A method and device for detecting delay of electric power communication network based on fgOTN

Abstract

The present invention relates to the field of network delay detection technology, and specifically to a delay detection method and device for an electric power communication network based on fgOTN. First, a power communication network model is established based on power communication equipment and its link relationship, and any two nodes in the network are selected as the sending end and the receiving end, and the delay when the data is transmitted at both ends is obtained. Secondly, all paths from the sending end to the receiving end are obtained under different traffic conditions based on historical data, and the average delay of all paths under different traffic conditions is obtained. At the same time, the historical data is analyzed based on the constructed delay detection model to obtain the optimal path delay under different traffic. Finally, the current traffic state and its delay in the power communication network are obtained, and the delay is compared with the average delay function and the optimal delay function to determine whether the delay is in a normal range; if the delay exceeds the range, the network abnormal state is predicted by calculating the probability of high burst delay.

Description

Technical Field

[0001] The present invention relates to the technical field of network delay detection, and in particular to a method and device for detecting delay in an electric power communication network based on fgOTN. Background Art

[0002] In modern power communication networks, the reliability and real-time performance of data transmission are key factors in ensuring the efficient and safe operation of power systems. With the construction of smart cities, the reliability requirements for power communication networks based on fine-grained optical transport networks (fgOTN) are increasing, especially in dealing with emergencies and high burst delays. Traditional data transmission technologies can no longer meet the growing demand.

[0003] The phenomenon of high burst delay in the network is mainly manifested as a significant increase or decrease in data transmission delay under certain specific time periods or event triggers, which is usually caused by the following factors: network congestion, data packet loss and link failure. In the power communication network under extreme climate or equipment failure conditions, when the traffic suddenly increases, the network's bandwidth and processing capacity cannot adapt to the demand in time, resulting in queuing of data packets during transmission, thus causing high burst delays, which in turn have a negative impact on the operation and dispatching decisions of power equipment, and even cause safety hazards in extreme cases.

[0004] Therefore, a delay detection method and device for electric power communication network based on fgOTN is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a delay detection method and device for an electric power communication network based on fgOTN. First, a power communication network model is established based on the power communication equipment and the link relationship between them, and the delay is obtained by selecting any two nodes in the network as the sending end and the receiving end. Secondly, all paths from the sending end to the receiving end are obtained under different traffic conditions based on historical data, and the average delay of all paths under different traffic conditions is obtained. At the same time, the historical data is analyzed based on the constructed delay detection model to obtain the optimal path delay under different traffic. Finally, the current traffic state and its delay in the power communication network are obtained, and the delay is compared with the average delay function and the optimal delay function to determine whether the delay is in the normal range; if the delay exceeds the range, the network abnormal state is predicted by calculating the probability of high burst delay.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for detecting delay of a power communication network based on fgOTN, comprising:

[0008] S1. Constructing a power communication network based on power communication equipment and its link relationships , where the nodes in the network Indicates power communication equipment, connected Indicates the link between power equipment, link weight Indicates the length of the link;

[0009] S2. Select any two nodes in the power communication network , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message;

[0010] determining a first delay according to the first data and the second data;

[0011] S3. Obtain the historical data from the sender at different traffic levels To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end The average delay is determined by the delays of all paths;

[0012] According to the average delay, obtaining an average delay function about the average delay under different traffic flows;

[0013] S4. Construct a delay detection model to detect the delay of the nodes under different traffic conditions in historical data. To Node Analyze the path and obtain the slave node To Node The optimal path of

[0014] Obtaining an optimal delay of the optimal path and generating an optimal delay function;

[0015] S5. According to the current flow of the electric power communication network, obtain the regional relationship between the first delay and the average delay function and the optimal delay function under the same flow;

[0016] S6. Based on the regional relationship, the probability of high burst delay occurring in the electric power communication network is obtained.

[0017] Preferably, the first data is the sending end The time when the data message is sent, the second data is the time when the receiving end The time when the data message is received.

[0018] Preferably, the delay detection model is trained by using a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume;

[0019] The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer;

[0020] The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set;

[0021] The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay;

[0022] The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm;

[0023] The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol;

[0024] Find the slave node based on the path delay information and link weight To Node The optimal path.

[0025] Preferably, the specific calculation process of the path delay information is:

[0026] ;

[0027] in, For slave nodes To Node The total delay of the path, For Node To Node The total number of nodes between For Node To Node The distance is the propagation speed of the signal, is the traffic load rate, For the node The average length of the data packets at For the node The bandwidth of the link, For the node The packet loss rate at For the node Delay variance.

[0028] Preferably, the average delay function of the path delay under different traffic conditions is expressed as:

[0029] ;

[0030] in, The flow rate is The average delay under For different traffic states between devices, For slave nodes To Node The number of all paths, For the The path has a flow rate of Transmission delay in the state;

[0031] The optimal delay function of the optimal path is the minimum delay among all paths. The specific calculation formula is:

[0032] ;

[0033] in, The flow rate is The optimal path delay from the sender to the receiver is is the number of all paths from the sender to the receiver.

[0034] Preferably, a first flow state and a first delay in the power communication network are obtained;

[0035] Obtaining a first average delay and a first optimal delay under the first traffic state according to the average delay function and the optimal delay function;

[0036] Constructing a first interval, a second interval, and a third interval according to the first average delay and the first optimal delay;

[0037] The first interval is when the first delay is lower than the optimal delay; the second interval is greater than or equal to the optimal delay and less than or equal to the first average delay; the third interval is greater than the first average delay;

[0038] When the first delay is within the second interval, it is a normal delay;

[0039] When the first delay is in the first interval or the second interval, the probability of high burst delay occurring in the power communication network increases.

[0040] Preferably, the high burst delay probability reflects the possibility of high burst delay occurring under the current traffic state;

[0041] The specific formula for the high burst delay probability is:

[0042] ;

[0043] in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current flow, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

[0044] A fgOTN-based power communication network delay detection device, comprising:

[0045] Network establishment module, used to build a power communication network based on power communication equipment and its link relationship , where the nodes in the network Indicates power communication equipment, connected Indicates the link between power equipment, link weight Indicates the length of the link;

[0046] A delay acquisition module is used to select any two nodes in the power communication network , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message;

[0047] determining a first delay according to the first data and the second data;

[0048] The average delay module is used to obtain the average delay from the sender under different traffic conditions from historical data. To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end The average delay is determined by the delays of all paths;

[0049] According to the average delay, obtaining an average delay function about the average delay under different traffic flows;

[0050] The optimal delay module is used to detect the delay of the slave nodes under different traffic conditions in historical data based on the constructed delay detection model. To Node Analyze the path and obtain the slave node To Node The optimal path of

[0051] Obtaining an optimal delay of the optimal path and generating an optimal delay function;

[0052] A delay area analysis module, used for obtaining the regional relationship between the first delay and the average delay function and the optimal delay function under the same flow rate according to the current flow rate of the electric power communication network;

[0053] The high burst delay occurrence probability module is used to obtain the probability of high burst delay occurring in the power communication network based on the regional relationship.

[0054] Preferably, the delay detection model is trained by using a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume;

[0055] The step of performing the model training on the delay detection model through the historical data set includes:

[0056] Data preprocessing: normalize all data to obtain the first preprocessing set;

[0057] Data set division: the first preprocessed set is divided into a training set and a test set in a ratio of 7:3;

[0058] Training model: inputting the training set into the delay detection model until the model training stop condition is met to obtain an initial delay detection model; the model training stop condition is that the loss of the delay detection model reaches convergence;

[0059] Test model: input the test set into the delay detection model for testing, and optimize the delay detection model according to the test results to obtain the optimal delay detection model;

[0060] The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer;

[0061] The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set;

[0062] The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay;

[0063] The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm;

[0064] The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol;

[0065] Find the slave node based on the path delay information and link weight To Node The optimal path.

[0066] Preferably, the high burst delay probability reflects the possibility of high burst delay occurring under the current traffic state;

[0067] The specific formula for the high burst delay probability is:

[0068] ;

[0069] in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current flow, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] 1. The present invention first constructs a power communication network model, taking into account the physical properties of the links between nodes and key factors such as traffic load and bandwidth utilization in historical data, and realizes path delay analysis between any nodes in the network under different traffic conditions, thereby improving the reliability and stability of the power communication network.

[0072] 2. The present invention obtains the optimal path identification mechanism by combining the physical length of the link, traffic load rate, bandwidth utilization rate and data packet transmission volume through the constructed delay detection model. By deeply analyzing the delay characteristics of the link between nodes in the power communication network, the model optimizes the delay of data transmission. At the same time, through the optimal delay under different traffic conditions, the optimal delay function is generated to cope with network changes under different traffic conditions and enhance the stability of the network.

[0073] 3. The present invention monitors the current traffic of the power communication network in real time, obtains the regional relationship between the first delay and the average delay function and the optimal delay function under the same traffic conditions, and improves the network's emergency response capability to sudden delays. Based on the dual analysis framework of the average delay function and the optimal delay function, this method can accurately evaluate the delay status of the current network, and timely warn of possible network anomalies such as packet loss or network congestion by calculating the probability of high sudden delays. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 A schematic flow chart of a method for detecting delay in a power communication network based on fgOTN provided in an embodiment of the present invention;

[0075] Figure 2 A schematic diagram of the structure of a fgOTN-based power communication network delay detection device provided in an embodiment of the present invention;

[0076] Figure 3 A schematic diagram of the delay detection model structure provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] The present invention provides a method for detecting delay of a power communication network based on fgOTN. The method is applied to a device for detecting delay of a power communication network based on fgOTN. The flowchart of the specific method and system is shown in FIG. Figure 1 and Figure 2 The historical data involved in this invention can be obtained from the National Bureau of Statistics, comprehensive data search engines and power industry data platforms.

[0079] Embodiment 1

[0080] As an embodiment of the present invention, refer to Figure 1 S10 in the embodiment is applied to a network establishment module of a power communication network delay detection device based on fgOTN, wherein the network establishment module is used to construct a power communication network based on power communication equipment and its link relationship , where the nodes in the network Indicates power communication equipment, connected Represents the link between power equipment, link weight Indicates the length of the link. This embodiment collects the site information, connection information between sites, flow information and delay data of power communication;

[0081] Site information includes geographic location, device model, device status and the site's traffic handling capabilities; connection information between sites includes link length, link bandwidth, link delay, link load and physical characteristics of the link; traffic information includes the traffic load rate of each link, the number of data packets transmitted per second, bandwidth utilization and packet loss rate; delay data includes real-time delay and historical delay data.

[0082] Reference Figure 1 S20 in the above description is applied to a delay acquisition module of a delay detection device of a power communication network based on fgOTN, wherein the delay acquisition module is used to select any two nodes in the power communication network. , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message;

[0083] determining a first delay according to the first data and the second data;

[0084] Furthermore, the first data is the sending end The time when the data message is sent, the second data is the time when the receiving end The time when the data message is received.

[0085] In this embodiment, the delay acquisition module calculates the transmission delay between the two ends by recording the data message transmission time of the sending end and the receiving end, which simplifies the delay acquisition process and improves the accuracy of data transmission delay measurement.

[0086] Reference Figure 1 S30 in the example is applied to an average delay module of a delay detection device for a power communication network based on fgOTN, wherein the average delay module is used to obtain the average delay from the sending end under different traffic conditions from historical data. To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end Determine the average delay through the delays of all paths; obtain the average delay function about the average delay under different traffic according to the average delay;

[0087] Reference Figure 1 S40 in the embodiment is applied to an optimal delay module of a delay detection device of a power communication network based on fgOTN, wherein the optimal delay module is used to detect the delay of the slave nodes under different flow rates in the historical data according to the constructed delay detection model. To Node Analyze the path and obtain the slave node To Node The optimal path of the optimal path; obtaining the optimal delay of the optimal path, and generating an optimal delay function;

[0088] Furthermore, the delay detection model is constructed by training a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume;

[0089] The step of performing the model training on the delay detection model through the historical data set includes:

[0090] Data preprocessing: normalize all data to obtain the first preprocessing set;

[0091] Data set division: the first preprocessed set is divided into a training set and a test set in a ratio of 7:3;

[0092] Training model: inputting the training set into the delay detection model until the model training stop condition is met to obtain an initial delay detection model; the model training stop condition is that the loss of the delay detection model reaches convergence;

[0093] Test model: input the test set into the delay detection model for testing, and optimize the delay detection model according to the test results to obtain the optimal delay detection model;

[0094] The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer;

[0095] The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set;

[0096] The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay;

[0097] The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm;

[0098] The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol;

[0099] Find the slave node based on the path delay information and link weight To Node The optimal path.

[0100] In this embodiment, the optimal delay module combines the dynamic routing algorithm and the Dijkstra algorithm to analyze the optimal path between any nodes under different traffic from historical data, and obtain the optimal delay, thereby improving the delay optimization capability of the network. The delay detection model is trained and tested based on historical data, the delay pattern in the historical data is analyzed, the network performance is optimized, and the demand for low-latency and highly reliable communication of the smart grid is met.

[0101] Furthermore, the specific calculation process of the path delay information is as follows:

[0102] ;

[0103] in, For slave nodes To Node The total delay of the path, For Node To Node The total number of nodes between For Node To Node The distance is the propagation speed of the signal, is the traffic load rate (number of packets transmitted per second), For the node The average length of the data packets at For the node The bandwidth of the link, For the node The packet loss rate at For the node Delay variance.

[0104] The total delay of the above path is the sum of the delays between all nodes from the sender to the receiver. The normal delay range in the communication network is usually between 1 millisecond and 200 milliseconds. The data is shown in Table 1;

[0105] Table 1 Path total delay data table

[0106] Distance length (m) Traffic load rate Packet loss rate Delay variance (s) Total path delay (s) 1000 0.5 0.01 0.005 0.025 2000 0.7 0.02 0.002 0.061 1500 0.4 0.05 0.003 0.024 3000 0.6 0.015 0.008 0.088 2500 0.8 0.02 0.006 0.096

[0107] Among the above data, other standard data can be obtained from communication equipment, and traffic load rate, packet loss rate and delay variance are obtained by measuring and calculating historical data;

[0108] In this embodiment, this calculation formula comprehensively considers the physical transmission delay, node processing delay, packet loss rate and delay volatility, so as to achieve accurate calculation of path delay; wherein the physical transmission delay is determined by the distance and propagation speed, the node processing delay is related to the data packet length, bandwidth and traffic load, and the packet rate and delay volatility are measured by the delay variance.

[0109] In this embodiment, this method improves the accuracy and reliability of delay detection in the power communication network by accurately calculating the path delay information. By calculating the calculation model of the total delay between nodes, the transmission delay at the physical level of the node, as well as the impact of link load and network fluctuations on the delay are comprehensively considered. This method also considers the impact of packet loss rate and delay variance, thereby effectively improving the accuracy of delay detection and the efficiency of network performance optimization.

[0110] Furthermore, the average delay function of path delay under different traffic conditions is expressed as:

[0111] ;

[0112] in, The flow rate is The average delay under For different traffic states between devices, For slave nodes To Node The number of all paths, For the The path has a flow rate of Transmission delay in the state;

[0113] The optimal delay function of the optimal path is the minimum delay among all paths. The specific calculation formula is:

[0114] ;

[0115] in, The flow rate is The optimal path delay from the sender to the receiver is is the number of all paths from the sender to the receiver.

[0116] In this embodiment, the formula calculates the average delay and the optimal path under different traffic flows; the average delay function reflects the network performance under the traffic flow, and the optimal delay function is used to optimize the route selection to ensure the transmission efficiency and stability of the network under different load conditions.

[0117] In this embodiment, the delay analysis and optimization capabilities of the power communication network under multiple traffic conditions are improved by calculating the average delay and the optimal delay. The average delay function takes into account the transmission delay of all paths between devices and reflects the overall performance of the network under different traffic conditions. The optimal delay function reduces the delay fluctuation under high traffic conditions by selecting the path with the smallest delay. The combination of the two enables the system to not only analyze the network delay characteristics under different loads, but also dynamically select the optimal path in a complex network environment to ensure the efficiency and stability of data transmission.

[0118] Reference Figure 1 S50 in which S50 is applied to a delay area analysis module of a delay detection device for a power communication network based on fgOTN, wherein the delay area analysis module is used to obtain a regional relationship between the first delay and the average delay function and the optimal delay function under the same flow rate according to the current flow rate of the power communication network;

[0119] Further, obtaining a first flow state and a first delay in the power communication network;

[0120] Obtaining a first average delay and a first optimal delay under the first traffic state according to the average delay function and the optimal delay function;

[0121] Constructing a first interval, a second interval, and a third interval according to the first average delay and the first optimal delay;

[0122] The first interval is when the first delay is lower than the optimal delay; the second interval is greater than or equal to the optimal delay and less than or equal to the first average delay; the third interval is greater than the first average delay;

[0123] When the first delay is within the second interval, it is a normal delay;

[0124] When the first delay is in the first interval or the second interval, the probability of high burst delay occurring in the power communication network increases.

[0125] In this embodiment, the delay area analysis module is introduced to determine whether the delay is within the normal range by comparing the relationship between the first delay and the average delay function and the optimal delay function under the same flow. First, the first delay is obtained through the power communication network flow, and the first average delay and the first optimal delay under the flow state are obtained through the average delay function and the optimal delay function. If the first delay is within the interval between the first average delay and the first optimal delay, it is considered that the network delay is within the normal range; if the first delay exceeds this interval, it indicates that the power communication network has a high probability of burst delay.

[0126] Through this delay area analysis mechanism, this method can dynamically evaluate the delay status under different traffic conditions and provide timely high-delay warnings, which helps to prevent and optimize network performance under high-traffic conditions and improve the stability and controllability of the power communication network.

[0127] Reference Figure 1 S60 in which S60 is applied to a high burst delay occurrence probability module of a power communication network delay detection device based on fgOTN, wherein the high burst delay occurrence probability module is used to obtain the probability of high burst delay occurrence in the power communication network according to the regional relationship;

[0128] Furthermore, the high burst delay probability reflects the possibility of high burst delay occurring under the current traffic state;

[0129] The specific formula for the high burst delay probability is:

[0130] ;

[0131] in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current traffic, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

[0132] Under different traffic, the data delay is different. The larger the traffic, the longer the corresponding delay, as shown in Table 2;

[0133] Table 2 High burst delay probability

[0134] Measured delay (ms) Average latency (ms) Optimal delay (ms) Adjustment coefficient Measured delay (ms) High burst delay probability 75 90 80 0.5 10 0.7732 85 90 80 0.5 10 0.2085 95 90 80 0.5 10 0.6915 115 130 110 0.5 12 0.2743 140 130 110 0.5 12 0.7475

[0135] In the above table, the measured delay is the delay under the current traffic, and the adjustment coefficient is fixed at 0.5 to reflect the importance of the measured delay compared with the optimal delay.

[0136] In this embodiment, by introducing a high burst delay probability model, the probability of high burst delay risk in the power communication network is evaluated. This method quantifies the degree to which the current delay deviates from the optimal state by comparing the current delay with the optimal delay and average delay under the same traffic state in historical data, thereby predicting the possibility of high burst delay in the network. When the delay is significantly low, it may cause problems such as packet loss, network congestion recovery, or router cache clearing, increasing network instability; when the delay is high, the probability of burst delay increases, indicating that performance fluctuations may occur in the network.

[0137] The present invention first constructs a network establishment module for the power communication network by combining key parameters such as the geographical location, equipment characteristics and link performance of each node. Secondly, the delay acquisition module improves the reliability and accuracy of data measurement by calculating the transmission delay between nodes. Then, the average delay function and the optimal delay function are obtained through the transmission delay between nodes, which effectively alleviates the delay fluctuation under high traffic conditions and improves the transmission efficiency and stability of the network. In addition, the combination of the delay area analysis module and the high burst delay probability module provides an effective mechanism for real-time monitoring of the delay status and warning of potential problems, greatly enhancing the ability of the power communication network to respond to emergencies.

[0138] Embodiment 2

[0139] A fgOTN-based power communication network delay detection device, comprising:

[0140] Network establishment module, used to build a power communication network based on power communication equipment and its link relationship , where the nodes in the network Indicates power communication equipment, connected Indicates the link between power equipment, link weight Indicates the length of the link;

[0141] A delay acquisition module is used to select any two nodes in the power communication network , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message;

[0142] determining a first delay according to the first data and the second data;

[0143] The average delay module is used to obtain the average delay from the sender under different traffic conditions from historical data. To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end Determine the average delay through the delays of all paths; obtain the average delay function about the average delay under different traffic according to the average delay;

[0144] The optimal delay module is used to detect the delay of the slave nodes under different traffic conditions in historical data based on the constructed delay detection model. To Node Analyze the path and obtain the slave node To Node The optimal path of

[0145] Obtaining an optimal delay of the optimal path and generating an optimal delay function;

[0146] A delay area analysis module, used for obtaining the regional relationship between the first delay and the average delay function and the optimal delay function under the same flow rate according to the current flow rate of the electric power communication network;

[0147] The high burst delay occurrence probability module is used to obtain the probability of high burst delay occurring in the power communication network based on the regional relationship.

[0148] Furthermore, the delay detection model is constructed by training a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume;

[0149] The step of performing the model training on the delay detection model through the historical data set includes:

[0150] Data preprocessing: normalize all data to obtain the first preprocessing set;

[0151] Data set division: the first preprocessed set is divided into a training set and a test set in a ratio of 7:3;

[0152] Training model: inputting the training set into the delay detection model until the model training stop condition is met to obtain an initial delay detection model; the model training stop condition is that the loss of the delay detection model reaches convergence;

[0153] Test model: input the test set into the delay detection model for testing, and optimize the delay detection model according to the test results to obtain the optimal delay detection model;

[0154] The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer;

[0155] The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set;

[0156] The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay;

[0157] The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm;

[0158] The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol;

[0159] Find the slave node based on the path delay information and link weight To Node The optimal path.

[0160] Furthermore, the specific formula of the high burst delay probability is:

[0161] ;

[0162] in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current flow, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

[0163] The fgOTN-based power communication network delay detection device of the present invention has demonstrated its remarkable beneficial effects in many aspects. First, the network establishment module provides a comprehensive data basis for delay analysis by integrating the geographical location, equipment status and link characteristics of the power communication equipment. Secondly, the delay acquisition module effectively simplifies the delay measurement process by recording the transmission time of the data packet, and improves the calculation accuracy of the transmission delay between nodes. In addition, the combination of the average delay module and the optimal delay module makes the evaluation of network performance under different traffic conditions more accurate. At the same time, the integration of the delay area analysis module and the high burst delay probability module enhances the power communication network's ability to respond to emergencies. These comprehensive functions not only meet the needs of smart grids for low latency and high reliability, but also promote technological progress and application development in the field of power communications.

[0164] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A delay detection method for power communication network based on fgOTN, The invention comprises: S1. Constructing a power communication network based on power communication equipment and its link relationships , where the nodes in the network Indicates power communication equipment, connected Represents the link between power equipment, link weight Indicates the length of the link; S2. Select any two nodes in the power communication network , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message; determining a first delay according to the first data and the second data; S3. Obtain the historical data from the sender at different traffic levels To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end The average delay is determined by the delays of all paths; According to the average delay, obtaining an average delay function about the average delay under different traffic flows; S4. Construct a delay detection model to detect the delay of the nodes under different traffic conditions in historical data. To Node Analyze the path and obtain the slave node To Node The optimal path of Obtaining an optimal delay of the optimal path and generating an optimal delay function; S5. According to the current flow of the electric power communication network, obtain the regional relationship between the first delay and the average delay function and the optimal delay function under the same flow; S6. Based on the regional relationship, the probability of high burst delay occurring in the electric power communication network is obtained.

2. According to claim 1, a method for detecting delay in a power communication network based on fgOTN is characterized in that: The first data is the sending end The time when the data message is sent, the second data is the time when the receiving end The time when the data message is received.

3. The method for delay detection of a power communication network based on fgOTN according to claim 1, characterized in that: The delay detection model is trained by using a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume; The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer; The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set; The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay; The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm; The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol; Find the slave node based on the path delay information and link weight To Node The optimal path.

4. The method for delay detection of a power communication network based on fgOTN according to claim 3 is characterized in that: The specific calculation process of the path delay information is as follows: ； in, For slave nodes To Node The total delay of the path, For Node To Node The total number of nodes between For Node To Node The distance is the propagation speed of the signal, is the traffic load rate, For the node The average length of the data packets at For the node The bandwidth of the link, For the node The packet loss rate at For the node Delay variance.

5. The method for delay detection of a power communication network based on fgOTN according to claim 1, characterized in that: The average delay function of path delay based on different traffic conditions is expressed as: ； in, The flow rate is The average delay under For different traffic states between devices, For slave nodes To Node The number of all paths, For the The path has a flow rate of Transmission delay in the state; The optimal delay function of the optimal path is the minimum delay among all paths. The specific calculation formula is: ； in, The flow rate is The optimal path delay from the sender to the receiver is is the number of all paths from the sender to the receiver.

6. The method for delay detection of a power communication network based on fgOTN according to claim 1, characterized in that: Acquire a first flow state and a first delay in the electric power communication network; Obtaining a first average delay and a first optimal delay under the first traffic state according to the average delay function and the optimal delay function; Constructing a first interval, a second interval, and a third interval according to the first average delay and the first optimal delay; The first interval is when the first delay is lower than the optimal delay; the second interval is greater than or equal to the optimal delay and less than or equal to the first average delay; the third interval is greater than the first average delay; When the first delay is within the second interval, it is a normal delay; When the first delay is in the first interval or the second interval, the probability of high burst delay occurring in the power communication network increases.

7. The method for detecting delay of a power communication network based on fgOTN according to claim 6 is characterized in that: The high burst delay probability reflects the possibility of high burst delay occurring under the current traffic state; The specific formula for the high burst delay probability is: ； in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current traffic, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

8. A delay detection device for a power communication network based on fgOTN, characterized in that: include: Network establishment module, used to build a power communication network based on power communication equipment and its link relationship , where the nodes in the network Indicates power communication equipment, connected Represents the link between power equipment, link weight Indicates the length of the link; A delay acquisition module is used to select any two nodes in the power communication network , For the sender, For the receiving end; from the sending end To the receiving end Transmitting a data message, and recording the first data and the second data in the data message; determining a first delay according to the first data and the second data; The average delay module is used to obtain the average delay from the sender under different traffic conditions from historical data. To the receiving end All paths of the data packet are recorded, and the transmission delay of the data packet is recorded; according to the message from the sending end Transmit to the receiving end The average delay is determined by the delays of all paths; According to the average delay, obtaining an average delay function about the average delay under different traffic flows; The optimal delay module is used to detect the delay of the slave nodes under different traffic conditions in historical data based on the constructed delay detection model. To Node Analyze the path and obtain the slave node To Node The optimal path of Obtaining an optimal delay of the optimal path and generating an optimal delay function; A delay area analysis module, used for obtaining the regional relationship between the first delay and the average delay function and the optimal delay function under the same flow rate according to the current flow rate of the electric power communication network; The high burst delay occurrence probability module is used to obtain the probability of high burst delay occurring in the power communication network based on the regional relationship.

9. The fgOTN-based power communication network delay detection device according to claim 8, characterized in that: The delay detection model is trained by using a historical data set; the data in the historical data set includes the physical length of the link, the flow load rate, the bandwidth utilization rate and the data packet transmission volume; The step of training the delay detection model using the historical data set includes: Data preprocessing: normalize all data to obtain the first preprocessing set; Data set division: the first preprocessed set is divided into a training set and a test set in a ratio of 7:3; Training model: inputting the training set into the delay detection model until the model training stop condition is met to obtain an initial delay detection model; the model training stop condition is that the loss of the delay detection model reaches convergence; Test model: input the test set into the delay detection model for testing, and optimize the delay detection model according to the test results to obtain the optimal delay detection model; The delay detection model includes a preprocessing layer, a feature analysis layer and a path delay calculation layer; The preprocessing layer is used to perform data cleaning, standardization, normalization and missing value processing on the historical data set; The feature analysis layer is used to analyze the preprocessed data to obtain a first feature vector related to network delay; The path delay calculation layer includes a dynamic routing algorithm and a Dijkstra algorithm; The dynamic routing algorithm is used to collect path delay information of each link through a routing information protocol; Find the slave node based on the path delay information and link weight To Node The optimal path.

10. The fgOTN-based power communication network delay detection device according to claim 8, characterized in that: The high burst delay probability reflects the possibility of high burst delay occurring under the current traffic state; The specific formula for the high burst delay probability is: ； in, is a high burst delay probability, is the current traffic status, For delay The cumulative distribution function of is the delay under the current flow, is the average delay, is the adjustment coefficient, is the optimal delay, is the delay standard deviation.

Images