Power distribution data channel monitoring method and device

Abstract

The invention discloses a power distribution data channel monitoring method and device, and relates to the field of data transmission monitoring, and the method comprises the steps: obtaining the node operation data of each node in a current power distribution data channel; the node operation data comprises load time sequence change data, node response time, a node internal occupancy rate and a node access log; determining a load state of each node according to the node operation data, and dividing each node into a first type of nodes and a second type of nodes based on a clustering algorithm according to the load state of each node; updating the first type of nodes according to the node operation data of the second type of nodes; according to the node access log, determining the abnormal performance association degree of each node based on an isolated forest algorithm; and according to the abnormal performance association degree of each node, carrying out current limiting on the node of which the abnormal performance association degree exceeds a preset threshold value. According to the application, the problem of monitoring and adjusting distortion caused by lack of comprehensiveness of power distribution data channel monitoring in the prior art can be solved.

Description

Technical Field

[0001] The present application relates to the field of data transmission monitoring, and in particular to a method and device for monitoring a power distribution data channel. Background Art

[0002] In recent years, with the expansion and development of smart grid, more and more power grid acquisition terminals and mobile terminals are connected to the system distribution network as nodes, which brings new challenges to the efficiency, quality and stability of data transmission in the system distribution network. Therefore, it is necessary to monitor and process the transmission of distribution data in the power grid system to prevent the decline of efficiency, quality and stability of data transmission in the system distribution network. Among them, a more effective method is to directly monitor the distribution data channel.

[0003] The existing method of monitoring the power distribution data channel is generally to study the protocol of the corresponding channel and perform monitoring and processing in combination with the content specified in the protocol. For example, the μTESLA broadcast authentication protocol in the wireless protocol is to perform corresponding monitoring and processing by studying the shared key between its base station and node and the delayed authentication duration. However, this monitoring method is usually limited to the corresponding protocol and cannot be extended to all protocols. At the same time, since a single protocol will limit the corresponding monitoring method to a single type or some types of data, and the protocol usually does not have the ability to distinguish other nodes outside the base station, the comprehensiveness of the distribution data channel monitoring is lacking, resulting in distorted monitoring of the distribution data channel. Therefore, how to improve the comprehensiveness of the distribution data channel monitoring to improve the accuracy of the distribution data channel monitoring and adjustment is still one of the problems that need to be solved in the existing technology. Summary of the invention

[0004] The present application provides a method and device for monitoring a power distribution data channel to solve the technical problem that the prior art lacks comprehensiveness in monitoring a power distribution data channel, resulting in distorted monitoring and adjustment.

[0005] According to a first aspect of an embodiment of the present application, a method for monitoring a power distribution data channel is provided, comprising:

[0006] Obtaining node operation data of each node in the current power distribution data channel; wherein the node operation data includes load timing change data, node response time, node internal occupancy rate and node access log;

[0007] Determine the load status of each node according to the node operation data, and divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node; wherein the load rate of the first type of node is higher than that of the second type of node;

[0008] updating the first type of nodes according to the node operation data of the second type of nodes;

[0009] Based on the node access logs and the Isolation Forest algorithm, determine the abnormal performance correlation degree of each node;

[0010] Based on the abnormal performance correlation degree of each node, throttle the nodes whose abnormal performance correlation degree exceeds a preset threshold.

[0011] This application first obtains the node operation data of each node in the current power distribution data channel, and then determines the load status of each node and divides the nodes into the first type of nodes and the second type of nodes based on the clustering algorithm. Compared with the prior art that cannot distinguish other nodes outside the base station and thus has a one-size-fits-all adjustment method, this application classifies the nodes into two categories according to the node operation data, and then adjusts the first type of nodes through the second type of nodes. It can adjust the nodes with relatively poor status according to the nodes with relatively better status in the channel, making the monitoring and adjustment more adaptable to the nodes in the current power distribution data channel, thereby improving the accuracy of power distribution data channel monitoring and adjustment; at the same time, based on the node access logs and the Isolation Forest algorithm, determine the abnormal performance correlation degree of each node, and then throttle the nodes that exceed the threshold, which can further determine the abnormal nodes and take measures after classifying and updating the adjustment of the nodes, thereby improving the accuracy of power distribution data channel monitoring and adjustment.

[0012] In some embodiments of this application, the determining the load status of each node according to the node operation data specifically includes:

[0013] Based on the load time-series change data of each node and the time-series prediction algorithm, obtain the predicted load change trend of each node;

[0014] According to the node response time and the node internal occupancy rate of each node, determine the channel network status of the current power distribution data channel;

[0015] According to the predicted load change trend of each node and the channel network status, determine the load status of each node.

[0016] This application first obtains the predicted load change trend of each node based on the load time-series change data of each node and the time-series prediction algorithm, and then determines the channel network status of the current power distribution data channel according to the node response time and the node internal occupancy rate of each node. It can respectively determine the status of each node and the overall status in the channel, and then accurately determine the load status of each node, thereby providing a necessary condition for subsequent node division based on the clustering algorithm, and further improving the accuracy of power distribution data channel monitoring and adjustment.

[0017] In some embodiments of this application, the dividing the nodes into the first type of nodes and the second type of nodes based on the load status of each node and the clustering algorithm specifically includes:

[0018] Perform temporal feature analysis on the temporal variation data of the load of each node to obtain the load change characteristics of each node;

[0019] Perform temporal feature analysis on the node access logs of each node to obtain the node access characteristics of each node;

[0020] Based on the load change characteristics, node access characteristics, and load status of each node, and based on the clustering algorithm, divide each node into a first type of node and a second type of node.

[0021] In this application, temporal feature analysis is respectively performed on the temporal variation data of the load of each node and the node access logs to obtain the load change characteristics and node access characteristics, and then combined with the load status of each node, and node division is performed based on the clustering algorithm, which can comprehensively consider various data characteristics of each node, and then accurately divide each node, so that the adjustment after node division is more suitable for the nodes in the current channel, improving the accuracy of power distribution data channel monitoring and adjustment.

[0022] In some embodiments of this application, the updating of the first type of nodes according to the node operation data of the second type of nodes specifically includes:

[0023] Obtain the channel setting parameters of each second type of node, and based on the channel setting parameters and node operation data of each second type of node, and based on the clustering algorithm, obtain the first setting parameters;

[0024] Update the channel setting parameters of each first type of node according to the first setting parameters.

[0025] In this application, first obtain the channel setting parameters of each second type of node, and combine its node operation data to obtain the first setting parameters based on the clustering algorithm, which can screen and summarize the parameter settings of the second type of nodes, and then when updating the channel setting parameters of each first type of node according to the first setting parameters, make the monitoring and adjustment of parameter update more suitable for the nodes in the current channel, thereby improving the accuracy of power distribution data channel monitoring and adjustment.

[0026] In some embodiments of this application, the determination of the abnormal performance correlation degree of each node based on the isolation forest algorithm according to the node access logs specifically includes:

[0027] Based on the node access logs, determine the abnormal access probability of each node based on the isolation forest algorithm;

[0028] Based on the abnormal access probability and node operation data of each node, determine the abnormal performance correlation degree of each node.

[0029] This application first determines the abnormal access probability of each node according to the node access log and based on the isolation forest algorithm, and then combines the node operation data to more accurately determine the abnormal performance correlation degree of each node, so that the adjustment of each node according to the abnormal performance correlation degree is more accurate, thereby improving the accuracy of the monitoring and adjustment of the distribution data channel.

[0030] According to the second aspect of the embodiments of this application, a distribution data channel monitoring device is provided, including a data acquisition module, a node division module, a node update module, a correlation determination module, and a node current limiting module;

[0031] The data acquisition module is used to acquire the node operation data of each node in the current distribution data channel; wherein, the node operation data includes load time series change data, node response time, node internal occupancy rate, and node access log;

[0032] The node division module is used to determine the load status of each node according to the node operation data, and based on the load status of each node, divide each node into a first type of node and a second type of node by means of a clustering algorithm; wherein, the load rate of the first type of node is higher than that of the second type of node.

[0033] The node update module is used to update the first type of node according to the node operation data of the second type of node.

[0034] The correlation determination module is used to determine the abnormal performance correlation degree of each node according to the node access log and based on the isolation forest algorithm.

[0035] The node current limiting module is used to limit the current of the nodes whose abnormal performance correlation degree exceeds a preset threshold according to the abnormal performance correlation degree of each node.

[0036] In some embodiments of this application, the node division module includes a node change prediction unit, a channel state determination unit, and a node state determination unit;

[0037] The node change prediction unit is used to obtain the predicted load change trend of each node according to the load time series change data of each node and based on the time series prediction algorithm.

[0038] The channel state determination unit is used to determine the channel network state of the current distribution data channel according to the node response time and node internal occupancy rate of each node.

[0039] The node state determination unit is used to determine the load status of each node according to the predicted load change trend of each node and the channel network state.

[0040] In some embodiments of the present application, the node division module includes a load characteristic analysis unit, an access characteristic analysis unit, and a node clustering and division unit;

[0041] The load characteristic analysis unit is configured to perform time-series characteristic analysis on the load time-series change data of each node to obtain the load change characteristics of each node;

[0042] The access characteristic analysis unit is configured to perform time-series characteristic analysis on the node access logs of each node to obtain the node access characteristics of each node;

[0043] The node clustering and division unit is configured to divide each node into a first type of node and a second type of node based on the load change characteristics, node access characteristics, and load status of each node and based on a clustering algorithm.

[0044] In some embodiments of the present application, the node update module includes a setting parameter determination unit and a first type of node update unit;

[0045] The setting parameter determination unit is configured to obtain the channel setting parameters of each second type of node, and based on the channel setting parameters of each second type of node and the node operation data, obtain a first setting parameter based on a clustering algorithm;

[0046] The first type of node update unit is configured to update the channel setting parameters of each first type of node according to the first setting parameter.

[0047] In some embodiments of the present application, the association determination module includes an abnormal probability determination unit and an abnormal association determination unit;

[0048] The abnormal probability determination unit is configured to determine the abnormal access probability of each node based on the node access logs and based on an isolation forest algorithm;

[0049] The abnormal association determination unit is configured to determine the abnormal performance association degree of each node according to the abnormal access probability of each node and the node operation data.

[0050] This application first obtains the node operation data of each node in the current power distribution data channel, then determines the load status of each node and divides the nodes into the first type of nodes and the second type of nodes based on the clustering algorithm. Compared with the prior art that cannot distinguish other nodes outside the base station and thus has a uniform adjustment method, this application classifies the nodes into two categories according to the node operation data, and then adjusts the first type of nodes through the second type of nodes, which can adjust the nodes with relatively poor status according to the nodes with relatively better status in the channel, making the monitoring and adjustment more adaptable to the nodes in the current power distribution data channel, thereby improving the accuracy of the monitoring and adjustment of the power distribution data channel; at the same time, based on the node access logs and the isolation forest algorithm, the abnormal performance correlation degree of each node is determined, and then the nodes exceeding the threshold are throttled, which can further determine the abnormal nodes and take measures after classifying and updating the adjustment of the nodes, thereby improving the accuracy of the monitoring and adjustment of the power distribution data channel. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 : A flowchart of a power distribution data channel monitoring method shown in some embodiments of this application;

[0052] Figure 2 : A module structure diagram of a power distribution data channel monitoring device shown in some embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0053] The embodiments of this application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by combining the drawings are exemplary and are only used to explain some embodiments of this application, and should not be construed as a limitation on the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments shown in this application belong to the protection scope of this application.

[0054] In the description of this application, it should be understood that the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In the description of this application, unless otherwise specifically defined, the meaning of "a plurality" and "several" is two or more.

[0055] The existing method for monitoring the power distribution data channel is generally to study the protocol corresponding to the channel, and to perform monitoring and processing in combination with the content specified in the protocol. However, this monitoring method usually requires that the channel type matches the protocol type, which means that it is difficult to expand to all protocols. At the same time, the corresponding monitoring method of a single protocol is limited to monitoring a single type or some types of data, and the protocol usually does not have the ability to distinguish other nodes outside the base station, resulting in a lack of comprehensiveness in the monitoring of the power distribution data channel, thereby causing distortion in the monitoring of the power distribution data channel. Therefore, how to improve the comprehensiveness of the monitoring of the power distribution data channel to improve the accuracy of the monitoring and adjustment of the power distribution data channel is still one of the problems that need to be solved urgently in the existing technology.

[0056] Based on the above technical background, please refer to Figure 1 The embodiment of the present application provides a method for monitoring a power distribution data channel, including steps S101 to S105, each of which is as follows:

[0057] Step S101: Obtain node operation data of each node in the current power distribution data channel; wherein the node operation data includes load timing change data, node response time, node internal occupancy rate and node access log.

[0058] Step S102: Determine the load status of each node according to the node operation data, and divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node; wherein the load rate of the first type of node is higher than that of the second type of node.

[0059] In certain embodiments of the present application, determining the load status of each node according to the node operation data specifically includes:

[0060] According to the load time series change data of each node, based on the time series prediction algorithm, the predicted load change trend of each node is obtained;

[0061] Determine the channel network status of the current power distribution data channel according to the node response time and the node internal occupancy rate of each node;

[0062] The load status of each node is determined according to the predicted load change trend of each node and the channel network status.

[0063] In certain embodiments of the present application, the time series prediction algorithm is implemented based on a time series prediction model, and the time series prediction model includes but is not limited to a moving average model, an exponential smoothing model, an XGBoost model, an ARIMA model or a random forest model, and the preferred embodiment is an exponential smoothing model.

[0064] In some embodiments of the present application, determining the channel network status of the current power distribution data channel according to the node response time and the internal occupancy rate of each node specifically includes:

[0065] If the node response time of all nodes is not greater than the preset timeout time, and the internal occupancy rate of all nodes is not greater than the preset occupancy threshold, it is determined that the channel network status of the current power distribution data channel is qualified;

[0066] In other cases, it is determined that the channel network status of the current power distribution data channel is unqualified.

[0067] In some embodiments of the present application, determining the load status of each node according to the predicted load change trend of each node and the channel network status specifically includes: judging each node separately. If the predicted load change trend of the current node is an upward trend and the channel network status is unqualified, it is determined that the load status of the current node is overloaded; in other cases, it is determined that the load status of the current node is normal.

[0068] The present application first obtains the predicted load change trend of each node according to the load time-series change data of each node and based on the time-series prediction algorithm, and then determines the channel network status of the current power distribution data channel according to the node response time and the internal occupancy rate of each node, so as to be able to determine the status of each node and the overall status in the channel respectively, and then accurately determine the load status of each node, thereby providing necessary conditions for subsequent node division based on the clustering algorithm, and further improving the accuracy of power distribution data channel monitoring and regulation.

[0069] In some embodiments of the present application, dividing each node into a first type of node and a second type of node based on the clustering algorithm according to the load status of each node specifically includes:

[0070] Performing time-series feature analysis on the load time-series change data of each node to obtain the load change characteristics of each node;

[0071] Performing time-series feature analysis on the node access logs of each node to obtain the node access characteristics of each node;

[0072] According to the load change characteristics, node access characteristics and load status of each node, based on the clustering algorithm, each node is divided into a first type of node and a second type of node.

[0073] In some embodiments of the present application, the time-series feature analysis is implemented based on a time-series feature extraction model, and the time-series feature extraction model includes but is not limited to an LSTM model and its improved models, an ARIMA model, a model based on the Shapelet algorithm, or a TimesNet model. The preferred implementation is the TimesNet model.

[0074] In some embodiments of the present application, the clustering algorithm includes, but is not limited to, the K-means clustering algorithm, hierarchical clustering algorithm, density-based clustering algorithm (DBSCAN), spectral clustering algorithm, or probability model-based clustering algorithm. The preferred embodiment is the probability model-based clustering algorithm.

[0075] In the present application, the time series feature analysis is respectively performed on the load time series change data and node access logs of each node to obtain the load change feature and node access feature. Then, combined with the load status of each node, node division is performed based on the clustering algorithm, which can comprehensively consider various data features of each node, and then accurately perform node division on each node, so that the adjustment after node division is more adapted to the nodes in the current channel, improving the accuracy of power distribution data channel monitoring and adjustment.

[0076] Step S103: Update the first type of nodes according to the node operation data of the second type of nodes.

[0077] In some embodiments of the present application, the updating of the first type of nodes according to the node operation data of the second type of nodes specifically includes:

[0078] Obtain the channel setting parameters of each second type of node, and based on the channel setting parameters and node operation data of each second type of node, obtain the first setting parameter based on the clustering algorithm;

[0079] Update the channel setting parameters of each first type of node according to the first setting parameter.

[0080] In some embodiments of the present application, the clustering algorithm includes, but is not limited to, the K-means clustering algorithm, hierarchical clustering algorithm, density-based clustering algorithm (DBSCAN), spectral clustering algorithm, or probability model-based clustering algorithm. The preferred embodiment is the K-means clustering algorithm.

[0081] In the present application, first obtain the channel setting parameters of each second type of node, and combined with its node operation data, obtain the first setting parameter based on the clustering algorithm, which can screen and summarize the parameter settings of the second type of nodes. Then, when updating the channel setting parameters of each first type of node according to the first setting parameter, the parameter update monitoring and adjustment is more adapted to the nodes in the current channel, thereby improving the accuracy of power distribution data channel monitoring and adjustment.

[0082] Step S104: Determine the abnormal performance correlation degree of each node based on the isolation forest algorithm according to the node access log.

[0083] Specifically, the abnormal performance correlation degree refers to the probability that there is an association between abnormal access in the node access log of each node and the performance degradation of the current power distribution data channel.

[0084] In some embodiments of the present application, based on the node access logs and the isolation forest algorithm, determining the abnormal performance correlation degree of each node specifically includes:

[0085] Based on the node access logs and the isolation forest algorithm, determining the abnormal access probability of each node;

[0086] Based on the abnormal access probability of each node and the node operation data, determining the abnormal performance correlation degree of each node.

[0087] In some embodiments of the present application, the determining the abnormal access probability of each node based on the node access logs and the isolation forest algorithm specifically is: based on the node access logs and the isolation forest algorithm, obtaining the abnormal access logs of each node; based on the abnormal access logs of each node, determining the abnormal access probability of each node.

[0088] The present application first determines the abnormal access probability of each node according to the node access logs and based on the isolation forest algorithm, and then combines the node operation data, which can more accurately determine the abnormal performance correlation degree of each node, and further makes the adjustment of each node according to the abnormal performance correlation degree more accurate, thereby improving the accuracy of the monitoring and adjustment of the power distribution data channel.

[0089] Step S105: According to the abnormal performance correlation degree of each node, perform current limiting on the nodes whose abnormal performance correlation degree exceeds a preset threshold.

[0090] The present application first obtains the node operation data of each node in the current power distribution data channel, then determines the load status of each node and divides the nodes into the first type of nodes and the second type of nodes based on the clustering algorithm. Compared with the prior art that cannot distinguish other nodes outside the base station and thus has a one-size-fits-all adjustment method, the present application classifies the nodes into two categories according to the node operation data, and then adjusts the first type of nodes through the second type of nodes, which can adjust the nodes with relatively poor status according to the nodes with relatively better status in the channel, making the monitoring and adjustment more suitable for the nodes in the current power distribution data channel, thereby improving the accuracy of the monitoring and adjustment of the power distribution data channel; at the same time, based on the node access logs and the isolation forest algorithm, determining the abnormal performance correlation degree of each node, and then performing current limiting on the nodes exceeding the threshold, can further determine the abnormal nodes and take measures after classifying and updating the adjustment of the nodes, thereby improving the accuracy of the monitoring and adjustment of the power distribution data channel.

[0091] Corresponding to the foregoing method, please refer to Figure 2 , the embodiments of the present application provide a power distribution data channel monitoring device, including a data acquisition module 210, a node division module 220, a node update module 230, a correlation determination module 240, and a node current limiting module 250;

[0092] The data acquisition module 210 is configured to acquire the node operation data of each node in the current power distribution data channel; wherein, the node operation data includes load time series change data, node response time, node internal occupancy rate, and node access log;

[0093] The node division module 220 is configured to determine the load status of each node according to the node operation data, and divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node; wherein, the load rate of the first type of node is higher than that of the second type of node;

[0094] The node update module 230 is configured to update the first type of node according to the node operation data of the second type of node.

[0095] The association determination module 240 is configured to determine the abnormal performance association degree of each node based on the isolation forest algorithm according to the node access log.

[0096] The node current limiting module 250 is configured to limit the current of the node whose abnormal performance association degree exceeds a preset threshold according to the abnormal performance association degree of each node.

[0097] In some embodiments of the present application, the node division module 220 includes a node change prediction unit, a channel state determination unit, and a node state determination unit;

[0098] The node change prediction unit is configured to obtain the predicted load change trend of each node based on a time series prediction algorithm according to the load time series change data of each node.

[0099] The channel state determination unit is configured to determine the channel network state of the current power distribution data channel according to the node response time and the node internal occupancy rate of each node.

[0100] The node state determination unit is configured to determine the load status of each node according to the predicted load change trend of each node and the channel network state.

[0101] In some embodiments of the present application, the node division module 220 includes a load feature analysis unit, an access feature analysis unit, and a node clustering division unit;

[0102] The load feature analysis unit is configured to perform time series feature analysis on the load time series change data of each node to obtain the load change feature of each node.

[0103] The access feature analysis unit is configured to perform time series feature analysis on the node access log of each node to obtain the node access feature of each node.

[0104] The node clustering and partitioning unit is used to divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load change characteristics, node access characteristics, and load status of each node.

[0105] In some embodiments of the present application, the node update module 230 includes a setting parameter determination unit and a first type of node update unit;

[0106] The setting parameter determination unit is used to obtain the channel setting parameters of each second type of node, and based on the channel setting parameters of each second type of node and the node operation data, obtain a first setting parameter based on a clustering algorithm;

[0107] The first type of node update unit is used to update the channel setting parameters of each first type of node according to the first setting parameter.

[0108] In some embodiments of the present application, the association determination module 240 includes an abnormal probability determination unit and an abnormal association determination unit;

[0109] The abnormal probability determination unit is used to determine the abnormal access probability of each node based on the isolation forest algorithm according to the node access log;

[0110] The abnormal association determination unit is used to determine the abnormal performance association degree of each node according to the abnormal access probability of each node and the node operation data.

[0111] The present application first obtains the node operation data of each node in the current power distribution data channel, and then determines the load status of each node and divides the nodes into a first type of node and a second type of node based on a clustering algorithm. Compared with the prior art that cannot distinguish other nodes outside the base station and thus has a uniform adjustment method, the present application classifies the nodes into two categories according to the node operation data, and then adjusts the first type of node through the second type of node, which can adjust the nodes with relatively poor status according to the nodes with relatively better status in the channel, making the monitoring and adjustment more suitable for the nodes in the current power distribution data channel, thereby improving the accuracy of power distribution data channel monitoring and adjustment; at the same time, based on the node access log and the isolation forest algorithm, the abnormal performance association degree of each node is determined, and then the nodes exceeding the threshold are throttled, which can further determine the abnormal nodes and take measures after classifying and updating the adjustment of the nodes, thereby improving the accuracy of power distribution data channel monitoring and adjustment.

[0112] It should be understood that the device provided in the embodiments of the present application corresponds to the foregoing method. A power distribution data channel monitoring device provided in the embodiments of the present application can implement a power distribution data channel monitoring method provided in any one of the embodiments of the present application.

[0113] Adaptively, an embodiment of the present application further provides a computer device and a computer-readable storage medium.

[0114] The computer device includes: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor;

[0115] Wherein, when the processor executes the computer program, a power distribution data channel monitoring method of the present application is implemented.

[0116] The computer-readable storage medium stores multiple instructions, and the instructions are suitable for being loaded by the processor to execute a power distribution data channel monitoring method of the present application.

[0117] The above are partial embodiments of the present application, which further elaborate on the purpose, technical solutions, and beneficial effects of the present application. It should be clear that the above partial embodiments of the present application should not be construed as a limitation of the present application. In particular, for those skilled in the art, any changes, modifications, equivalent replacements, and variations made within the spirit and principle of the present application shall be included within the protection scope of the present application.

Claims

1. A method for monitoring a power distribution data channel, characterized in that: include: Obtaining node operation data of each node in the current power distribution data channel; wherein the node operation data includes load timing change data, node response time, node internal occupancy rate and node access log; Determine the load status of each node according to the node operation data, and divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node; wherein the load rate of the first type of node is higher than that of the second type of node; updating the first type of nodes according to the node operation data of the second type of nodes; According to the node access log, based on the isolation forest algorithm, determine the abnormal performance correlation of each node; According to the abnormal performance correlation of each node, the nodes whose abnormal performance correlation exceeds the preset threshold are subject to current limiting.

2. A method for monitoring a power distribution data channel according to claim 1, characterized in that: Determining the load status of each node according to the node operation data specifically includes: According to the load time series change data of each node, based on the time series prediction algorithm, the predicted load change trend of each node is obtained; Determine the channel network status of the current power distribution data channel according to the node response time and the node internal occupancy rate of each node; The load status of each node is determined according to the predicted load change trend of each node and the channel network status.

3. A power distribution data channel monitoring method according to claim 1, characterized in that: The method of dividing each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node specifically includes: Perform time series characteristic analysis on the load time series variation data of each node to obtain the load variation characteristics of each node; Perform time series feature analysis on the node access logs of each node to obtain the node access features of each node; According to the load change characteristics, node access characteristics and load status of each node, each node is divided into the first type of nodes and the second type of nodes based on the clustering algorithm.

4. A method for monitoring a power distribution data channel according to claim 1, characterized in that: The updating of the first-type nodes according to the node operation data of the second-type nodes specifically includes: Acquire the channel setting parameters of each second-type node, and obtain the first setting parameters based on the clustering algorithm according to the channel setting parameters of each second-type node and the node operation data; According to the first setting parameters, the channel setting parameters of each first-type node are updated.

5. A method for monitoring a power distribution data channel according to claim 1, characterized in that: Determining the abnormal performance correlation of each node based on the node access log and the isolation forest algorithm specifically includes: According to the node access log, based on the isolation forest algorithm, determine the abnormal access probability of each node; According to the abnormal access probability of each node and the node operation data, the abnormal performance correlation of each node is determined.

6. A power distribution data channel monitoring device, characterized in that: It includes a data acquisition module, a node division module, a node update module, an association determination module and a node current limiting module; The data acquisition module is used to acquire the node operation data of each node in the current power distribution data channel; wherein the node operation data includes load timing change data, node response time, node internal occupancy rate and node access log; The node division module is used to determine the load status of each node according to the node operation data, and divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load status of each node; wherein the load rate of the first type of node is higher than that of the second type of node; The node updating module is used to update the first type of nodes according to the node operation data of the second type of nodes; The association determination module is used to determine the abnormal performance association of each node based on the node access log and the isolation forest algorithm; The node current limiting module is used to limit the current of nodes whose abnormal performance correlation exceeds a preset threshold according to the abnormal performance correlation of each node.

7. A power distribution data channel monitoring device according to claim 6, characterized in that: The node division module includes a node change prediction unit, a channel state determination unit and a node state determination unit; The node change prediction unit is used to obtain the predicted load change trend of each node based on the load time series change data of each node and the time series prediction algorithm; The channel state determination unit is used to determine the channel network state of the current power distribution data channel according to the node response time of each node and the node internal occupancy rate; The node status determination unit is used to determine the load status of each node according to the predicted load change trend of each node and the channel network status.

8. A power distribution data channel monitoring device according to claim 6, characterized in that: The node division module includes a load feature analysis unit, an access feature analysis unit and a node cluster division unit; The load characteristic analysis unit is used to perform time series characteristic analysis on the load time series variation data of each node to obtain the load variation characteristics of each node; The access feature analysis unit is used to perform time series feature analysis on the node access log of each node to obtain the node access feature of each node; The node clustering division unit is used to divide each node into a first type of node and a second type of node based on a clustering algorithm according to the load change characteristics, node access characteristics and load status of each node.

9. A power distribution data channel monitoring device according to claim 6, characterized in that: The node update module includes a setting parameter determination unit and a first type node update unit; The setting parameter determination unit is used to obtain the channel setting parameters of each second-type node, and obtain the first setting parameters based on the clustering algorithm according to the channel setting parameters of each second-type node and the node operation data; The first-type node updating unit is used to update the channel setting parameters of each first-type node according to the first setting parameters.

10. A power distribution data channel monitoring device according to claim 6, characterized in that: The association determination module includes an abnormal probability determination unit and an abnormal association determination unit; The abnormal probability determination unit is used to determine the abnormal access probability of each node based on the node access log and the isolation forest algorithm; The abnormal correlation determination unit is used to determine the abnormal performance correlation of each node according to the abnormal access probability of each node and the node operation data.

Images