An intelligent diagnostic system for the communication network of an automatic fuel dispenser

Abstract

The present invention discloses an intelligent diagnosis system for the communication network of an automatic fuel dispenser, belonging to the technical field of intelligent diagnosis; filing and registering the terminal devices in a gas station, and constructing a virtual channel mapping diagram; using a communication parameter acquisition device embedded in the terminal device to collect the link communication attributes of the communication links in the connection edges, and constructing a time acquisition sequence; calculating the communication quality of the communication links based on the link communication attributes under the time acquisition sequence, and calculating the comprehensive communication quality of the connection edges; setting the communication anomaly judgment rules for the connection edges, and constructing an anomaly judgment Boolean matrix; constructing the eigenvectors of the connection edges under the time acquisition sequence, calculating the communication anomaly similarity between the connection edges, presetting a threshold value, analyzing and giving early warnings, which not only realizes the closed-loop process from link acquisition to diagnosis and early warning, but also improves the diagnostic intelligence level of the communication network of the automatic fuel dispenser, and enhances the accuracy, responsiveness and forward-looking of network operation and maintenance.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent diagnosis, and in particular to an automatic fuel dispenser communication network intelligent diagnosis system. Background Art

[0002] In modern gas stations, the frequency of data exchange between gas pumps and backend servers and network switching equipment has increased significantly, and information transmission relies on the stability and real-time performance of the communication network. In this context, traditional network anomaly detection methods based on manual troubleshooting or regular inspections are no longer able to meet the high-frequency, low-tolerance business needs. Therefore, research on intelligent diagnostic technologies for automatic gas pump communication networks has important technical value and practical significance for ensuring the efficient operation of terminal equipment, improving service continuity, and enhancing system security. Although some gas stations have introduced network diagnostic tools based on ping detection, packet loss rate, or traffic monitoring in recent years, these technologies often lack the ability to comprehensively analyze communication link properties, cannot accurately identify anomaly types and associated links, and have a low level of intelligence, making them difficult to adapt to the complex and dynamic gas station communication environment.

[0003] Existing communication diagnosis technologies generally have the following shortcomings: most methods still judge network link anomalies at the level of static threshold detection or single-point indicator analysis, failing to achieve dynamic modeling and intelligent judgment of time series communication data, resulting in difficulty in ensuring the accuracy and timeliness of diagnosis; existing solutions lack the ability to model the entire gas station terminal network structure, especially in scenarios with complex relationships between nodes and diverse links, and are unable to form an intuitive communication graph model, which affects the mining of high-level information such as abnormal propagation paths and link similarities; traditional diagnostic methods find it difficult to comprehensively consider multiple communication dimensions such as signal strength, transmission rate, bit error rate, and make joint judgments, resulting in poor recognition of multi-attribute coupled anomalies. Summary of the Invention

[0004] The purpose of the present invention is to provide an automatic fuel dispenser communication network intelligent diagnosis system to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An intelligent diagnosis system for the communication network of an automatic fuel dispenser, comprising: a mapping construction module, a data acquisition module, a communication quality calculation and rule setting module, and a matrix construction and analysis warning module;

[0007] The mapping construction module is configured to file and register the terminal devices in the gas station and construct a virtual channel mapping map;

[0008] The data acquisition module: Based on the virtual channel mapping diagram, and using the communication parameter acquisition device embedded in the terminal device, acquire the link communication attributes of the communication links in the connection edges, and construct a time acquisition sequence;

[0009] The communication quality calculation and rule setting module: Based on the link communication attributes under the time acquisition sequence, calculate the communication quality of the communication links; Based on the communication quality, calculate the comprehensive communication quality of the connection edges, and set the communication anomaly judgment rules for the connection edges;

[0010] The matrix construction and analysis warning module: Based on the communication anomaly judgment rules, construct an anomaly judgment Boolean matrix; Based on the anomaly judgment Boolean matrix, construct the eigenvectors of the connection edges under the time acquisition sequence; Based on the eigenvectors of different connection edges, calculate the communication anomaly similarity between the connection edges, preset a threshold, analyze and issue a warning.

[0011] Furthermore, the mapping diagram construction module includes a mapping diagram construction unit;

[0012] The mapping diagram construction unit: File and record the terminal devices in the gas station for constructing a virtual channel mapping diagram. The terminal devices include fuel dispensers, network switches, and servers; The fuel dispenser communicates with the server via at least one of the network switches;

[0013] The method for constructing the virtual channel mapping diagram is as follows:

[0014] Map a terminal device to a graph node, and map the communication links between the fuel dispenser, network switch, and server to the connection edges between the graph nodes.

[0015] Furthermore, the data acquisition module includes a data acquisition unit;

[0016] The data acquisition unit: Based on the virtual channel mapping diagram, and using the communication parameter acquisition device embedded in the terminal device, acquire the link communication attributes of the communication links in the connection edges. The link communication attributes include signal strength, transmission rate, and bit error rate; Construct a time acquisition sequence.

[0017] Furthermore, the communication quality calculation and rule setting module includes a communication quality calculation unit and a rule setting unit;

[0018] The communication quality calculation unit: Based on the signal strength, transmission rate, and bit error rate of the communication link under the time acquisition sequence, calculate the communication quality of the communication link; Based on the communication quality, calculate the comprehensive communication quality of the connection edge under the time acquisition sequence;

[0019] The rule setting unit: Set the communication anomaly judgment rules for the connection edges.

[0020] Further, the matrix construction and analysis and warning module includes a matrix construction unit and an analysis and warning unit;

[0021] The matrix construction unit: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix, where the rows of the anomaly judgment Boolean matrix are all communication links in the connection edges, and the columns of the anomaly judgment Boolean matrix are anomaly types, and the anomaly types include signal strength anomaly, transmission rate anomaly, bit error rate anomaly, and multi-attribute anomaly;

[0022] The analysis and warning unit: Based on the anomaly judgment Boolean matrix, construct a feature vector of the connection edges under the time acquisition sequence; use the cosine similarity to calculate the communication anomaly similarity between the feature vectors of different connection edges under the time acquisition sequence, preset a similarity threshold, if the communication anomaly similarity is greater than or equal to the similarity threshold, then judge that there is a communication anomaly similarity between the connection edges; real-time diagnose whether there is a network communication anomaly and the cause of the anomaly in the connection edges, obtain all the connection edges with communication anomaly similarity to the connection edges, and conduct unified warning.

[0023] An intelligent diagnosis method for the communication network of an automatic fuel dispenser, the method includes the following steps: Step S1: File and record the terminal devices in the gas station and construct a virtual channel mapping diagram; Step S2: Based on the virtual channel mapping diagram, and using the communication parameter acquisition device embedded in the terminal devices, collect the link communication attributes of the communication links in the connection edges and construct a time acquisition sequence; Step S3: Based on the link communication attributes under the time acquisition sequence, calculate the communication quality of the communication links; based on the communication quality, calculate the comprehensive communication quality of the connection edges and set the communication anomaly judgment rule for the connection edges; Step S4: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix; based on the anomaly judgment Boolean matrix, construct a feature vector of the connection edges under the time acquisition sequence; based on the feature vectors of different connection edges, calculate the communication anomaly similarity between the connection edges, preset a threshold, analyze and conduct warning.

[0024] As a preferred solution of the intelligent diagnosis method for the communication network of the automatic fuel dispenser described in the present invention, file and record the terminal devices in the gas station for constructing a virtual channel mapping diagram, and the terminal devices include fuel dispensers, network switches, and servers; the fuel dispensers communicate with the servers via at least one of the network switches;

[0025] The method for constructing the virtual channel mapping diagram is as follows:

[0026] Map a terminal device to a graph node, and map the communication links between the fuel dispenser, the network switch, and the server to the connection edges between the graph nodes.

[0027] As a preferred solution of the intelligent diagnosis method for the communication network of an automatic fuel dispenser according to the present invention, based on the virtual channel mapping graph, and by using the communication parameter acquisition device embedded in the terminal device, the link communication attributes of the communication links in the connection edges are collected, and the link communication attributes include signal strength, transmission rate, and error rate; the i-th connection edge is denoted as , where , represents the j-th communication link in the i-th connection edge, represents the connection edge ; the total number of communication links in

[0028] A time acquisition sequence is constructed, denoted as , where represents the t-th time acquisition sequence, and T represents the total number of time acquisition sequences. The signal strength, transmission rate, and error rate of the communication link at the time acquisition sequence are respectively denoted as and .

[0029] As a preferred solution of the intelligent diagnosis method for the communication network of an automatic fuel dispenser according to the present invention, based on the signal strength of the communication link , the transmission rate , and the error rate at the time acquisition sequence , the communication quality of the communication link is calculated, and the calculation formula is as follows:

[0030] ;

[0031] where represents the communication quality of the communication link , represents the preset signal strength influence factor, represents the preset transmission rate influence factor, represents the preset error rate influence factor, and respectively represent the minimum and maximum values of the signal strength within the preset normal range, and respectively represent the minimum and maximum values of the transmission rate within the preset normal range, and respectively represent the minimum and maximum values of the error rate within the preset normal range;

[0032] It should be noted that in the actual communication network of automatic fuel dispensers, the conditions of different communication links are complex. This formula can quantify and synthesize various communication attributes to comprehensively and accurately evaluate the quality of each communication link. By adjusting the influencing factors and , the influence weights of each attribute on communication quality can be flexibly adjusted according to actual needs. For example, if there is a large interference in the surrounding environment of a gas station and the signal strength fluctuation has a serious impact on communication, the value of can be appropriately increased to highlight the importance of signal strength in the evaluation of communication quality, providing a quantitative basis for accurately judging whether the communication link is abnormal subsequently.

[0033] Based on the communication quality of the communication link , calculate the comprehensive communication quality of the connection edge under the time acquisition sequence . The calculation formula is as follows:

[0034] ;

[0035] where represents the comprehensive communication quality of the connection edge under the time acquisition sequence , represents the total number of communication links in the connection edge .

[0036] It should be noted that in the gas station communication network, a connection edge may contain multiple communication links, and the communication quality of each link may be different. By calculating the comprehensive communication quality of the connection edge, the communication status of the connection edge can be grasped as a whole. For example, when judging whether the connection edge between the fuel dispenser and the network switch is normal, the comprehensive communication quality can comprehensively reflect the overall situation of all communication links within the connection edge, avoiding ignoring the overall communication status due to local problems of individual links. Combined with the preset comprehensive communication quality threshold, once the comprehensive communication quality of the connection edge exceeds the threshold, it can be quickly determined that there is a network communication abnormality in the connection edge, providing an important basis for subsequent analysis of abnormal causes and early warning, helping to timely discover and solve network communication problems, and ensuring the stable operation of the gas station communication network.

[0037] Set the communication abnormality judgment rule for the connection edge as follows:

[0038] Preset the comprehensive communication quality threshold. If the comprehensive communication quality of the connection edge is greater than or equal to the comprehensive communication quality threshold, it is determined that there is a network communication abnormality in the connection edge , and analyze the abnormal reasons as follows:

[0039] If , it is determined that the cause of the abnormality is that the signal strength of the communication link is abnormal;

[0040] If , it is determined that the cause of the abnormality is that the transmission rate of the communication link is abnormal;

[0041] If , it is determined that the cause of the abnormality is that the bit error rate of the communication link is abnormal;

[0042] If there are multiple abnormalities in the link communication attributes, it is determined that the connection edge has a network communication abnormality due to multiple attribute abnormalities in the communication link .

[0043] As a preferred solution of the intelligent diagnosis method for the communication network of an automatic fuel dispenser described in the present invention, based on the communication abnormality judgment rule, an abnormality judgment Boolean matrix is constructed. The rows of the abnormality judgment Boolean matrix are all communication links in the connection edge , and the columns of the abnormality judgment Boolean matrix are abnormality types, where the abnormality types include signal strength abnormality, transmission rate abnormality, bit error rate abnormality, and multi-attribute abnormality;

[0044] If the connection edge has a network communication abnormality due to the m-th abnormality type of the j-th communication link, the matrix element corresponding to the j-th row and the m-th column is recorded as 1, otherwise it is 0;

[0045] Based on the abnormality judgment Boolean matrix, a feature vector of the connection edge under the time acquisition sequence is constructed and denoted as , where represents the ratio of the number of edges with signal strength abnormality in the connection edge under the time acquisition sequence to the total number of edges in the connection edge , represents the ratio of the number of edges with transmission rate abnormality in the connection edge under the time acquisition sequence to the total number of edges in the connection edge , represents the ratio of the number of edges with bit error rate abnormality in the connection edge under the time acquisition sequence to the total number of edges in the connection edge , represents the ratio of the number of edges with multi-attribute abnormality in the connection edge under the time acquisition sequence to the total number of edges in the connection edge Ratio of the total number of edges therein;

[0046] Using cosine similarity, calculate the time acquisition sequence Lower connection edge Feature vector of And the time acquisition sequence Lower connection edge Feature vector of Calculate the communication anomaly similarity between them, preset a similarity threshold, if the communication anomaly similarity is greater than or equal to the similarity threshold, then determine the connection edge And the connection edge There is a communication anomaly similarity;

[0047] Let t = t + 1, and diagnose in real time whether there is a network communication anomaly and the cause of the anomaly for the connection edge Obtain all connection edges that are similar to the communication anomaly of the connection edge And issue a unified warning.

[0048] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: In an intelligent diagnosis system for an automatic fuel dispenser communication network provided by the present invention, by filing and registering the terminal devices in the gas station and constructing a virtual channel mapping diagram, digital modeling and visual management of the gas station network structure are realized, providing a unified data basis for subsequent communication monitoring and anomaly location; on this basis, combined with the communication parameter acquisition device embedded in the terminal device, link communication attributes such as signal strength, transmission rate, and bit error rate are collected, and a time acquisition sequence is constructed, thereby realizing dynamic and continuous monitoring of the network state and laying a data foundation for subsequent time series analysis and quality assessment; further, by calculating the communication quality of each link and summarizing it into the comprehensive communication quality of the connection edge, and then combining the preset threshold and attribute itemization rules, accurate positioning and classification identification of the cause of the anomaly are realized, avoiding diagnostic deviations caused by misjudgment of a single anomaly index; subsequently, by constructing an anomaly judgment Boolean matrix and a feature vector, a structured and standardized anomaly expression method is formed, making the network communication anomaly quantifiable and comparable; finally, using the anomaly feature vector similarity between connection edges, areas or links with similar communication anomalies are identified, realizing early warning and cluster analysis of potential anomalies. The present invention not only realizes a closed-loop process from link acquisition to diagnosis and warning, but also improves the diagnostic intelligence level of the automatic fuel dispenser communication network, enhancing the accuracy, responsiveness, and forward-looking of network operation and maintenance. Description of the Drawings

[0049] The drawings are used to provide further understanding of the present invention, and constitute a part of the specification, and are used together with the embodiments of the present invention to explain the present invention, and do not constitute a limitation to the present invention.

[0050] Figure 1 It is a schematic structural diagram of an intelligent diagnosis system for the communication network of an automatic fuel dispenser according to the present invention;

[0051] Figure 2 It is a schematic diagram of the steps of an intelligent diagnosis method for the communication network of an automatic fuel dispenser according to the present invention. Specific embodiments

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Please refer to Figure 1 , in the first embodiment: An intelligent diagnosis system for the communication network of an automatic fuel dispenser is provided. The system includes: a mapping diagram construction module, a data acquisition module, a communication quality calculation and rule setting module, and a matrix construction and analysis warning module;

[0054] The mapping diagram construction module: files and records the terminal devices in the gas station and constructs a virtual channel mapping diagram;

[0055] The data acquisition module: based on the virtual channel mapping diagram and using the communication parameter acquisition device embedded in the terminal device, acquires the link communication attributes of the communication links in the connection edges and constructs a time acquisition sequence;

[0056] The communication quality calculation and rule setting module: calculates the communication quality of the communication link based on the link communication attributes under the time acquisition sequence; calculates the comprehensive communication quality of the connection edge based on the communication quality, and sets the communication anomaly judgment rule for the connection edge;

[0057] The matrix construction and analysis warning module: constructs an anomaly judgment Boolean matrix based on the communication anomaly judgment rule; constructs a feature vector of the connection edge under the time acquisition sequence based on the anomaly judgment Boolean matrix; calculates the communication anomaly similarity between the connection edges based on the feature vectors of different connection edges, presets a threshold, analyzes and issues a warning.

[0058] Further, the mapping diagram construction module includes a mapping diagram construction unit;

[0059] The mapping diagram construction unit: files and records the terminal devices in the gas station for constructing a virtual channel mapping diagram. The terminal devices include fuel dispensers, network switches, and servers; the fuel dispensers communicate with the servers via at least one of the network switches;

[0060] The method for constructing a virtual channel mapping graph is as follows:

[0061] Map a terminal device to a graph node, and map the communication links between the fuel dispenser, network switch, and server to the connection edges between graph nodes.

[0062] Furthermore, the data acquisition module includes a data acquisition unit;

[0063] The data acquisition unit: Based on the virtual channel mapping graph, and using the communication parameter acquisition device embedded in the terminal device, acquire the link communication attributes of the communication links in the connection edges, where the link communication attributes include signal strength, transmission rate, and bit error rate; construct a time acquisition sequence.

[0064] Furthermore, the communication quality calculation and rule setting module includes a communication quality calculation unit and a rule setting unit;

[0065] The communication quality calculation unit: Based on the signal strength, transmission rate, and bit error rate of the communication link under the time acquisition sequence, calculate the communication quality of the communication link; based on the communication quality, calculate the comprehensive communication quality of the connection edge under the time acquisition sequence;

[0066] The rule setting unit: Set the communication anomaly judgment rule for the connection edge.

[0067] Furthermore, the matrix construction and analysis warning module includes a matrix construction unit and an analysis warning unit;

[0068] The matrix construction unit: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix, where the rows of the anomaly judgment Boolean matrix are all the communication links in the connection edge, and the columns of the anomaly judgment Boolean matrix are anomaly types, and the anomaly types include signal strength anomaly, transmission rate anomaly, bit error rate anomaly, and multi-attribute anomaly;

[0069] The analysis warning unit: Based on the anomaly judgment Boolean matrix, construct the eigenvector of the connection edge under the time acquisition sequence; use the cosine similarity to calculate the communication anomaly similarity between the eigenvectors of different connection edges under the time acquisition sequence, preset a similarity threshold, if the communication anomaly similarity is greater than or equal to the similarity threshold, then judge that there is a communication anomaly similarity between the connection edges; real-time diagnose whether there is a network communication anomaly and the anomaly cause in the connection edge, obtain all the connection edges with communication anomaly similarity to the connection edge, and perform unified warning.

[0070] Please refer to Figure 2 , in the second embodiment: Provide an intelligent diagnosis method for the communication network of an automatic fuel dispenser, and this method includes the following steps:

[0071] Step S1: Archive and record the terminal devices in the gas station, and construct a virtual channel mapping diagram.

[0072] Specifically, archive and record the terminal devices in the gas station for constructing a virtual channel mapping diagram. The terminal devices include fuel dispensers, network switches, and servers; the fuel dispensers communicate with the servers via at least one of the network switches.

[0073] The method for constructing the virtual channel mapping diagram is as follows:

[0074] Map one terminal device to one graph node, and map the communication links between the fuel dispensers, network switches, and servers to the connection edges between the graph nodes.

[0075] Step S2: Based on the virtual channel mapping diagram, and using the communication parameter acquisition device embedded in the terminal device, collect the link communication attributes of the communication links in the connection edges, and construct a time acquisition sequence.

[0076] Specifically, based on the virtual channel mapping diagram, and using the communication parameter acquisition device embedded in the terminal device, collect the link communication attributes of the communication links in the connection edges. The link communication attributes include signal strength, transmission rate, and error rate; denote the i-th connection edge as , where , represents the j-th communication link in the i-th connection edge, represents the connection edge ;

[0077] Furthermore, construct a time acquisition sequence, denoted as , where represents the t-th time acquisition sequence, T represents the total number of time acquisition sequences, and respectively denote the signal strength, transmission rate, and error rate of the communication link at the time acquisition sequence as and .

[0078] Step S3: Calculate the communication quality of the communication link based on the link communication attributes under the time acquisition sequence; calculate the comprehensive communication quality of the connection edge based on the communication quality, and set the communication anomaly judgment rule for the connection edge.

[0079] Specifically, based on the signal strength , transmission rate , and error rate of the communication link under the time acquisition sequence , calculate the communication quality of the communication link The communication quality is calculated by the following formula:

[0080] ;

[0081] Wherein, represents the communication quality of the communication link ; represents a preset signal strength influence factor, represents a preset transmission rate influence factor, represents a preset bit error rate influence factor, and respectively represent the minimum and maximum values of the signal strength within a preset normal range, and respectively represent the minimum and maximum values of the transmission rate within a preset normal range, and respectively represent the minimum and maximum values of the bit error rate within a preset normal range;

[0082] In the present invention, this formula comprehensively considers the signal strength, transmission rate, and bit error rate, and these three attributes are the key factors for measuring the communication link quality. In the actual communication network of automatic fuel dispensers, the signal strength affects the stability of data transmission, the transmission rate determines the efficiency of data transmission, and the bit error rate reflects the accuracy of data transmission. By incorporating these three attributes into the calculation, the actual quality of the communication link can be comprehensively and accurately reflected. In the gas station environment, if there are interference sources around affecting the signal strength, and at the same time network congestion leads to a decrease in the transmission rate, this formula can comprehensively consider these factors and accurately evaluate the degree of influence on the communication link. Compared with the evaluation method that only relies on a single index, it can more truly reflect the link status.

[0083] Normalize each attribute, mapping the actual value to the interval [0, 1]. This processing method makes the data of different attributes comparable. The numerical ranges of the signal strength, transmission rate, and bit error rate of the equipment in different gas stations may vary greatly due to factors such as model and location. After normalization, these differences can be eliminated, facilitating unified evaluation and comparison. For the links between different fuel dispensers and the server, even if the original data is different, after normalization processing, the communication quality of each link can be measured under the same standard, and its advantages and disadvantages can be judged more accurately.

[0084] By introducing the preset influence factors and , users can flexibly adjust the influence weights of each attribute on the communication quality according to the actual situation. In different gas station scenarios, the importance of each communication attribute may be different. In areas with strong signal interference, the signal strength has a more crucial impact on the communication quality, and Value; in scenarios with high requirements for real-time data transmission, the weight of the transmission rate can be increased accordingly. This flexibility enables the formula to better adapt to diverse actual application scenarios and improve the accuracy and pertinence of the evaluation.

[0085] Further, based on the communication link communication quality , calculate the comprehensive communication quality of the connection edge under the time acquisition sequence , and the calculation formula is as follows:

[0086] ;

[0087] Wherein, represents the comprehensive communication quality of the connection edge under the time acquisition sequence , represents the total number of communication links in the connection edge .

[0088] In the present invention, a connection edge is composed of multiple communication links. By summing and averaging the communication quality of all communication links within the connection edge, this formula can reflect the communication status of the connection edge as a whole. In an actual network, the quality of each link within a connection edge may vary greatly. Focusing only on the abnormal conditions of individual links is likely to overlook the overall problem. By calculating the comprehensive communication quality, it is possible to avoid misjudging the status of the connection edge due to local fluctuations of individual links, more accurately grasp the overall communication stability of the connection edge, and ensure a comprehensive and reliable evaluation of the network connection. Representing the communication quality of the connection edge with a single comprehensive value simplifies the index for evaluating the status of the connection edge. In a large-scale gas station communication network, the number of connection edges is numerous. If complex multi-link individual evaluations are performed on each connection edge, not only is the calculation volume large, but it is also difficult to quickly judge the overall status. The formula for the comprehensive communication quality of the connection edge provides a simple and intuitive index. Combined with a preset threshold, it can quickly determine whether there is an abnormality in the connection edge, improve the efficiency of network anomaly diagnosis, facilitate the timely discovery and handling of potential network problems, and ensure the normal operation of the network. The comprehensive communication quality of the connection edge, as a quantitative index, provides an important basis for subsequent analysis of the cause of anomalies and early warning. When the comprehensive communication quality of the connection edge exceeds the threshold, the system can further deeply analyze the communication quality of each link and, combined with the anomaly judgment rules, quickly locate the cause of the anomaly, such as abnormal signal strength, abnormal transmission rate, or abnormal bit error rate. At the same time, based on the comprehensive communication quality of the connection edge, it is also possible to calculate the similarity of communication anomalies between connection edges, realize early warning and cluster analysis of potential anomalies, and improve the intelligent level of the intelligent diagnosis system for the communication network of automatic fuel dispensers.

[0089] Further, set the communication anomaly judgment rules for the connection edge , which are specifically as follows:

[0090] Preset a comprehensive communication quality threshold. If the connected edge has a comprehensive communication quality greater than or equal to the comprehensive communication quality threshold, it is determined that there is a network communication anomaly in the connected edge and analyze the cause of the anomaly as follows:

[0091] If , it is determined that the cause of the anomaly is the signal strength anomaly of the communication link ;

[0092] If , it is determined that the cause of the anomaly is the transmission rate anomaly of the communication link ;

[0093] If <s , it is determined that the cause of the anomaly is the bit error rate anomaly of the communication link ;

[0094] If there are multiple anomalies in the link communication attributes, it is determined that the connected edge has a network communication anomaly due to multiple attribute anomalies in the communication link .

[0095] Step S4: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix; based on the anomaly judgment Boolean matrix, construct a feature vector of the connected edge under the time acquisition sequence; based on the feature vectors of different connected edges, calculate the communication anomaly similarity between the connected edges, preset a threshold, analyze and give an early warning.

[0096] Specifically, based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix. The rows of the anomaly judgment Boolean matrix are all communication links in the connected edge , and the columns of the anomaly judgment Boolean matrix are anomaly types, where the anomaly types include signal strength anomaly, transmission rate anomaly, bit error rate anomaly, and multiple attribute anomalies;

[0097] If the connected edge has a network communication anomaly due to the m-th anomaly type of the j-th communication link, the matrix element corresponding to the j-th row and the m-th column is recorded as 1, otherwise 0;

[0098] Further, based on the anomaly judgment Boolean matrix, construct a feature vector of the connected edge under the time acquisition sequence , denoted as , where represents the ratio of the number of edges with signal strength anomaly in the connected edge under the time acquisition sequence to the total number of edges in the connected edge , Represents the time acquisition sequence Lower connecting edge The ratio of the number of edges with abnormal transmission rate in to the total number of edges in the connecting edges in Represents the time acquisition sequence Lower connecting edge The ratio of the number of edges with abnormal bit error rate in to the total number of edges in the connecting edges in Represents the time acquisition sequence Lower connecting edge The ratio of the number of edges with multi-attribute abnormality in to the total number of edges in the connecting edges in

[0099] Using cosine similarity, calculate the eigenvector of the time acquisition sequence Lower connecting edge of the eigenvector and the time acquisition sequence Lower connecting edge of the eigenvector The communication anomaly similarity between them, preset the similarity threshold. If the communication anomaly similarity is greater than or equal to the similarity threshold, then judge that the connecting edge and the connecting edge have similar communication anomalies;

[0100] Furthermore, let t = t + 1, and diagnose in real time whether the connecting edge has network communication anomalies and the reasons for the anomalies, obtain all the connecting edges that are similar to the connecting edge with communication anomalies, and issue a unified warning.

[0101] In the present invention, by constructing a Boolean matrix with communication anomaly types as columns and links as rows, extracting the eigenvectors of the connecting edges, using cosine similarity to calculate the similarity of multiple connecting edges in the abnormal pattern, and then implementing clustering warning based on the similarity threshold, the pattern recognition and spatial similarity analysis of communication anomalies are realized.

[0102] Its function is to transform isolated anomaly instances into "abnormal pattern groups", identify abnormal regions or device groups with consistent behaviors, and support more targeted regional maintenance and risk isolation.

[0103] Through the abnormal similarity analysis mechanism, false alarms and missed alarms caused by isolated judgment are effectively avoided, the stability and credibility of the warning system are improved, and maintenance personnel can give priority to dealing with "high-risk similar groups", significantly improving the operation and maintenance response efficiency and management accuracy.

[0104] It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device.

[0105] Finally, it should be noted that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent diagnostic system for the communication network of an automatic fuel dispenser, characterized in that, The system includes: a mapping graph construction module, a data acquisition module, a communication quality calculation and rule setting module, and a matrix construction and analysis warning module; The mapping graph construction module: files and records the terminal devices in the gas station and constructs a virtual channel mapping graph; The data acquisition module: based on the virtual channel mapping graph and using the communication parameter acquisition device embedded in the terminal device, acquires the link communication attributes of the communication links in the connection edges and constructs a time acquisition sequence; The communication quality calculation and rule setting module: calculates the communication quality of the communication links based on the link communication attributes in the time acquisition sequence; calculates the comprehensive communication quality of the connection edges based on the communication quality, and sets the communication anomaly judgment rules for the connection edges; The matrix construction and analysis warning module: constructs an anomaly judgment Boolean matrix based on the communication anomaly judgment rules; constructs the feature vectors of the connection edges in the time acquisition sequence based on the anomaly judgment Boolean matrix; calculates the communication anomaly similarity between the connection edges based on the feature vectors of different connection edges, preset a threshold value, analyzes and issues a warning; The mapping graph construction module includes a mapping graph construction unit; The mapping graph construction unit: files and records the terminal devices in the gas station for constructing a virtual channel mapping graph, and the terminal devices include fuel dispensers, network switches, and servers; the fuel dispensers communicate with the servers via at least one of the network switches; The method for constructing the virtual channel mapping graph is as follows: Maps one terminal device to one graph node, and maps the communication links between the fuel dispensers, network switches, and servers to the connection edges between the graph nodes; The communication quality calculation and rule setting module includes a communication quality calculation unit and a rule setting unit; The communication quality calculation unit: calculates the communication quality of the communication links based on the signal strength, transmission rate, and error rate of the communication links in the time acquisition sequence; calculates the comprehensive communication quality of the connection edges in the time acquisition sequence based on the communication quality; The rule setting unit: sets the communication anomaly judgment rules for the connection edges; The matrix construction and analysis warning module includes a matrix construction unit and an analysis warning unit; The matrix construction unit: constructs an anomaly judgment Boolean matrix based on the communication anomaly judgment rules, where the rows of the anomaly judgment Boolean matrix are all the communication links in the connection edges, and the columns of the anomaly judgment Boolean matrix are the anomaly types, and the anomaly types include signal strength anomaly, transmission rate anomaly, error rate anomaly, and multi-attribute anomaly; The analysis warning unit: constructs the feature vectors of the connection edges in the time acquisition sequence based on the anomaly judgment Boolean matrix; uses cosine similarity to calculate the communication anomaly similarity between the feature vectors of different connection edges in the time acquisition sequence, preset a similarity threshold value, if the communication anomaly similarity is greater than or equal to the similarity threshold value, then it is judged that there is a communication anomaly similarity between the connection edges; real-time diagnoses whether there is a network communication anomaly and the cause of the anomaly for the connection edges, obtains all the connection edges with communication anomaly similarity to the connection edges, and issues a unified warning.

2. The intelligent diagnosis system for the communication network of an automatic fuel dispenser according to claim 1, wherein: The data acquisition module includes a data acquisition unit; The data acquisition unit: Based on the virtual channel mapping graph, and using the communication parameter acquisition device embedded in the terminal device, acquire the link communication attributes of the communication links in the connection edges, where the link communication attributes include signal strength, transmission rate, and bit error rate; construct a time acquisition sequence.

3. An intelligent diagnosis method for the communication network of an automatic fuel dispenser, which executes an intelligent diagnosis system for the communication network of an automatic fuel dispenser as described in any one of claims 1-2, characterized in that, The method includes the following steps: Step S1: File and record the terminal devices in the gas station, and construct a virtual channel mapping graph; Step S2: Based on the virtual channel mapping graph, and using the communication parameter acquisition device embedded in the terminal device, acquire the link communication attributes of the communication links in the connection edges, and construct a time acquisition sequence; Step S3: Calculate the communication quality of the communication link based on the link communication attributes under the time acquisition sequence; based on the communication quality, calculate the comprehensive communication quality of the connection edge, and set the communication anomaly judgment rule for the connection edge; Step S4: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix; based on the anomaly judgment Boolean matrix, construct the eigenvector of the connection edge under the time acquisition sequence; based on the eigenvectors of different connection edges, calculate the communication anomaly similarity between the connection edges, preset a threshold, analyze, and issue a warning.

4. The intelligent diagnosis method for the communication network of an automatic fuel dispenser according to claim 3, wherein The specific implementation process of the said Step S1 includes: File and record the terminal devices in the gas station for constructing a virtual channel mapping graph. The terminal devices include fuel dispensers, network switches, and servers; the fuel dispensers communicate with the servers via at least one of the network switches; The method for constructing the virtual channel mapping graph is as follows: Map one terminal device to one graph node, and map the communication links between the fuel dispensers, network switches, and servers to the connection edges between the graph nodes.

5. The intelligent diagnosis method for the communication network of an automatic fuel dispenser according to claim 4, characterized in that The specific implementation process of the said Step S2 includes: Based on the virtual channel mapping diagram, and by using the communication parameter acquisition device embedded in the terminal device, collect the link communication attributes of the communication links in the connection edge, where the link communication attributes include signal strength, transmission rate, and bit error rate; denote the i-th connection edge as ed i , where ed i = {v i.j |j ∈ [1, J]}, v i.j represents the j-th communication link in the i-th connection edge, and J represents the total number of communication links in the connection edge ed i ; Construct a time acquisition sequence, denoted as Y = {Y t |t ∈ [1, T]}, where Y t represents the t-th time acquisition sequence, and T represents the total number of time acquisition sequences. Respectively, record the signal strength, transmission rate, and bit error rate of the communication link v t under the time acquisition sequence Y i.j as and 6. The intelligent diagnosis method for the communication network of an automatic fuel dispenser according to claim 5, characterized in that, The specific implementation process of the said Step S3 includes: Based on the time acquisition sequence Y t Lower communication link v i.j Signal strength Transmission rate And bit error rate Calculate the communication quality of communication link v i.j The calculation formula is as follows: Among them, represents the communication quality of communication link v i.j , α1 represents a preset signal strength influence factor, α2 represents a preset transmission rate influence factor, α3 represents a preset bit error rate influence factor, SI min and SI max respectively represent the minimum and maximum values of the signal strength within a preset normal range, TS min and TS max respectively represent the minimum and maximum values of the transmission rate within a preset normal range, ER min and ER max respectively represent the minimum and maximum values of the bit error rate within a preset normal range; Based on communication link v i.j Communication quality Calculate time acquisition sequence Y t Lower connection edge ed i The comprehensive communication quality, and the calculation formula is as follows: Among them, represents the time acquisition sequence Y t the lower connection edge ed i the comprehensive communication quality of, J represents the connection edge ed i the total number of communication links in; Set the communication exception judgment rule for the connection edge ed i as follows: Preset a comprehensive communication quality threshold. If the comprehensive communication quality of the connection edge ed i is greater than or equal to the comprehensive communication quality threshold, it is determined that there is a network communication anomaly in the connection edge ed i and the cause of the anomaly is analyzed as follows:​ If it is determined that the cause of the abnormality is that the signal strength of the communication link v i.j is abnormal; If it is determined that the cause of the abnormality is that the transmission rate of the communication link v i.j is abnormally high; If then it is determined that the cause of the anomaly is the abnormal bit error rate of the communication link v i.j ; If there are multiple anomalies in the link communication attributes, then determine the connection edge ed i Due to the communication link v i.j There is a network communication anomaly due to multiple attribute anomalies 7. An intelligent diagnosis method for the communication network of an automatic fuel dispenser according to claim 6, characterized in that, The specific implementation process of the said Step S4 includes: Based on the communication anomaly judgment rule, construct an anomaly judgment Boolean matrix, where the rows of the anomaly judgment Boolean matrix are all communication links in the connection edge ed i in, and the columns of the anomaly judgment Boolean matrix are anomaly types, and the anomaly types include signal strength anomaly, transmission rate anomaly, bit error rate anomaly, and multi-attribute anomaly; If the connection edge is ed i If there is a network communication anomaly caused by the m-th anomaly type of the j-th communication link, the matrix element corresponding to the j-th row and the m-th column is recorded as 1, otherwise it is 0; Construct a time acquisition sequence Y based on the abnormal judgment Boolean matrix t Lower connection edge ed i The eigenvector of, denoted as Wherein Represents the time acquisition sequence Y t Lower connection edge ed i The ratio of the number of edges with abnormal signal strength in to the total number of edges in the connection edge ed i Represents the time acquisition sequence Y t Lower connection edge ed i The ratio of the number of edges with abnormal transmission rate in to the total number of edges in the connection edge ed i Represents the time acquisition sequence Y t Lower connection edge ed i The ratio of the number of edges with abnormal bit error rate in to the total number of edges in the connection edge ed i Represents the time acquisition sequence Y t Lower connection edge ed i The ratio of the number of edges with multi-attribute abnormality in to the total number of edges in the connection edge ed i ​​​​ Calculate the time acquisition sequence Y using cosine similarity t Lower connection edge ed i Feature vector of And the time acquisition sequence Y t Lower connection edge ed i+1 Feature vector of The communication anomaly similarity between them, preset a similarity threshold. If the communication anomaly similarity is greater than or equal to the similarity threshold, then it is determined that the connection edge ed i And the connection edge ed i+1 There is a communication anomaly similarity; Let t = t + 1, and diagnose the connection edge ed in real time i Whether there is network communication anomaly and the reason for the anomaly, and obtain all the connection edges similar to the connection edge ed i with communication anomalies, and issue a unified early warning

Images