-speed leakage), stratified sampling can be performed based on the sample gas leakage rate corresponding to each sample infrared image. All sample infrared images can be divided into training and test sets. For example, 70% of the sample infrared images of each leakage rate level can be uniformly selected as the training set and 30% as the test set to ensure that the ratio of training and test sets is consistent for each leakage rate level, thus completing the division of the overall dataset.

[0099] Furthermore, the initial model is pre-trained based on the training set and tested based on the test set to obtain a gas leak monitoring model.

[0100] In some embodiments, acquiring a multi-condition sample infrared image dataset includes: acquiring gas plume motion videos of sample hydrogen-blended natural gas pipelines under multiple leakage conditions based on a micro-leakage monitoring system for hydrogen-blended natural gas pipelines; performing frame-by-frame processing on each gas plume motion video to obtain sample infrared images of sample hydrogen-blended natural gas pipelines under multiple leakage conditions, thereby constituting a multi-condition sample infrared image dataset.

[0101] The hydrogen-blended natural gas pipeline leakage monitoring method provided in this embodiment has completed the acquisition of infrared images and other numerical data of small leaks in hydrogen-blended natural gas under multiple operating conditions, and constructed a large-scale multi-condition sample infrared image dataset of pipeline leaks in hydrogen-blended natural gas stations, which is beneficial for the training and application of subsequent gas leak monitoring models. The experimental data were obtained through hydrogen-blended natural gas leakage monitoring based on passive infrared imaging, which helps in the development of subsequent deep learning models to quantify the degree of gas leakage, identify the type of leaked gas, and achieve accurate identification and monitoring of small leaks. In addition, non-contact infrared monitoring has obvious advantages: the infrared imager adopts a non-contact monitoring method, which does not require direct contact with the pipeline, avoiding pipeline damage and safety risks that may be caused by contact. At the same time, it can monitor without interrupting pipeline operation, ensuring the continuity of natural gas hydrogen-blended pipeline transportation, reducing monitoring costs and interference with production operations.

[0102] This invention also provides a leak monitoring device for hydrogen-blended natural gas pipelines. Please refer to [link / reference]. Figure 5 , Figure 5 This is a schematic diagram of the structure of the hydrogen-blended natural gas pipeline leakage monitoring device provided by the present invention. In this embodiment, the hydrogen-blended natural gas pipeline leakage monitoring device includes an acquisition module 510 and a monitoring module 520.

[0103] The acquisition module 510 is used to acquire the infrared image to be detected.

[0104] The infrared image to be detected is obtained by taking pictures of the hydrogen-blended natural gas pipeline to be monitored where there is a gas leak.

[0105] The monitoring module 520 is used to input the infrared image to be detected into the gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored, as output by the gas leak monitoring model.

[0106] The gas leak monitoring model is trained based on sample infrared images, sample gas type labels corresponding to the sample infrared images, and sample leak risk level labels corresponding to the sample infrared images.

[0107] In some embodiments, the gas leak monitoring model is trained through the following steps: acquiring a multi-condition sample infrared image dataset; the multi-condition sample infrared image dataset includes sample infrared images of hydrogen-blended natural gas pipelines under multiple leak conditions, where a leak condition is determined based on the sample gas type, leak hole size, and pipeline operating pressure; determining the sample gas type label corresponding to each sample infrared image; determining the sample leak risk level label corresponding to each sample infrared image based on the leak condition corresponding to each sample infrared image; and pre-training the initial model based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leak risk level label corresponding to each sample infrared image to obtain the gas leak monitoring model.

[0108] In some embodiments, the sample leakage risk level label corresponding to a sample infrared image is determined by the following steps: determining the sample gas type and sample gas leakage rate under the corresponding leakage conditions; determining the lower explosive limit corresponding to the sample gas type based on Le Chatelier's law; and determining the sample leakage risk level label corresponding to the sample infrared image based on the sample gas leakage rate and the lower explosive limit corresponding to the sample gas type.

[0109] In some embodiments, the initial model is pre-trained based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image to obtain a gas leakage monitoring model. This includes: randomly sampling based on the sample gas type label corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set respectively; pre-training the initial model based on the training set, and testing the initial model based on the test set to obtain a gas leakage monitoring model.

[0110] In some embodiments, a gas leak monitoring model is obtained by pre-training an initial model based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image. This includes: performing stratified sampling based on the sample gas leakage rate corresponding to each sample infrared image, dividing all sample infrared images into a training set and a test set respectively; pre-training the initial model based on the training set, and testing the initial model based on the test set to obtain the gas leak monitoring model.

[0111] In some embodiments, acquiring a multi-condition sample infrared image dataset includes: acquiring gas plume motion videos of sample hydrogen-blended natural gas pipelines under multiple leakage conditions based on a micro-leakage monitoring system for hydrogen-blended natural gas pipelines; performing frame-by-frame processing on each gas plume motion video to obtain sample infrared images of sample hydrogen-blended natural gas pipelines under multiple leakage conditions, thereby constituting a multi-condition sample infrared image dataset.

[0112] The present invention also provides an electronic device. Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention, such as... Figure 6 As shown, the electronic device may include a processor 610, a communication interface 620, a memory 630, and a communication bus 640, wherein the processor 610, the communication interface 620, and the memory 630 communicate with each other via the communication bus 640. The processor 610 can call logical instructions in the memory 630 to execute a hydrogen-blended natural gas pipeline leak monitoring method.

[0113] Furthermore, the logical instructions in the aforementioned memory 630 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0114] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the hydrogen-blended natural gas pipeline leakage monitoring method provided by the above methods.

[0115] The present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the hydrogen-blended natural gas pipeline leakage monitoring method provided by the above methods.

[0116] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0117] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for monitoring leaks in hydrogen-blended natural gas pipelines, characterized in that, include: Acquire an infrared image to be detected; the infrared image to be detected is obtained by taking a picture of the hydrogen-blended natural gas pipeline to be monitored where there is a gas leak; The infrared image to be detected is input into the gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored, as output by the gas leak monitoring model. The gas leak monitoring model is trained based on sample infrared images, sample gas type labels corresponding to the sample infrared images, and sample leak risk level labels corresponding to the sample infrared images.

2. The method for monitoring leaks in hydrogen-blended natural gas pipelines according to claim 1, characterized in that, The gas leak monitoring model was trained through the following steps: Acquire a multi-condition sample infrared image dataset; the multi-condition sample infrared image dataset includes sample infrared images of hydrogen-blended natural gas pipelines under multiple leakage conditions, wherein one leakage condition is determined based on the sample gas type, sample leak hole size, and sample pipeline operating pressure of the sample hydrogen-blended natural gas pipeline; Determine the sample gas type label corresponding to each sample infrared image; Based on the leakage condition corresponding to each sample infrared image, a sample leakage risk level label is determined for each sample infrared image; Based on each sample infrared image, the sample gas type label corresponding to each sample infrared image, and the sample leakage risk level label corresponding to each sample infrared image, the initial model is pre-trained to obtain the gas leakage monitoring model.

3. The method for monitoring leaks in hydrogen-blended natural gas pipelines according to claim 2, characterized in that, The sample leakage risk level label corresponding to an infrared image of the sample is determined by the following steps: Determine the type and leakage rate of the sample gas in the infrared image of the sample under the corresponding leakage condition; Based on Le Chatelier's rule, the lower explosive limit corresponding to the type of sample gas is determined; Based on the sample gas leakage rate and the lower explosion limit corresponding to the sample gas type, the sample leakage risk level label corresponding to the sample infrared image is determined.

4. The method for monitoring leaks in hydrogen-blended natural gas pipelines according to claim 2, characterized in that, The process of pre-training an initial model based on each sample infrared image, the corresponding sample gas type label for each sample infrared image, and the corresponding sample leakage risk level label for each sample infrared image to obtain the gas leakage monitoring model includes: Based on the sample gas type label corresponding to each sample infrared image, random sampling is performed to divide all sample infrared images into training set and test set respectively; The initial model is pre-trained based on the training set and tested based on the test set to obtain the gas leak monitoring model.

5. The method for monitoring leaks in hydrogen-blended natural gas pipelines according to claim 2, characterized in that, The process of pre-training an initial model based on each sample infrared image, the corresponding sample gas type label for each sample infrared image, and the corresponding sample leakage risk level label for each sample infrared image to obtain the gas leakage monitoring model includes: Based on the sample gas leakage rate corresponding to each sample infrared image, stratified sampling is performed to divide all sample infrared images into training set and test set respectively. The initial model is pre-trained based on the training set and tested based on the test set to obtain the gas leak monitoring model.

6. The method for monitoring leaks in hydrogen-blended natural gas pipelines according to claim 2, characterized in that, The acquisition of the multi-condition sample infrared image dataset includes: Based on the micro-leakage monitoring system for hydrogen-blended natural gas pipelines, video footage of gas plume movement in the sample hydrogen-blended natural gas pipelines under multiple leakage conditions was obtained. Each of the gas plume motion videos is processed by frame segmentation to obtain sample infrared images of the hydrogen-doped natural gas pipeline under multiple leakage conditions, thus forming the multi-condition sample infrared image dataset.

7. A hydrogen-blended natural gas pipeline leak monitoring device, characterized in that, include: The acquisition module is used to acquire an infrared image to be detected; the infrared image to be detected is obtained by taking a picture of the hydrogen-blended natural gas pipeline to be monitored where there is a gas leak; The monitoring module is used to input the infrared image to be detected into the gas leak monitoring model to obtain the leakage risk level of the hydrogen-blended natural gas pipeline to be monitored, as output by the gas leak monitoring model. The gas leak monitoring model is trained based on sample infrared images, sample gas type labels corresponding to the sample infrared images, and sample leak risk level labels corresponding to the sample infrared images.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the hydrogen-blended natural gas pipeline leakage monitoring method as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for monitoring leaks in hydrogen-blended natural gas pipelines as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for monitoring leaks in hydrogen-blended natural gas pipelines as described in any one of claims 1 to 6.

Images