An intelligent monitoring system and method for liquefied petroleum gas supply

Abstract

This application provides an intelligent monitoring system and method for liquefied petroleum gas (LPG) supply. The system determines the correlation coefficients of pressure decay between different sensor nodes based on the dynamic pressure characteristics of LPG at each sensor node. Furthermore, the system determines the thermodynamic distribution diagram of the LPG pressure decay from all the correlation coefficients and the topological structure of the sensor nodes. The system also evaluates the pressure leakage status of the transport segments between adjacent sensor nodes based on the mutual support of the pressure fluctuation amplitudes between all adjacent sensor nodes, deriving the confidence probability of pressure leakage in each transport segment and determining a correlation detection tree for LPG supply network leaks. Furthermore, the system performs fused leak detection on the LPG supply network using the thermodynamic distribution diagram of LPG pressure decay and the correlation detection tree to determine the leakage occurrence area of ​​the LPG supply network. This solution enables confident leak detection of the LPG supply network based on information from multiple monitoring points.

Description

Technical Field

[0001] This application relates to the field of pressure measurement technology, and more specifically, to an intelligent monitoring system and method for liquefied petroleum gas supply. Background Technology

[0002] In the safety inspection of liquefied petroleum gas (LPG) supply, pressure measurement is used to monitor the gas pressure in pipelines, storage tanks, and valves in real time, ensuring that it is maintained within the set safety range and promptly detecting any abnormal fluctuations. By analyzing continuous pressure data, the sealing performance and operational stability of the system can be effectively assessed, helping to identify pressure anomalies caused by minor cracks, loose joints, or equipment aging, thereby preventing potential leakage accidents. In addition, when a sudden drop or abnormality occurs in the pressure of the LPG supply pipeline, pressure measurement can quickly locate the leak point, improve the response speed of leak detection, and further ensure the safe operation of the entire supply system.

[0003] In existing technologies, traditional liquefied petroleum gas (LPG) leak monitoring typically relies on a single sensor or monitoring data from a localized area. This limitation results in a narrow detection range, preventing comprehensive analysis of the entire supply network. In practical applications, monitoring systems often depend on pressure sensors at specific locations or deploy monitoring equipment only on a single pipeline. This means that leaks outside the monitoring area may be overlooked. Furthermore, localized detection cannot capture the full picture of pressure decay across the supply network and lacks a comprehensive understanding of pressure fluctuations between multiple nodes. Consequently, it has a weak ability to identify complex leak patterns, micro-cracks, or pressure changes caused by aging equipment, leading to low reliability of LPG supply network leak monitoring results. Therefore, how to achieve reliable leak monitoring of LPG supply networks based on multi-source monitoring information has become a challenge for the industry. Summary of the Invention

[0004] This application provides an intelligent monitoring system and method for liquefied petroleum gas (LPG) supply, which can realize confident leakage monitoring of the LPG supply network based on information from multiple monitoring points.

[0005] In a first aspect, this application provides an intelligent monitoring method for liquefied petroleum gas supply, comprising the following steps:

[0006] Multiple sensor nodes are deployed in the liquefied petroleum gas supply network, and the pressure status of liquefied petroleum gas at each sensor node is monitored by intelligent pressure sensors.

[0007] Extract the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state, determine the cross-correlation coefficient of pressure attenuation between different sensor nodes based on all dynamic pressure characteristics, and then determine the thermodynamic distribution map of liquefied petroleum gas pressure attenuation based on all cross-correlation coefficients and the topology of sensor nodes.

[0008] Based on the mutual support of pressure fluctuation amplitudes between all adjacent sensor nodes, the correlation confidence evaluation of the pressure leakage status of the transportation segment between each pair of adjacent sensor nodes is carried out to obtain the confidence probability of pressure leakage in each transportation segment. Then, the correlation detection tree of liquefied petroleum gas supply network leakage is determined by the confidence probability of pressure leakage in all transportation segments.

[0009] By combining the thermal distribution map of liquefied petroleum gas pressure decay and the associated detection tree, the leakage detection of the liquefied petroleum gas supply network is performed to obtain the leakage area of ​​the liquefied petroleum gas supply network.

[0010] Preferably, extracting the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state specifically includes:

[0011] For each sensor node, the dynamic pressure data of liquefied petroleum gas at the sensor node is obtained from the pressure state.

[0012] Based on the dynamic pressure data, the dynamic pressure characteristics of liquefied petroleum gas at the sensor nodes are determined, and then the dynamic pressure characteristics of liquefied petroleum gas at each sensor node are obtained.

[0013] Preferably, determining the cross-correlation coefficient of pressure attenuation between different sensor nodes based on all dynamic pressure characteristics specifically includes:

[0014] Select a sensor node as the target sensor node;

[0015] Determine the comparison matrix of dynamic pressure characteristics between the target sensor node and its adjacent sensor nodes based on all dynamic pressure characteristics;

[0016] The cross-correlation coefficient of pressure attenuation between the target sensor node and its adjacent sensor nodes is determined by the comparison matrix.

[0017] Continue to determine the cross-correlation coefficients of pressure attenuation between the remaining sensor nodes and adjacent sensor nodes.

[0018] Preferably, the thermodynamic distribution map of liquefied petroleum gas pressure decay, determined by all cross-correlation coefficients and the topology of sensor nodes, specifically includes:

[0019] The topology of the sensor nodes is determined based on the spatial distance between them.

[0020] For each sensor node, the topological relationship between the sensor node and its adjacent sensor nodes is determined based on the topology diagram.

[0021] The pressure attenuation factor of liquefied petroleum gas at the sensor node is determined by the cross-correlation coefficient of pressure attenuation between the sensor node and its adjacent sensor nodes and the topological relationship, thereby obtaining the pressure attenuation factor of liquefied petroleum gas at each sensor node.

[0022] The thermodynamic distribution map of LPG pressure decay was determined based on all pressure decay factors.

[0023] Preferably, the confidence evaluation of the pressure leakage status of each two adjacent sensor nodes in the transport segment is performed based on the mutual support of pressure fluctuation amplitudes among all adjacent sensor nodes, and the confidence probability of pressure leakage in each transport segment is obtained specifically including:

[0024] Determine the degree of mutual support between pressure fluctuation amplitudes of adjacent sensor nodes;

[0025] Based on all mutual support, construct a correlation matrix of pressure fluctuation behavior in the transportation segment between adjacent sensor nodes, and map the values ​​in the correlation matrix to fuzzy sets;

[0026] Initialize the membership degree of the pressure fluctuation amplitude for each transport segment;

[0027] The fuzzy evaluation matrix is ​​determined by all membership degrees and the fuzzy set;

[0028] Based on the fuzzy evaluation matrix, a fuzzy confidence evaluation is performed on the pressure leakage status of the transportation segment between every two adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transportation segment.

[0029] Preferably, the association detection tree for determining liquefied petroleum gas supply network leaks based on the confidence probability of pressure leaks in all transportation segments specifically includes:

[0030] Each transportation segment is treated as a tree node;

[0031] The connection relationship between adjacent tree nodes is determined by the confidence probability of pressure leakage between adjacent transport sections;

[0032] The associated detection tree for liquefied petroleum gas supply network leaks is determined based on all tree nodes and all connections.

[0033] Preferably, by using the thermal distribution map of liquefied petroleum gas pressure decay and the correlation detection tree to perform fusion leak detection on the liquefied petroleum gas supply network, the specific areas where leaks occur in the liquefied petroleum gas supply network include:

[0034] Determine the fusion detection weights of the heat map and the associated detection tree;

[0035] Based on the fusion detection weights of the thermal distribution map and the associated detection tree, the abnormal information of the thermal distribution map and the associated detection tree is integrated to obtain the integrated information for leak detection.

[0036] The integrated information from the leak detection was used to determine the area where the leak occurred in the liquefied petroleum gas supply network.

[0037] Secondly, this application provides an intelligent monitoring system for liquefied petroleum gas supply, comprising:

[0038] The monitoring module is used to deploy multiple sensor nodes in the liquefied petroleum gas supply network and monitor the pressure status of liquefied petroleum gas at each sensor node through intelligent pressure sensors.

[0039] The processing module is used to extract the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state, determine the cross-correlation coefficient of pressure decay between different sensor nodes based on all the dynamic pressure characteristics, and then determine the thermodynamic distribution map of liquefied petroleum gas pressure decay based on all the cross-correlation coefficients and the topology of the sensor nodes.

[0040] The processing module is also used to perform a correlation confidence evaluation on the pressure leakage status of the transportation section between each pair of adjacent sensor nodes based on the mutual support of the pressure fluctuation amplitude between all adjacent sensor nodes, to obtain the confidence probability of pressure leakage in each transportation section, and then determine the correlation detection tree of liquefied petroleum gas supply network leakage through the confidence probability of pressure leakage in all transportation sections.

[0041] The execution module is used to perform fusion leak detection on the liquefied petroleum gas supply network by using the thermal distribution map of the liquefied petroleum gas pressure decay and the associated detection tree, so as to obtain the leakage area of ​​the liquefied petroleum gas supply network.

[0042] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described intelligent monitoring method for liquefied petroleum gas supply.

[0043] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent monitoring method for liquefied petroleum gas supply.

[0044] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0045] In this embodiment, multiple sensor nodes are deployed in the liquefied petroleum gas (LPG) supply network. Intelligent pressure sensors monitor the pressure state of LPG at each sensor node. Dynamic pressure characteristics of the LPG at each sensor node are extracted from the pressure state. Based on all dynamic pressure characteristics, cross-correlation coefficients of pressure attenuation between different sensor nodes are determined. Then, a thermal distribution map of LPG pressure attenuation is determined using all cross-correlation coefficients and the topology of the sensor nodes. Based on the mutual support of pressure fluctuation amplitudes between all adjacent sensor nodes, a correlation confidence evaluation is performed on the pressure leakage state of the transportation segment between each pair of adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transportation segment. Then, a correlation detection tree for leakage in the LPG supply network is determined using the confidence probabilities of pressure leakage in all transportation segments. Finally, a fusion leakage detection is performed on the LPG supply network using the thermal distribution map of LPG pressure attenuation and the correlation detection tree to obtain the leakage occurrence area of ​​the LPG supply network.

[0046] Therefore, this application utilizes a thermal distribution map of liquefied petroleum gas (LPG) pressure decay and a correlation detection tree to perform fusion leak detection on the LPG supply network, thereby identifying the leak areas. First, by determining the thermal distribution map of LPG pressure decay based on the dynamic pressure characteristics of LPG at each sensor node and the topology of the sensor nodes, it helps to comprehensively reflect the pressure change trend of LPG in the supply network, thus revealing potential leak areas and pressure anomaly areas, and providing the pressure decay pattern of the entire network. This provides reliable data support for subsequent accurate detection and decision-making, significantly improving the monitoring accuracy and safety of the LPG supply network. Then, based on the mutual support of pressure fluctuation amplitudes between adjacent sensor nodes, a correlation confidence evaluation is performed on the pressure leakage status of the transportation section between adjacent sensor nodes, obtaining the pressure of the transportation section. The confidence probability of a leak is determined, and a correlation detection tree for leaks in the LPG supply network is generated. Through a comprehensive evaluation of pressure fluctuation amplitude and mutual support, the leak status of each transportation segment is quantified. A probabilistic model enhances the accuracy of leak detection, improving the ability to identify complex leak patterns and pressure anomalies caused by minute cracks, thus significantly improving detection reliability. Finally, by integrating the thermal distribution map of LPG pressure decay and the correlation detection tree, leak detection of the LPG supply network is performed, identifying the leak occurrence areas. A comprehensive analysis of pressure changes and leak confidence allows for precise identification of leak locations, improving detection accuracy and reliability, effectively reducing the risk of missed detections, ensuring comprehensive monitoring and timely early warning of the LPG supply network, and guaranteeing the safe operation of the system. In summary, this application's solution can achieve confidence-based leak monitoring of the LPG supply network based on multi-source monitoring point information, thereby improving the reliability of LPG supply safety monitoring. Attached Figure Description

[0047] Figure 1 This is an exemplary flowchart of an intelligent monitoring method for liquefied petroleum gas supply according to some embodiments of this application;

[0048] Figure 2 This is a schematic flowchart illustrating the process of determining a heat distribution map according to some embodiments of this application;

[0049] Figure 3 This is a flowchart illustrating the process of determining an association detection tree according to some embodiments of this application;

[0050] Figure 4 This is a schematic diagram of the structure of an intelligent monitoring system for liquefied petroleum gas supply according to some embodiments of this application;

[0051] Figure 5 This is a schematic diagram of the structure of a computer device for implementing an intelligent monitoring method for liquefied petroleum gas supply, according to some embodiments of this application. Detailed Implementation

[0052] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] refer to Figure 1 The figure is an exemplary flowchart of an intelligent monitoring method for liquefied petroleum gas supply according to some embodiments of this application. The intelligent monitoring method 100 for liquefied petroleum gas supply mainly includes the following steps:

[0054] In step 101, multiple sensor nodes are deployed in the liquefied petroleum gas supply network, and the pressure status of liquefied petroleum gas at each sensor node is monitored by intelligent pressure sensors.

[0055] It should be noted that the liquefied petroleum gas (LPG) supply network in this application refers to the pipeline network used for LPG supply; the pressure state of LPG in this application refers to characteristic data measuring the pressure state of LPG (i.e., the pressure state is composed of static pressure data and dynamic pressure data measuring the pressure state of LPG). By extracting the pressure change characteristics of LPG during supply and transportation, it is possible to quickly detect whether there are abnormal fluctuations or leakage risks in the supply network (i.e., the pipeline network); in addition, the dynamic pressure data refers to the real-time data of pressure fluctuations during the flow of LPG in the pipeline network. It is different from static pressure data and can capture the pressure fluctuation characteristics caused by fluid flow and equipment operation factors. It can provide information on the dynamic behavior of LPG in the pipeline and is often used to analyze early warning of pipeline leakage; the static pressure data refers to the steady-state pressure data during the flow of LPG in the pipeline network.

[0056] In practice, distributed sensor nodes can be deployed at key pipe sections, bends, and interfaces of the liquefied petroleum gas supply network. The pressure data at each sensor node location can be collected in real time by intelligent pressure sensors. The pressure data includes static pressure data and dynamic pressure data, and the collected pressure data is used to reflect the pressure status of liquefied petroleum gas at the sensor node.

[0057] It should be noted that by deploying multiple sensor nodes in the supply network, this application enables comprehensive real-time monitoring of the entire liquefied petroleum gas pipeline network, avoiding the limitations of traditional single sensors, effectively improving the accuracy of the data, and providing comprehensive pressure data support for subsequent leak detection.

[0058] In step 102, the dynamic pressure characteristics of liquefied petroleum gas at each sensor node are extracted from the pressure state. Based on all the dynamic pressure characteristics, the cross-correlation coefficients of pressure attenuation between different sensor nodes are determined. Then, the thermodynamic distribution map of liquefied petroleum gas pressure attenuation is determined by all the cross-correlation coefficients and the topology of the sensor nodes.

[0059] In some embodiments, extracting the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state can be achieved by the following steps:

[0060] For each sensor node, the dynamic pressure data of liquefied petroleum gas at the sensor node is obtained from the pressure state.

[0061] Based on the dynamic pressure data, the dynamic pressure characteristics of liquefied petroleum gas at the sensor nodes are determined, and then the dynamic pressure characteristics of liquefied petroleum gas at each sensor node are obtained.

[0062] It should be noted that the dynamic pressure data in this application refers to the set of instantaneous pressure change amplitudes during the flow of liquefied petroleum gas in a pipeline. Dynamic pressure data can reflect the dynamic characteristics of liquefied petroleum gas flow. In addition, dynamic pressure data usually has a high time resolution and has periodic and nonlinear characteristics. The dynamic pressure characteristics in this application refer to the characteristics that can reflect the degree of transient pressure fluctuation of liquefied petroleum gas.

[0063] In specific implementation, the dynamic pressure characteristics of liquefied petroleum gas at the sensor node can be determined based on the dynamic pressure data in the following way: the mean dynamic pressure, variance dynamic pressure, and dominant frequency of dynamic pressure fluctuations can be extracted from the dynamic pressure data using existing statistical analysis techniques, and then the extracted mean dynamic pressure, variance dynamic pressure, and dominant frequency of dynamic pressure fluctuations can be used as the dynamic pressure characteristics of liquefied petroleum gas at the sensor node.

[0064] In some embodiments, determining the cross-correlation coefficient of pressure attenuation between different sensor nodes based on all dynamic pressure characteristics can be achieved using the following steps:

[0065] Select a sensor node as the target sensor node;

[0066] Determine the comparison matrix of dynamic pressure characteristics between the target sensor node and its adjacent sensor nodes based on all dynamic pressure characteristics;

[0067] The cross-correlation coefficient of pressure attenuation between the target sensor node and its adjacent sensor nodes is determined by the comparison matrix.

[0068] Continue to determine the cross-correlation coefficients of pressure attenuation between the remaining sensor nodes and adjacent sensor nodes.

[0069] It should be noted that the comparison matrix in this application is a matrix that quantifies the degree of difference in dynamic pressure characteristics between different sensor nodes; the cross-correlation coefficient of pressure attenuation in this application is an indicator that measures the similarity of the attenuation trend of liquefied petroleum gas pressure with distance or structural topology changes between different sensor nodes.

[0070] In specific implementation, determining the comparison matrix of dynamic pressure characteristics between the target sensor node and its adjacent sensor nodes based on all dynamic pressure characteristics can be achieved in the following way: Obtain the adjacent sensor nodes of the target sensor node, and filter out the dynamic pressure characteristics corresponding to each adjacent sensor node from all dynamic pressure characteristics. Use the Euclidean distance between the dynamic pressure characteristics of the target sensor node and each adjacent sensor node as the comparison value of the dynamic pressure characteristics between the target sensor node and each adjacent sensor node, and use the matrix formed by all comparison values ​​as the comparison matrix of dynamic pressure characteristics between the target sensor node and its adjacent sensor nodes. Determining the cross-correlation coefficient of pressure attenuation between the target sensor node and its adjacent sensor nodes through the comparison matrix can be achieved in the following way: Obtain the corresponding sensor value from each dynamic pressure characteristic. The average dynamic pressure of liquefied petroleum gas at the target sensor node is used as the absolute difference between the average dynamic pressure of the target sensor node and each adjacent sensor node. Then, the comparison value of the dynamic pressure characteristics between the target sensor node and each adjacent sensor node is obtained from the comparison matrix. The product of the pressure attenuation value and the comparison value is used as the cross-correlation coefficient of pressure attenuation between the target sensor node and each adjacent sensor node. The cross-correlation coefficient of pressure attenuation between the remaining sensor nodes and adjacent sensor nodes can be determined by repeatedly selecting the remaining sensor nodes as target sensor nodes.

[0071] In some embodiments, reference Figure 2As shown in the figure, this is a schematic flowchart of the process for determining the thermal distribution map in some embodiments of this application. In this embodiment, the determination of the thermal distribution map of liquefied petroleum gas pressure decay based on all cross-correlation coefficients and the topology of sensor nodes can be achieved by the following steps:

[0072] In step 1021, the topology diagram of the sensor nodes is determined based on the spatial distance between the sensor nodes;

[0073] In step 1022, for each sensor node, the topological relationship between the sensor node and its adjacent sensor nodes is determined according to the topology diagram.

[0074] In step 1023, the pressure attenuation factor of liquefied petroleum gas at the sensor node is determined by the cross-correlation coefficient of pressure attenuation between the sensor node and adjacent sensor nodes and the topological relationship, thereby obtaining the pressure attenuation factor of liquefied petroleum gas at each sensor node.

[0075] In step 1024, a thermodynamic distribution map of liquefied petroleum gas pressure decay is determined based on all pressure decay factors.

[0076] It should be noted that the topology diagram in this application refers to a graphic representation of the spatial connection and relative positional relationships between sensor nodes, used to clarify the layout of each sensor node in the liquefied petroleum gas supply network; the pressure attenuation factor in this application is an indicator that measures the degree of attenuation of liquefied petroleum gas pressure at a sensor node during transmission along adjacent nodes; and the thermal distribution diagram of pressure attenuation in this application refers to a state diagram used to show the distribution of pressure attenuation at each sensor node in the liquefied petroleum gas supply network.

[0077] In specific implementation, determining the topology of sensor nodes based on the spatial distance between them can be achieved as follows: The spatial coordinates of each sensor node are obtained through a pipeline geographic information system. The geographical distance between each sensor node is used as the topological distance of the node topology. Each sensor node is treated as a graph node, and the graph nodes are connected using the topological distance to obtain a connection graph. This connection graph is then used as the topology diagram of the sensor nodes. Determining the topological relationship between a sensor node and its adjacent nodes based on this topology diagram can be achieved as follows: The natural exponential function value of the inverse of the topological distance between a sensor node and its adjacent nodes in the topology diagram can be used as the topological relationship between them. The relationship can also be determined through the pressure attenuation relationship between sensor nodes and their adjacent nodes. The pressure attenuation factor of liquefied petroleum gas at the sensor node can be determined by the following method: the product of the cross-correlation coefficient of pressure attenuation between the sensor node and its adjacent sensor nodes and the topological relationship can be used as the pressure attenuation factor of liquefied petroleum gas at the sensor node. The heat map of liquefied petroleum gas pressure attenuation can be determined by the following method: the pressure attenuation factors of all sensor nodes are mapped to the topological structure diagram, and the pressure attenuation degree of each sensor node is visually presented with color gradient using a heat map drawing tool (such as the imshow function of Matplotlib). The image output by the heat map drawing tool is used as the heat map of pressure attenuation of the liquefied petroleum gas supply network. The above heat map can realize the visual monitoring of the pressure anomaly area of ​​the entire network.

[0078] It should be noted that, through the extraction of dynamic pressure characteristics and the calculation of cross-correlation coefficients, this application can accurately reveal the pressure attenuation relationship between various sensor nodes, and construct a thermal distribution map through the topology. This pressure attenuation trend analysis of the entire network helps to identify potential abnormal areas in the liquefied petroleum gas supply network, improves the comprehensive understanding of the pressure status of the supply network, and enhances the accuracy of leak location.

[0079] In step 103, the pressure leakage status of the transportation segment between each pair of adjacent sensor nodes is evaluated based on the mutual support of the pressure fluctuation amplitudes between all adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transportation segment. Then, the correlation detection tree of liquefied petroleum gas supply network leakage is determined by the confidence probability of pressure leakage in all transportation segments.

[0080] In some embodiments, the confidence probability of pressure leakage in each transport segment between two adjacent sensor nodes is obtained by performing a correlation confidence evaluation based on the mutual support of pressure fluctuation amplitudes among all adjacent sensor nodes, which can be achieved by the following steps:

[0081] Determine the degree of mutual support between pressure fluctuation amplitudes of adjacent sensor nodes;

[0082] Based on all mutual support, construct a correlation matrix of pressure fluctuation behavior in the transportation segment between adjacent sensor nodes, and map the values ​​in the correlation matrix to fuzzy sets;

[0083] Initialize the membership degree of the pressure fluctuation amplitude for each transport segment;

[0084] The fuzzy evaluation matrix is ​​determined by all membership degrees and the fuzzy set;

[0085] Based on the fuzzy evaluation matrix, a fuzzy confidence evaluation is performed on the pressure leakage status of the transportation segment between every two adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transportation segment.

[0086] It should be noted that the mutual support in this application is a correlation index that measures the pressure changes between adjacent sensor nodes; the correlation matrix in this application refers to a matrix used to represent the correlation of pressure fluctuation behavior between adjacent sensor nodes; the fuzzy evaluation matrix in this application refers to a matrix used to represent the relationship between pressure fluctuation behavior and pressure leakage status in each transportation segment; and the confidence probability of pressure leakage in this application measures the degree of confidence in the occurrence of a liquefied petroleum gas supply network leakage event.

[0087] In specific implementation, determining the mutual support of pressure fluctuation amplitudes between adjacent sensor nodes can be achieved as follows: For each sensor node, acquire pressure fluctuation data of the sensor node and its adjacent sensor nodes. Extract the pressure fluctuation amplitude features of the corresponding sensor node from each pressure fluctuation data. These features are specifically a time series, where the amplitude values ​​are the maximum fluctuation amplitudes within each minute of the pressure fluctuation data. Then, use the Pearson correlation coefficient of the pressure fluctuation amplitude features between the sensor node and its adjacent sensor nodes as the mutual support of pressure fluctuation amplitudes between them. This yields the mutual support of pressure fluctuation amplitudes between each sensor node and its adjacent sensor nodes. Based on all the mutual support values, construct a correlation matrix of pressure fluctuation behavior in the transportation segment between adjacent sensor nodes, and then... The numerical mapping in the matrix to a fuzzy set can be achieved as follows: A correlation matrix is ​​constructed using these mutual support values, where each element represents the pressure fluctuation relationship between adjacent sensor nodes. The numerical values ​​in this correlation matrix are then mapped to a fuzzy set. A membership function (such as a Gaussian fuzzy function) is used to represent the uncertainty of the pressure fluctuation behavior. It should be noted that the Gaussian fuzzy function can generate a continuous membership value based on the magnitude of the fluctuation. The membership value ranges from 0 to 1, representing the degree of matching between the pressure fluctuation magnitude and a certain fuzzy state. For example, when the fluctuation magnitude is small and the consistency between adjacent nodes is strong, the membership value is close to 1; conversely, it is close to 0. In this way, the originally precise numerical relationship is transformed into a fuzzy membership value, thereby better handling the uncertainty and fuzziness in pressure fluctuations and aiding subsequent fuzzy inference and confidence evaluation.

[0088] Furthermore, in practical implementation, the initialization of the membership degree of the pressure fluctuation amplitude of each transportation segment can be achieved in the following way: Common fuzzy logic methods (such as minimum value or weighted average) can be used to initialize the membership degree of the pressure fluctuation amplitude of each transportation segment. These membership degrees represent the degree of pressure fluctuation amplitude in each pipeline segment. The determination of the fuzzy evaluation matrix through all membership degrees and the fuzzy set can be achieved in the following way: The membership degree of the pressure fluctuation amplitude of each transportation segment can be combined with the fuzzy set, and then an existing fuzzy inference model (such as the Takagi-Sugeno model) can be used to compare the input membership degrees with a rule base (such as a fuzzy rule base based on expert experience). The Takagi-Sugeno model combines multiple rules for fuzzy inference. Each rule typically takes the form of "if...then...", such as "if the pressure fluctuation is large, the leakage probability is high." Then, using the input membership values, the fuzzy inference model activates each rule and calculates its output. The output is usually a linear function or a constant value, representing the pressure leakage probability of each transport segment. Based on these multiple rules, the Takagi-Sugeno model uses a weighted average method to fuse the outputs of each rule, generating a final fuzzy evaluation value for each transport segment. This value reflects the relationship between the pressure fluctuation amplitude and the pressure leakage state. Finally, the... All fuzzy evaluation results are summarized into a fuzzy evaluation matrix for further pressure leakage status determination. Based on the fuzzy evaluation matrix, a fuzzy confidence evaluation is performed on the pressure leakage status of the transport segment between every two adjacent sensor nodes. The confidence probability of pressure leakage in each transport segment can be obtained in the following way: the fuzzy evaluation matrix contains the fuzzy relationship between the pressure fluctuation amplitude and leakage status of each transport segment. To perform fuzzy confidence evaluation, the evaluation system first calculates the leakage probability of each transport segment based on the membership degree and evaluation value in the fuzzy evaluation matrix, and then combines each fuzzy evaluation value with the confidence level using techniques such as weighted average or fuzzy addition. The comprehensive leakage probability of each transportation segment is obtained. For example, fuzzy inference algorithms (such as weighted average or maximum membership method) can be used to synthesize the fuzzy membership degrees between each sensor node to generate a more accurate leakage confidence score. This process can output a probability value based on the pressure fluctuation and membership degree evaluation value of each transportation pipeline segment, and then normalize the output probability value. The normalized probability value is then used as the confidence probability of pressure leakage in the corresponding transportation segment. The confidence probability represents the possibility of pressure leakage in each transportation segment. This fuzzy confidence evaluation method can handle the uncertainty and fuzziness of the data and improve the accuracy and robustness of leakage status determination.

[0089] In some embodiments, reference Figure 3As shown in the figure, this is a flowchart illustrating the process of determining the correlation detection tree in some embodiments of this application. In this embodiment, determining the correlation detection tree for liquefied petroleum gas supply network leaks based on the confidence probability of pressure leaks in all transportation sections can be achieved using the following steps:

[0090] In step 1031, each transport segment is treated as a tree node;

[0091] In step 1032, the connection relationship between adjacent tree nodes is determined by the confidence probability of pressure leakage between adjacent transport sections;

[0092] In step 1033, the associated detection tree for liquefied petroleum gas supply network leaks is determined based on all tree nodes and all connections.

[0093] It should be noted that the correlation detection tree in this application is a tree structure used to measure the correlation and impact of leaks between different transportation segments in the liquefied petroleum gas supply network.

[0094] In practical implementation, treating each transportation segment as a tree node can be achieved as follows: Each transportation segment can be treated as a tree node, and an initial graph structure containing all transportation segments can be created. The connection relationships between adjacent tree nodes can be determined by the confidence probability of pressure leakage between adjacent transportation segments, which can be used as the connection weights between adjacent transportation segments. The connection weights reflect the connection relationships between adjacent tree nodes. Specifically, a weighted graph model in graph theory can be used, where the weight of each connection is the confidence probability of pressure leakage between two adjacent transportation segments. The liquefied petroleum gas supply network is then determined based on all tree nodes and all connection relationships. The associated detection tree for leaks can be implemented in the following way: a tree can be generated using an existing tree-forming algorithm based on all tree nodes and the connection relationships between them, and the generated tree can be used as the associated detection tree for leaks in the liquefied petroleum gas supply network. It should be noted that the technical principle of the Prim algorithm used in this application to generate the associated detection tree is as follows: a weighted graph is constructed based on the connection relationships (i.e., confidence probabilities) between sensor nodes, where the weight of each connection corresponds to the confidence probability of pressure leakage between adjacent sensor nodes. The Prim algorithm will first sort all edges according to their weights from largest to smallest, and then gradually select the edge with the largest weight and add it to the tree to avoid forming loops, until all tree nodes are included.

[0095] It should be noted that this application uses the mutual support and confidence of pressure fluctuation amplitudes of adjacent sensor nodes to quantify and synthesize the leakage risk of each transportation segment, generate the confidence probability of leakage, and finally construct an association detection tree. This process can provide a probability-based leakage assessment for the entire supply network, enhancing the robustness and leakage prediction capability of the detection system.

[0096] In step 104, the liquefied petroleum gas supply network is fused and leaked by using the thermal distribution map of the liquefied petroleum gas pressure decay and the associated detection tree to obtain the leakage area of ​​the liquefied petroleum gas supply network.

[0097] In some embodiments, the fusion leak detection of the liquefied petroleum gas supply network using the thermal distribution map of liquefied petroleum gas pressure decay and the associated detection tree can be achieved by the following steps:

[0098] Determine the fusion detection weights of the heat map and the associated detection tree;

[0099] Based on the fusion detection weights of the thermal distribution map and the associated detection tree, the abnormal information of the thermal distribution map and the associated detection tree is integrated to obtain the integrated information for leak detection.

[0100] The integrated information from the leak detection was used to determine the area where the leak occurred in the liquefied petroleum gas supply network.

[0101] It should be noted that the fusion detection weight in this application is an indicator that quantifies the contribution of the thermal distribution map and the correlation detection tree to leak detection; the leak occurrence area in this application refers to the specific transportation section area in the liquefied petroleum gas supply network that is determined to have pressure leakage anomalies through fusion detection results.

[0102] In specific implementation, the fusion detection weights of the thermal distribution map and the associated detection tree can be determined in the following way: The thermal distribution map and the associated detection tree can be weighted according to the importance of the pressure attenuation region in the thermal distribution map and the strength of the pressure leakage confidence probability in the associated detection tree. The weight assigned to the thermal distribution map is used as the fusion detection weight of the thermal distribution map, and the weight assigned to the associated detection tree is used as the fusion detection weight of the associated detection tree. It should be noted that the specific implementation process of weight allocation is as follows: First, the pressure attenuation factor of each sensor node in the thermal distribution map is standardized, and the importance score of its region is set according to the significance of the pressure attenuation amplitude. Simultaneously, the pressure leakage confidence probability of each transport section in the associated detection tree is normalized to quantify its leakage correlation strength. Then, a weighted average method is used to normalize the importance scores of both to the same scale. The total weight is set to 1, and the weights of the thermal distribution map and the associated detection tree are allocated using linear weighting or entropy weighting. The integrated information for leak detection is obtained by integrating the abnormal information of the thermal distribution map and the associated detection tree according to their respective fusion detection weights. This can be achieved by using an anomaly detection algorithm from the prior art to identify abnormal areas in the thermal distribution map and the associated detection tree, then weighting and integrating these abnormal areas according to their respective fusion detection weights, and using the integrated information as the integrated information for leak detection. The leak location of the liquefied petroleum gas supply network can be determined using the integrated information for leak detection by comparing the integrated information (i.e., the abnormal area) with the topology map of the sensor nodes.

[0103] It should be noted that by integrating the information from the thermal distribution map and the associated detection tree, this application can effectively combine the pressure decay pattern with the leakage confidence level to perform global leakage detection. This multi-dimensional fusion detection method can accurately identify leakage areas in a wider range, reduce the risk of missed detection, and improve the safety and reliability of the liquefied petroleum gas supply network.

[0104] On the other hand, in some embodiments, this application provides an intelligent monitoring system for liquefied petroleum gas supply, with reference to... Figure 4 The figure is a schematic diagram of the structure of an intelligent monitoring system for liquefied petroleum gas supply according to some embodiments of this application. The intelligent monitoring system 400 for liquefied petroleum gas supply includes: a monitoring module 401, a processing module 402, and an execution module 403, which are described below:

[0105] Monitoring module 401, in this application, is mainly used to deploy multiple sensor nodes in the liquefied petroleum gas supply network and monitor the pressure status of liquefied petroleum gas at each sensor node through intelligent pressure sensors.

[0106] Processing module 402, in this application, is used to extract the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state, determine the cross-correlation coefficient of pressure attenuation between different sensor nodes based on all dynamic pressure characteristics, and then determine the thermodynamic distribution map of liquefied petroleum gas pressure attenuation based on all cross-correlation coefficients and the topology of sensor nodes.

[0107] In this application, the processing module 402 is also used to perform a correlation confidence evaluation on the pressure leakage status of the transportation section between each pair of adjacent sensor nodes based on the mutual support of the pressure fluctuation amplitude between all adjacent sensor nodes, to obtain the confidence probability of pressure leakage in each transportation section, and then determine the correlation detection tree of liquefied petroleum gas supply network leakage through the confidence probability of pressure leakage in all transportation sections.

[0108] The execution module 403 in this application is mainly used to perform fusion leak detection on the liquefied petroleum gas supply network by using the thermal distribution map of the liquefied petroleum gas pressure decay and the associated detection tree, so as to obtain the leakage area of ​​the liquefied petroleum gas supply network.

[0109] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described intelligent monitoring method for liquefied petroleum gas supply.

[0110] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing an intelligent monitoring method for liquefied petroleum gas supply according to some embodiments of this application. The intelligent monitoring method for liquefied petroleum gas supply in the above embodiments can... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0111] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0112] The communication bus 502 can be used to transmit information between the aforementioned components.

[0113] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0114] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. In the above embodiments, the intelligent monitoring method for liquefied petroleum gas supply can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0115] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area network (WLAN), etc.

[0116] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0117] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0118] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described intelligent monitoring method for liquefied petroleum gas supply.

[0119] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0120] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A smart monitoring method for liquefied petroleum gas supply, characterized in that, Includes the following steps: Multiple sensor nodes are deployed in the liquefied petroleum gas supply network, and the pressure status of liquefied petroleum gas at each sensor node is monitored by intelligent pressure sensors. Extract the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state, determine the cross-correlation coefficient of pressure attenuation between different sensor nodes based on all dynamic pressure characteristics, and then determine the thermodynamic distribution map of liquefied petroleum gas pressure attenuation based on all cross-correlation coefficients and the topology of sensor nodes. Based on the mutual support of pressure fluctuation amplitudes between all adjacent sensor nodes, the correlation confidence evaluation of the pressure leakage status of the transportation segment between each pair of adjacent sensor nodes is carried out to obtain the confidence probability of pressure leakage in each transportation segment. Then, the correlation detection tree of liquefied petroleum gas supply network leakage is determined by the confidence probability of pressure leakage in all transportation segments. By combining the thermal distribution map of liquefied petroleum gas pressure decay and the associated detection tree, the leakage detection of the liquefied petroleum gas supply network is performed to obtain the leakage area of ​​the liquefied petroleum gas supply network.

2. The method as described in claim 1, characterized in that, Extracting the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state specifically includes: For each sensor node, the dynamic pressure data of liquefied petroleum gas at the sensor node is obtained from the pressure state. Based on the dynamic pressure data, the dynamic pressure characteristics of liquefied petroleum gas at the sensor nodes are determined, and then the dynamic pressure characteristics of liquefied petroleum gas at each sensor node are obtained.

3. The method as described in claim 1, characterized in that, The cross-correlation coefficients for pressure attenuation between different sensor nodes are determined based on all dynamic pressure characteristics, specifically including: Select a sensor node as the target sensor node; Determine the comparison matrix of dynamic pressure characteristics between the target sensor node and its adjacent sensor nodes based on all dynamic pressure characteristics; The cross-correlation coefficient of pressure attenuation between the target sensor node and its adjacent sensor nodes is determined by the comparison matrix. Continue to determine the cross-correlation coefficients of pressure attenuation between the remaining sensor nodes and adjacent sensor nodes.

4. The method as described in claim 1, characterized in that, The thermodynamic distribution map of LPG pressure decay, determined by all cross-correlation coefficients and the topology of sensor nodes, specifically includes: The topology of the sensor nodes is determined based on the spatial distance between them. For each sensor node, the topological relationship between the sensor node and its adjacent sensor nodes is determined based on the topology diagram. The pressure attenuation factor of liquefied petroleum gas at the sensor node is determined by the cross-correlation coefficient of pressure attenuation between the sensor node and its adjacent sensor nodes and the topological relationship, thereby obtaining the pressure attenuation factor of liquefied petroleum gas at each sensor node. The thermodynamic distribution map of LPG pressure decay was determined based on all pressure decay factors.

5. The method as described in claim 1, characterized in that, Based on the mutual support of pressure fluctuation amplitudes among all adjacent sensor nodes, a correlation confidence evaluation is performed on the pressure leakage status of the transport segment between every two adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transport segment, specifically including: Determine the degree of mutual support between pressure fluctuation amplitudes of adjacent sensor nodes; Based on all the mutual support, construct the correlation matrix of pressure fluctuation behavior in the transportation segment between adjacent sensor nodes, and map the values ​​in the correlation matrix to fuzzy sets; Initialize the membership degree of the pressure fluctuation amplitude for each transport segment; The fuzzy evaluation matrix is ​​determined by all membership degrees and the fuzzy set; Based on the fuzzy evaluation matrix, a fuzzy confidence evaluation is performed on the pressure leakage status of the transportation segment between every two adjacent sensor nodes to obtain the confidence probability of pressure leakage in each transportation segment.

6. The method as described in claim 1, characterized in that, The association detection tree for determining leaks in the LPG supply network based on the confidence probability of pressure leaks across all transportation segments specifically includes: Each transportation segment is treated as a tree node; The connection relationship between adjacent tree nodes is determined by the confidence probability of pressure leakage between adjacent transport sections; The associated detection tree for liquefied petroleum gas supply network leaks is determined based on all tree nodes and all connections.

7. The method as described in claim 1, characterized in that, By combining the thermal distribution map of liquefied petroleum gas pressure decay and the correlation detection tree to perform fusion leak detection on the liquefied petroleum gas supply network, the specific leak locations in the liquefied petroleum gas supply network are obtained, including: Determine the fusion detection weights of the heat map and the associated detection tree; Based on the fusion detection weights of the thermal distribution map and the associated detection tree, the abnormal information of the thermal distribution map and the associated detection tree is integrated to obtain the integrated information for leak detection. The integrated information from the leak detection was used to determine the area where the leak occurred in the liquefied petroleum gas supply network.

8. An intelligent monitoring system for liquefied petroleum gas supply, characterized in that, include: The monitoring module is used to deploy multiple sensor nodes in the liquefied petroleum gas supply network and monitor the pressure status of liquefied petroleum gas at each sensor node through intelligent pressure sensors. The processing module is used to extract the dynamic pressure characteristics of liquefied petroleum gas at each sensor node from the pressure state, determine the cross-correlation coefficient of pressure decay between different sensor nodes based on all the dynamic pressure characteristics, and then determine the thermodynamic distribution map of liquefied petroleum gas pressure decay based on all the cross-correlation coefficients and the topology of the sensor nodes. The processing module is also used to perform a correlation confidence evaluation on the pressure leakage status of the transportation section between each pair of adjacent sensor nodes based on the mutual support of the pressure fluctuation amplitude between all adjacent sensor nodes, to obtain the confidence probability of pressure leakage in each transportation section, and then determine the correlation detection tree of liquefied petroleum gas supply network leakage through the confidence probability of pressure leakage in all transportation sections. The execution module is used to perform fusion leak detection on the liquefied petroleum gas supply network by using the thermal distribution map of the liquefied petroleum gas pressure decay and the associated detection tree, so as to obtain the leakage area of ​​the liquefied petroleum gas supply network.

9. A computer device comprising a memory and a processor, the memory storing code, characterized in that, The processor is configured to acquire the code and execute the intelligent monitoring method for liquefied petroleum gas supply as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the intelligent monitoring method for liquefied petroleum gas supply as described in any one of claims 1 to 7.

Images