Intelligent management method and system based on private network communication

Abstract

The invention specifically relates to an intelligent management method and system based on private network communication. The system comprises a data collection module; a data analysis and judgment module; the node analysis module is used for analyzing the internal temperature of the node switch to obtain a temperature anomaly coefficient; and a switch fault determination module. According to the method, multi-dimensional network data such as round-trip time, packet loss rate and throughput are obtained, and the operation state of the network can be accurately reflected through preprocessing and deep analysis; the method comprises the following steps of: performing detailed temperature monitoring on different hardware areas in the switch, analyzing related parameters to obtain a temperature abnormal coefficient, and combining threshold comparison of a switch fault judgment module to accurately judge whether the switch hardware has a fault or not; compared with traditional simple temperature monitoring, the accuracy and timeliness of fault diagnosis are improved, potential faults can be found in advance, and the probability of network faults is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of private network communication, and particularly to an intelligent management method and system based on private network communication. Background Art

[0002] With the wide application of private network communication technology in various fields, the stability and reliability of the network are becoming increasingly important. In a private network communication network, as a core device, the normal operation of the switch directly affects the data transmission quality and efficiency of the entire network.

[0003] However, there are many deficiencies in the management and monitoring means of the existing private network communication network.

[0004] On the one hand, most traditional network monitoring only focuses on single or a few network performance indicators, such as only monitoring network throughput or packet loss rate, and cannot comprehensively reflect the real operating conditions of the network. When complex problems occur in the network, it is difficult to accurately judge the root cause of the problem. On the other hand, for the fault diagnosis of switch hardware, the existing methods often rely on manual inspection or simple temperature monitoring, and cannot timely and accurately detect potential faults, resulting in the inability to quickly locate and solve problems when network faults occur, seriously affecting the normal operation of the private network communication network.

[0005] Therefore, there is an urgent need for an intelligent management method and system that can comprehensively monitor network performance and accurately diagnose switch hardware faults to improve the management level and reliability of the private network communication network. Summary of the Invention

[0006] The purpose of the present invention is to propose an intelligent management method and system based on private network communication in order to solve the above problems.

[0007] In order to achieve the above purpose, the present invention adopts the following technical solutions: An intelligent management system based on private network communication, comprising: A data collection module: obtaining relevant network data of each node; A data analysis and judgment module: analyzing the relevant network data of the node to obtain a trigger evaluation coefficient, and triggering the analysis of the node switch based on the trigger evaluation coefficient; A node analysis module: analyzing the internal temperature of the node switch to obtain a temperature anomaly coefficient; A switch fault judgment module: judging whether there is a fault in the switch hardware based on the temperature anomaly coefficient.

[0008] Preferably, the data collection module specifically includes: The area is divided according to the working area corresponding to the subnet switch of each node, and the round-trip time data, packet loss rate data and throughput data of each node are obtained, and the data are pre-processed.

[0009] Preferably, the data analysis module specifically includes: After analyzing the round-trip time data, packet loss rate data and throughput data of the node, the round-trip delay value, the high packet loss value and the throughput deviation value are obtained, and the trigger evaluation coefficient is obtained after comprehensive analysis of the round-trip delay value, the high packet loss value and the throughput deviation value; A trigger evaluation coefficient threshold is preset, and the trigger evaluation coefficient is compared with the trigger evaluation coefficient threshold. If the trigger evaluation coefficient is greater than the trigger evaluation coefficient threshold, an analysis of the node hardware is triggered.

[0010] Preferably, the round-trip time data, packet loss rate data and throughput data of the node are analyzed to obtain the round-trip delay value, the high packet loss value and the throughput deviation value, and the specific process includes: The data round trip time of the nodes is obtained at preset time intervals, and the obtained round trip time is arranged in time series; a round trip time threshold is preset, and the round trip time threshold is subtracted from each round trip time obtained in turn, the obtained round trip time difference values are arranged according to the numerical value, and the round trip time difference values less than 0 are eliminated; the round trip time difference values greater than 0 are counted and divided by the total number of round trip time differences to obtain the time deviation degree; Extract the maximum round-trip time difference from the round-trip time differences greater than 0, obtain the duration corresponding to the maximum round-trip time difference, and multiply the maximum round-trip time difference by the duration corresponding to it to obtain the single maximum delay; The round-trip delay value is obtained by weighting the time deviation and the single maximum delay; During the preset test time period, the sender sends a specified number of test data packets according to the configuration requirements, and the receiver receives the data packets synchronously; after the test is completed, the difference between the total number of sent data packets and the actual number of received data packets is calculated, and the difference is divided by the total number of sent data packets to obtain the packet loss rate during network transmission; Based on the above content, a preset number of packet loss tests are performed to obtain a corresponding number of packet loss rates; Preset an allowable fluctuation range of the packet loss rate, match each packet loss rate with the allowable fluctuation range of the packet loss rate in turn, and record the packet loss rate that is not within the allowable fluctuation range of the packet loss rate as an abnormal packet loss rate; Arrange the abnormal packet loss rates in descending order from left to right according to their numerical values, and mark the three largest abnormal packet loss rates; obtain the time intervals between the three largest abnormal packet loss rates, and mark them as the first time interval, the second time interval, and the third time interval, respectively; An elliptical model is established with the first time interval and the second time interval as the major semi-axis and the minor semi-axis of the ellipse respectively, and the third time interval is used as the height of the elliptical model to establish an ellipsoid. The volume of the ellipsoid is calculated and denoted as the packet loss height value.

[0011] Preferably, the process of obtaining the throughput deviation value includes: Obtain the amount of data successfully transmitted by the network within a preset duration; divide the successfully transmitted data amount by the preset duration to obtain the throughput value, and perform multiple tests to obtain a corresponding number of throughput values; Arrange each throughput value according to the time series; Obtain the standard throughput value based on the link nominal bandwidth, and subtract the standard throughput value from each throughput value in turn to obtain the throughput difference; eliminate the throughput differences greater than 0, take the absolute value of the throughput differences less than 0 and denote them as throughput difference pair values; Calculate the mean value of each throughput difference pair value to obtain the throughput difference pair mean; Calculate the difference between the maximum throughput difference pair value and the minimum throughput difference pair value to obtain the throughput difference pair extreme value; Perform weighted calculation on the throughput difference pair mean and the throughput difference pair extreme value to obtain the throughput deviation value.

[0012] Preferably, the process of obtaining the temperature anomaly coefficient includes: Divide the inside of the switch into regions according to the positions of each hardware to obtain the regions corresponding to each hardware, and denote them as hardware regions; The hardware includes: switching chips, CPUs, memory chips; Install temperature sensors at preset positions in the hardware regions to collect the temperatures at each position, preset the maximum allowable temperature of the hardware, compare the temperature at each position with the maximum allowable temperature of the hardware, denote the temperature greater than the maximum allowable temperature of the hardware as the abnormal point temperature, and extract the maximum abnormal point temperature among them; Obtain the maximum abnormal point temperature of each hardware and the position corresponding to the maximum abnormal point temperature of each hardware according to the above steps in turn; Take the positions corresponding to the maximum abnormal point temperature of each hardware as the centers of circles, and connect the centers of circles with straight lines to obtain a triangular figure; Extract the minimum abnormal point temperature from the temperatures greater than the maximum allowable temperature of the hardware, and obtain the triangular figure constructed by the minimum abnormal point temperature according to the above content; Obtain the overlapping area of the triangle constructed by the maximum abnormal point temperature of each hardware and the triangle constructed by the minimum abnormal point temperature of each hardware, and denote it as the stacking area value; Calculate the difference between each maximum abnormal point temperature and the maximum allowable temperature of the hardware to obtain the point temperature difference value; Again, take the position corresponding to the maximum abnormal point temperature as the center of the circle, and the respective point temperature difference values as the radii to obtain three circles. Moreover, the three circles respectively cut the triangular figure to obtain three line segments; Take the two ends of the three line segments as connection points respectively, and connect the connection points of the three line segments in sequence to obtain a closed hexagon; Obtain the center of the hexagon and the temperature detected at the center, and denote it as the center temperature. Subtract the preset maximum allowable hardware temperature from the center temperature to obtain the center temperature difference, and denote the center temperature difference greater than 0 as the center abnormal temperature difference; Take the center of the hexagon as the center of the circle, and use the center abnormal temperature difference as the radius to draw a circle. Divide the area of the circle by the area of the hexagon to obtain the occupancy ratio; After comprehensively processing the stack face value, the center abnormal temperature difference, and the occupancy ratio, obtain the temperature anomaly coefficient.

[0013] Preferably, after normalizing the stack face value, the center abnormal temperature difference, and the occupancy ratio, take the stack face value and the center abnormal temperature difference as the two right-angled sides of a triangle respectively, connect the remaining side to form a complete right-angled triangle, take the occupancy ratio as the height of the right-angled triangle, construct a triangular pyramid model, calculate the volume of the triangular pyramid model, and denote it as the temperature anomaly coefficient.

[0014] Preferably, judging whether there is a fault in the switch hardware based on the temperature anomaly coefficient specifically includes: Preset a temperature anomaly coefficient threshold, compare the temperature anomaly coefficient with the temperature anomaly coefficient threshold. If the temperature anomaly coefficient is greater than the temperature anomaly coefficient threshold, it is determined that there is a fault in the switch hardware.

[0015] An intelligent management method based on private network communication includes: Data collection: Obtain the round-trip time data, packet loss rate data, and throughput data of each node; Data analysis block: After comprehensively analyzing the round-trip time data, packet loss rate data, and throughput data of the node, obtain a trigger evaluation coefficient, and trigger the analysis of the node switch based on the trigger evaluation coefficient; Node analysis: Analyze the internal temperature of the node switch to obtain the temperature anomaly coefficient; Switch hardware diagnosis: Judge whether there is a hardware fault in the switch hardware based on the temperature anomaly coefficient.

[0016] In summary, due to adopting the above technical solutions, the beneficial effects of the present invention are: 1. The present invention can accurately reflect the operating state of the network by obtaining multi-dimensional network data such as round-trip time, packet loss rate, and throughput, and through preprocessing and in-depth analysis. It also conducts detailed temperature monitoring on different hardware areas inside the switch, analyzes relevant parameters to obtain the temperature anomaly coefficient, and combines the threshold comparison of the switch fault judgment module to accurately determine whether there is a fault in the switch hardware. Compared with traditional simple temperature monitoring, it improves the accuracy and timeliness of fault diagnosis, helps to detect potential faults in advance, and reduces the probability of network faults occurring.

[0017] 2. The present invention realizes the intelligent management of the private network communication network based on the trigger evaluation coefficient and the temperature anomaly coefficient. When the network performance is abnormal, it triggers the analysis of the node switch in a timely manner. When it is determined that there is a fault in the switch hardware, it can quickly locate the problem, provide accurate decision-making support for network administrators, facilitate timely measures for maintenance and optimization, improve the stability and reliability of the network, and reduce the network operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In the following description of exemplary embodiments in conjunction with the drawings, more details, features, and advantages of the present application are disclosed. In the drawings: Figure 1 is a flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0019] The following will describe several embodiments of the present application in more detail with reference to the drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and for many different purposes and should not be limited to the embodiments set forth herein. These embodiments are provided so that this application will be thorough and complete and will fully convey the scope of the application to those skilled in the art. The embodiments do not limit the present application.

[0020] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0021] Please refer to Figure 1 as shown, the present invention provides a technical solution: An intelligent management system based on private network communication, comprising: A data collection module: obtaining relevant network data of each node; Specifically including: The area is divided according to the working area corresponding to the subnet switch of each node, and the round-trip time data, packet loss rate data and throughput data of each node are obtained, and the data is pre-processed; Preprocessing includes data cleaning, denoising, normalization and other operations, which helps to improve data quality and provide a more accurate data basis for subsequent analysis and processing; Data analysis and judgment module: analyzes the relevant network data of the node to obtain the trigger evaluation coefficient, and triggers the analysis of the node switch based on the trigger evaluation coefficient; Specifically include: After analyzing the round-trip time data, packet loss rate data and throughput data of the node, the round-trip delay value, the high packet loss value and the throughput deviation value are obtained; The specific process includes: The data round trip time of the nodes is obtained at preset time intervals, and the obtained round trip time is arranged in time series; a round trip time threshold is preset, and the round trip time threshold is subtracted from each round trip time obtained in turn, the obtained round trip time difference values are arranged according to the numerical value, and the round trip time difference values less than 0 are eliminated; the round trip time difference values greater than 0 are counted and divided by the total number of round trip time differences to obtain the time deviation degree; Extract the maximum round-trip time difference from the round-trip time differences greater than 0, obtain the duration corresponding to the maximum round-trip time difference, and multiply the maximum round-trip time difference by the duration corresponding to it to obtain the single maximum delay; The round-trip delay value is obtained by weighting the time deviation and the single maximum delay; Preset the weight factors of the time deviation and the single maximum delay, respectively multiply the time deviation and the single maximum delay with their corresponding weight factors and then sum them to obtain the round-trip delay value; The maximum single delay can reflect the degree of influence of the maximum delay in the network on the overall network performance during the observation period. Specifically, the larger the value, the longer the network has experienced a large delay during this period, which may cause serious lag in data transmission and affect the response speed and stability of various network applications. For example, it may cause video playback to freeze, online games to have obvious delays or even disconnection, indicating that the network condition may be poor, with large room for optimization or potential fault hazards. On the contrary, a smaller value means that the network delay is relatively good and the network performance is relatively stable. During the preset test time period, the sender sends a specified number of test data packets according to the configuration requirements, and the receiver receives the data packets synchronously; after the test is completed, the difference between the total number of sent data packets and the actual number of received data packets is calculated, and the difference is divided by the total number of sent data packets to obtain the packet loss rate during network transmission; Based on the above content, perform a preset number of packet loss tests to obtain the corresponding packet loss rates; Preset the allowable fluctuation range of the packet loss rate, and successively match each packet loss rate with the allowable fluctuation range of the packet loss rate. Mark the packet loss rates that are not within the allowable fluctuation range of the packet loss rate as abnormal packet loss rates; Arrange the abnormal packet loss rates in descending order from left to right according to their numerical values, and mark the three largest abnormal packet loss rates; Obtain the time intervals between the three largest abnormal packet loss rates, and mark them as the first time interval, the second time interval, and the third time interval respectively; Among them, the first time interval is the time interval between the leftmost abnormal packet loss rate and the middle abnormal packet loss rate, the second time interval is the time interval between the middle abnormal packet loss rate and the rightmost abnormal packet loss rate, and the third time interval is the time interval between the leftmost abnormal packet loss rate and the rightmost abnormal packet loss rate; Establish an ellipse model with the first time interval and the second time interval as the major semi-axis and the minor semi-axis of the ellipse respectively, use the third time interval as the height of the ellipse model to establish an ellipsoid, and calculate the volume of the ellipsoid, which is denoted as the packet loss height value; Obtain the amount of data successfully transmitted by the network within a preset duration; Divide the successfully transmitted data amount by the preset duration to obtain the throughput value, and perform multiple tests to obtain the corresponding number of throughput values; Arrange each throughput value according to the time series; Obtain the standard throughput value based on the link nominal bandwidth, and successively subtract the standard throughput value from each throughput value to obtain the throughput difference; Eliminate the throughput differences greater than 0, and take the absolute value of the throughput differences less than 0 and denote them as throughput difference pair values; Calculate the mean value of each throughput difference pair value to obtain the throughput difference pair mean; Calculate the difference between the largest throughput difference pair value and the smallest throughput difference pair value to obtain the throughput difference pair extreme value; Perform weighted calculation on the throughput difference pair mean and the throughput difference pair extreme value to obtain the throughput deviation value; Preset the weight factors of the throughput difference pair mean and the throughput difference pair extreme value. Multiply the throughput difference pair mean and the throughput difference pair extreme value by their corresponding weight factors respectively, and then sum them to obtain the throughput deviation value; Perform weighted calculation on the throughput difference pair mean and the throughput difference pair extreme value to obtain the throughput deviation value. The weighted calculation can allocate different weights to the throughput difference pair mean and the throughput difference pair extreme value according to actual needs to comprehensively reflect the situation of the throughput deviating from the nominal bandwidth; Preset the weight factors of the round-trip delay value, the packet loss height value, and the throughput deviation value. Multiply the round-trip delay value, the packet loss height value, and the throughput deviation value by their corresponding weight factors respectively, and then sum them to obtain the trigger evaluation coefficient; Preset a trigger evaluation coefficient threshold, compare the trigger evaluation coefficient with the trigger evaluation coefficient threshold, and if the trigger evaluation coefficient is greater than the trigger evaluation coefficient threshold, trigger an analysis of the node hardware; Node analysis module: analyzes the internal temperature of the node switch to obtain the temperature anomaly coefficient; The acquisition process includes: Divide the interior of the switch into regions according to the locations of various hardware to obtain regions corresponding to various hardware and record them as hardware regions; The hardware includes: switch chip, CPU, memory chip; The switch is divided into zones according to the hardware location, and the corresponding zones of each hardware are clearly defined. This is conducive to accurately locating and monitoring the operating status of each hardware, providing a basis for subsequent targeted collection of temperature data, so that temperature monitoring has a clear goal and scope; Temperature sensors are placed at preset locations in the hardware area to collect temperature information, so that the temperature conditions at different locations of each hardware can be obtained in real time. By presetting the maximum allowable temperature of the hardware and comparing the actual collected temperature with it, temperature anomalies can be discovered in a timely manner. Temperatures greater than the maximum allowable temperature are recorded as abnormal point temperatures, and the maximum abnormal point temperature is extracted. This allows you to focus on the abnormal location with the highest temperature on the hardware, which helps to quickly locate key locations that may have failures or overheating problems, so that timely measures can be taken for maintenance or adjustment to ensure the stable operation of the switch; Temperature sensors are placed at preset locations in the hardware area to collect the temperature of each location, preset the maximum allowable temperature of the hardware, compare the temperature of each location with the maximum allowable temperature of the hardware, record the temperature greater than the maximum allowable temperature of the hardware as an abnormal point temperature, and extract the maximum abnormal point temperature; According to the above steps, the maximum abnormal point temperature of each hardware and the position corresponding to the maximum abnormal point temperature of each hardware are obtained in sequence; Take the position corresponding to the maximum abnormal point temperature of each hardware as the center of the circle, connect the centers of each circle with a straight line to obtain a triangular figure; Extract the minimum abnormal point temperature from the temperatures greater than the maximum allowable temperature of the hardware, and obtain a triangular graph constructed by the minimum abnormal point temperature based on the above content; Obtain the overlapping area of the triangle constructed by the maximum abnormal point temperature of each hardware and the triangle constructed by the minimum abnormal point temperature of each hardware, and record it as the stacking face value; Calculate the difference between each maximum abnormal point temperature and the maximum allowable temperature of the hardware to obtain the point temperature difference; Again, the position corresponding to the maximum abnormal point temperature is taken as the center of the circle, and the corresponding temperature differences of each point are taken as the radius, so as to obtain three circles, and the three circles cut the triangle figure respectively to obtain three line segments; The conversion between temperature values and length values includes: Set different temperature conditions in each hardware area of the switch. The temperature change of the hardware can be controlled by heating or cooling devices. At the same time, use high-precision measuring tools, such as laser rangefinders or high-precision calipers, to measure the distance change between specific points on the hardware at different temperatures.

[0022] Through multiple experiments, record the corresponding length change data under different temperature changes; Analyze and fit the experimental data to find the mathematical relationship between the temperature change and the length change of each hardware; it can be assumed that there is a non-linear relationship between them: ; where is the amount of length change; is the amount of temperature change; 、 、 are fitting coefficients; The fitting coefficients can be determined by methods such as the least squares method to improve the accuracy of the conversion; After obtaining the conversion coefficient in this way, convert the point temperature difference value into a length value to construct a circle, so that the three circles can stably obtain three line segments after cutting the triangular graph respectively; Take the two ends of the three line segments as connection points respectively, and connect the connection points of the three line segments in sequence to obtain a closed hexagon, which is the hexagon with the largest area that can be constructed; Obtain the center of the hexagon and the temperature detected at the center, and record it as the center temperature. Subtract the preset maximum allowable temperature of the hardware from the center temperature to obtain the center temperature difference, and record the center temperature difference greater than 0 as the center abnormal temperature difference; Take the center of the hexagon as the center of the circle and the center abnormal temperature difference as the radius to make a circle, and divide the area of the circle by the area of the hexagon to obtain the occupancy ratio; After comprehensively processing the stacked face value, the center abnormal temperature difference and the occupancy ratio, obtain the temperature anomaly coefficient, including the following parts: After normalizing the stacked face value, the center abnormal temperature difference and the occupancy ratio, take the stacked face value and the center abnormal temperature difference as the two right-angled sides of a triangle respectively, connect the remaining side to form a complete right-angled triangle, take the occupancy ratio as the height of the right-angled triangle, construct a triangular pyramid model, and calculate the volume of the triangular pyramid model, and record it as the temperature anomaly coefficient; Switch fault judgment module: Judge whether there is a fault in the switch hardware based on the temperature anomaly coefficient; Specifically include: Preset the temperature anomaly coefficient threshold, compare the temperature anomaly coefficient with the temperature anomaly coefficient threshold. If the temperature anomaly coefficient is greater than the temperature anomaly coefficient threshold, it is determined that there is a fault in the switch hardware; An intelligent management method based on private network communication, including: Data collection: Obtain the round-trip time data, packet loss rate data, and throughput data of each node; Data analysis block: After comprehensively analyzing the round-trip time data, packet loss rate data, and throughput data of the node, obtain a trigger evaluation coefficient, and trigger the analysis of the node switch based on the trigger evaluation coefficient; Including: After analyzing the round-trip time data, packet loss rate data, and throughput data of the node, obtain a round-trip delay value, a high packet loss value, and a throughput deviation value; Preset the weight factors of the round-trip delay value, high packet loss value, and throughput deviation value, and calculate the sum of the products of the round-trip delay value, high packet loss value, and throughput deviation value and their corresponding weight factors respectively to obtain the trigger evaluation coefficient; Preset the trigger evaluation coefficient threshold, compare the trigger evaluation coefficient with the trigger evaluation coefficient threshold. If the trigger evaluation coefficient is greater than the trigger evaluation coefficient threshold, trigger the analysis of the node hardware; Node analysis: After analyzing the internal temperature of the node switch, obtain the temperature anomaly coefficient; Switch hardware diagnosis: Based on the temperature anomaly coefficient, determine whether there is a hardware fault in the switch hardware.

[0023] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The influence weight factors and specific coefficient values in the formula are set by those skilled in the art according to the actual situation and can be adjusted and modified later.

[0024] The above description of the embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent management system based on private network communication, characterized in that, Including: Data collection module: Obtain relevant network data of each node; Data analysis and judgment module: Analyze the relevant network data of the node to obtain a trigger evaluation coefficient, and trigger the analysis of the node switch based on the trigger evaluation coefficient; Node analysis module: Analyze the internal temperature of the node switch to obtain a temperature anomaly coefficient; Switch failure judgment module: Judge whether there is a failure in the switch hardware based on the temperature anomaly coefficient.

2. An intelligent management system based on private network communication according to claim 1, characterized in that, The data collection module specifically includes: Conduct area division according to the working areas corresponding to the subnet switches of each node, obtain the round-trip time data, packet loss rate data, and throughput data of each node, and preprocess the data.

3. An intelligent management system based on private network communication according to claim 2, characterized in that, The data analysis module specifically includes: Analyze the round-trip time data, packet loss rate data, and throughput data of the node to obtain a round-trip delay value, a high packet loss value, and a throughput deviation value, and comprehensively analyze the round-trip delay value, the high packet loss value, and the throughput deviation value to obtain a trigger evaluation coefficient; Preset a trigger evaluation coefficient threshold, compare the trigger evaluation coefficient with the trigger evaluation coefficient threshold. If the trigger evaluation coefficient is greater than the trigger evaluation coefficient threshold, trigger the analysis of the node hardware.

4. An intelligent management system based on private network communication according to claim 3, characterized in that, The process of obtaining the round-trip delay value, the high packet loss value, and the throughput deviation value by analyzing the round-trip time data, packet loss rate data, and throughput data of the node specifically includes: Obtain the data round-trip time of the node at a preset time interval, and arrange the obtained round-trip times in a time series; preset a round-trip time threshold, and successively subtract the round-trip time threshold from each obtained round-trip time, arrange the obtained round-trip time differences in ascending order of numerical value, and eliminate the round-trip time differences less than 0; count the number of round-trip time differences greater than 0 and divide it by the total number of round-trip time differences to obtain a time deviation degree; Extract the maximum round-trip time difference from the round-trip time differences greater than 0, and obtain the duration corresponding to the maximum round-trip time difference. Multiply the maximum round-trip time difference by its corresponding duration to obtain a single maximum delay amount; Perform a weighted calculation on the time deviation degree and the single maximum delay amount to obtain a round-trip delay value; Within a preset test time period, the sender sends a specified number of test data packets according to the configuration requirements, and the receiver synchronously receives the data packets; after the test is completed, calculate the difference between the total number of sent data packets and the actual number of received data packets, and divide this difference by the total number of sent data packets to obtain the packet loss rate during the network transmission process; Conduct a preset number of packet loss tests based on the above content to obtain corresponding packet loss rates; Preset an allowable fluctuation range of the packet loss rate, successively match each packet loss rate with the allowable fluctuation range of the packet loss rate, and record the packet loss rates that are not within the allowable fluctuation range of the packet loss rate as abnormal packet loss rates; Arrange the abnormal packet loss rates in descending order from left to right according to their numerical values, and mark the three largest abnormal packet loss rates; obtain the time intervals between the three largest abnormal packet loss rates, and mark them as the first time interval, the second time interval, and the third time interval respectively; An ellipse model is established with the first time interval and the second time interval as the major semi-axis and minor semi-axis of the ellipse respectively, and the third time interval is used as the height of the ellipse model to establish an ellipsoid. The volume of the ellipsoid is calculated and denoted as the packet loss height value.

5. An intelligent management system based on private network communication according to claim 4, characterized in that, The process of obtaining the throughput deviation value includes: Obtain the amount of data successfully transmitted by the network within a preset duration; divide the successfully transmitted data volume by the preset duration to obtain the throughput value, and conduct multiple tests to obtain the corresponding number of throughput values; Arrange each throughput value according to the time series; Obtain the standard throughput value based on the link nominal bandwidth, and successively subtract the standard throughput value from each throughput value to obtain the throughput difference; eliminate the throughput differences greater than 0, and take the absolute value of the throughput differences less than 0 and denote them as throughput difference pair values; Calculate the average value of each throughput difference pair value to obtain the throughput difference pair average value; Calculate the difference between the maximum throughput difference pair value and the minimum throughput difference pair value to obtain the throughput difference pair extreme value; Perform a weighted calculation on the throughput difference pair average value and the throughput difference pair extreme value to obtain the throughput deviation value.

6. An intelligent management system based on private network communication according to claim 1, characterized in that, The process of obtaining the temperature anomaly coefficient includes: Divide the interior of the switch into regions according to the locations of each hardware to obtain the regions corresponding to each hardware, and denote them as hardware regions; The hardware includes: switching chips, CPUs, and memory chips; Install temperature sensors at preset positions in the hardware regions to collect the temperatures at each position. Preset the maximum allowable temperature of the hardware, compare the temperature at each position with the maximum allowable temperature of the hardware, denote the temperature greater than the maximum allowable temperature of the hardware as the abnormal point temperature, and extract the maximum abnormal point temperature among them; Successively obtain the maximum abnormal point temperature of each hardware and the location corresponding to the maximum abnormal point temperature of each hardware according to the above steps; Taking the locations corresponding to the maximum abnormal point temperature of each hardware as the centers, connect the centers with straight lines to obtain a triangular figure; Extract the minimum abnormal point temperature from the temperatures greater than the maximum allowable temperature of the hardware, and obtain the triangular figure constructed by the minimum abnormal point temperature according to the above content; Obtain the overlapping area of the triangle constructed by the maximum abnormal point temperature of each hardware and the triangle constructed by the minimum abnormal point temperature of each hardware, and denote it as the stacking face value; Calculate the difference between the maximum abnormal point temperature and the maximum allowable temperature of the hardware to obtain the point temperature difference value; Again, taking the locations corresponding to the maximum abnormal point temperature as the centers, and using the corresponding point temperature difference values as the radii respectively to obtain three circles, and the three circles respectively cut the triangular figure to obtain three line segments; Taking the two ends of the three line segments as connection points respectively, connect the connection points of the three line segments in sequence to obtain a closed hexagon; Obtain the center of the hexagon and the temperature detected at the center, and denote it as the center temperature. Subtract the preset maximum allowable temperature of the hardware from the center temperature to obtain the center temperature difference, and denote the center temperature difference greater than 0 as the center abnormal temperature difference; Taking the center of the hexagon as the center and the center abnormal temperature difference as the radius to draw a circle, divide the area of the circle by the area of the hexagon to obtain the occupancy ratio; Perform comprehensive processing on the stacking face value, the center abnormal temperature difference, and the occupancy ratio to obtain the temperature anomaly coefficient.

7. An intelligent management system based on private network communication according to claim 1, characterized in that After normalizing the stacked face value, central differential temperature, and occupancy ratio, the stacked face value and central differential temperature are respectively used as the two right-angled sides of a triangle, and the remaining side is connected to form a complete right-angled triangle. The occupancy ratio is used as the height of the right-angled triangle to construct a triangular pyramid model, and the volume of the triangular pyramid model is calculated and denoted as the temperature anomaly coefficient.

8. An intelligent management system based on private network communication according to claim 7, characterized in that, Based on the temperature anomaly coefficient, it is determined whether there is a fault in the switch hardware, specifically including: A temperature anomaly coefficient threshold is preset, and the temperature anomaly coefficient is compared with the temperature anomaly coefficient threshold. If the temperature anomaly coefficient is greater than the temperature anomaly coefficient threshold, it is determined that there is a fault in the switch hardware.

9. An intelligent management method based on private network communication, which uses an intelligent management system based on private network communication according to any one of claims 1-8, is characterized in that, Including: Data collection: Obtain the round-trip time data, packet loss rate data, and throughput data of each node; Data analysis block: After comprehensively analyzing the round-trip time data, packet loss rate data, and throughput data of the node, a trigger evaluation coefficient is obtained, and based on the trigger evaluation coefficient, the node switch is triggered for analysis; Node analysis: After analyzing the internal temperature of the node switch, a temperature anomaly coefficient is obtained; Switch hardware diagnosis: Based on the temperature anomaly coefficient, it is determined whether there is a hardware fault in the switch hardware.

Images