Communication management machine upgrading method and system

Abstract

The invention provides a communication management machine upgrading method and system, and belongs to the technical field of data processing, and the method comprises the steps: obtaining an available upgrading file from an upgrading file server, and downloading the upgrading file to a local storage; acquiring historical flow data; according to the historical traffic data, predicting the traffic data by using a jump gating cycle unit; analyzing a communication load according to the predicted flow data, and determining a low-load period; checking by using a Z-Score threshold value, and judging whether the low-load time period is a real low-load time period or not; if yes, entering the next step to execute the upgrading plan; otherwise, delaying the upgrading plan; the upgrading file is decompressed and installed in a standby system partition, and a standby system is subjected to pre-start check and test; when the standby system runs stably, the system is guided to the standby system partition through a hardware switching mechanism, and system upgrading is carried out. A hardware switching mechanism is adopted, the upgrading process almost has no influence on a user, and the system can be seamlessly switched.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a communication management machine upgrading method and system. Background Art

[0002] In the operation and management of new energy stations, the communication management machine plays a vital role. It is responsible for connecting the local intelligent equipment and the main control room background to realize data collection, transmission and management. In order to ensure that the communication management machine system can operate long-term, stably and safely, the communication management machine needs to be updated.

[0003] Currently, the upgrade method of the communication management machine relies on manual intervention, such as manually starting the upgrade, adjusting configuration parameters, and fixing errors. Especially when facing complex hardware devices and different software versions, error handling and fault recovery during the upgrade process often require operator intervention.

[0004] However, the manual intervention upgrade method increases the probability of errors and the complexity of operations. The system usually goes down or becomes unavailable during the upgrade process, resulting in unstable services and a degraded user experience. Summary of the invention

[0005] The main purpose of the present invention is to provide a communication management machine upgrade method and system to solve the problem that the manual intervention upgrade method in the prior art increases the probability of errors and the complexity of operations. The system usually shuts down or becomes unavailable during the upgrade process, resulting in unstable services and reduced user experience.

[0006] In order to achieve the above object, according to one aspect of the present invention, a communication management machine upgrade method is provided, comprising: S1: Obtain available upgrade files from the upgrade file server and download them to local storage; S2: Get historical traffic data; S3: predicting the traffic data using a jump gated recurrent unit according to the historical traffic data; S4: Analyze the communication load and determine the low-load period based on the predicted traffic data; S5: Using the Z-Score threshold check, determine whether the low-load period is a real low-load period; if so, proceed to the next step to execute the upgrade plan; otherwise, postpone the upgrade plan; S6: decompressing the upgrade file and installing it to the standby system partition, and performing a pre-startup check and test on the standby system; S7: When the backup system runs stably, the system is booted to the backup system partition through the hardware switching mechanism to perform system upgrade.

[0007] Furthermore, the S3 specifically includes: S301: constructing historical traffic time series data based on the historical traffic data; S302: Using a perceptron-based skip network, calculate the skip probability of the historical traffic time series data input into the gated recurrent unit: ; Among them, S t represents the hidden state of the perceptron at time t, RELU represents the RELU activation function, and W 1 represents the weight matrix of the first layer of perceptron, x t represents the historical traffic time series data at time t, b 1 represents the bias term of the first layer of perceptron, π t represents the skip probability of the historical traffic time series data at time t, Softmax represents the Softmax activation function, and W 2 represents the weight matrix of the second layer perceptron, b 2 Represents the bias term of the second layer perceptron; S303: skipping the historical traffic time series data according to the skipping probability of the historical traffic time series data at each moment, and inputting the data into the gated cycle unit; S304: extracting the hidden state in the historical traffic time series data through a gated recurrent unit: ; Among them, z t represents the output vector of the update gate at time t, σ() represents the activation function, and W uz represents the weight matrix between the input layer and the update gate, X t represents the traffic time series data at time t, W hz represents the self-connection weight matrix of the update gate between time t and time t-1, h t-1 represents the hidden state vector at time t-1, b z represents the bias term of the update gate, r t represents the output vector of the reset gate at time t, W ur represents the weight matrix between the input layer and the reset gate, W hr represents the self-connection weight matrix between the reset gate at time t and time t-1, b r represents the bias term of the reset gate, c t represents the output vector of the candidate layer at time t, W uc Represents the weight matrix between the input layer and the candidate layer, W hc represents the self-connection weight matrix between the candidate layer at time t and time t-1, b c represents the bias term of the candidate layer, e represents the element-by-element product operation, and ht represents the hidden state vector at time t; S305: Predicting the traffic data according to the hidden state: ; in, represents the traffic data prediction value at time t+1, σ() represents the activation function, and W R represents the prediction weight matrix, b R Represents the bias term of the prediction layer.

[0008] Furthermore, the skip gated recurrent unit is optimized by using a Seagull optimization algorithm.

[0009] Furthermore, the S4 is specifically: determining the portion of the predicted flow data that is smaller than the flow threshold as the low-load period.

[0010] Furthermore, the S5 specifically includes: S501: Calculate the mean and standard deviation of historical traffic data; S502: For the predicted traffic data at each time point in the low-load period, calculate the Z-Score: ; Among them, Z t It represents the Z-Score of the predicted traffic data at time t. Y t represents the predicted flow data at time t, μ represents the mean of historical flow data, and σ represents the standard deviation of historical flow data; S503: Determine whether the Z-Score of the predicted traffic data at each time point in the low-load period is less than the dynamic threshold; if so, determine that the low-load period is a real low-load period, and proceed to the next step to execute the upgrade plan; otherwise, determine that the low-load period is a false low-load period, and postpone the upgrade plan.

[0011] Furthermore, the dynamic threshold is specifically: ; Among them, Z th represents the dynamic threshold, Z 0 represents the basic threshold, k represents the adjustment coefficient, and C represents the fluctuation coefficient; ; Among them, μ represents the mean of historical traffic data, and σ represents the standard deviation of historical traffic data.

[0012] Furthermore, the postponing of the upgrade plan in S5 specifically includes: postponing the upgrade plan to the next low-load period for execution.

[0013] Furthermore, the S7 specifically includes: S701: monitor the health status of the backup system; S702: Based on data consistency verification, confirm that the standby system has synchronized data; S703: Verify the startup environment of the standby system; S704: Update the boot program and switch to the backup system partition; S705: starting the backup system partition; S706: Install the upgrade file; S707: During the upgrade process, the usage of system resources is continuously monitored. If the system detects an abnormality or upgrade failure, a rollback mechanism is immediately initiated to restore the system to the state before the upgrade; S708: After the upgrade is completed, perform system verification.

[0014] Furthermore, S702 specifically includes: verifying data consistency by verifying the hash value to confirm that the standby system has synchronized the data.

[0015] According to one aspect of the present invention, there is provided a communication management machine upgrade system, comprising: processor; The memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the above-mentioned communication management machine upgrade method is implemented.

[0016] By applying the technical solution of the present invention, by using a backup system partition and adopting a hardware switching mechanism, the upgrade process has almost no impact on users, the system can switch seamlessly, avoiding downtime or service interruption that may be caused during the upgrade process, ensuring the continuous operation of the communication management machine, enhancing service stability, and improving user experience.

[0017] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings: Figure 1 A schematic diagram of a flow chart of a communication management machine upgrade method provided by an embodiment of the present invention is shown; Figure 2 A schematic structural diagram of a communication management machine upgrade system provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0019] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0020] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0021] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0022] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0023] Reference Manual Attached Figure 1 , showing a flow chart of a communication management machine upgrade method provided in an embodiment of the present invention.

[0024] An embodiment of the present invention provides a communication management machine upgrade method, comprising: S1: Obtain available upgrade files from the upgrade file server and download them to local storage.

[0025] S2: Get historical traffic data.

[0026] S3: Based on the historical traffic data, the traffic data is predicted using the jump gated recurrent unit.

[0027] Among them, the gated recurrent unit (GRU) is an improved recurrent neural network (RNN) structure. GRU dynamically determines which information needs to be retained and which needs to be forgotten by introducing two gating mechanisms - update gate and reset gate. The update gate controls the degree of integration of the current input and previous state information, while the reset gate determines how to combine the current input and past information. The GRU structure is relatively simple and has high computational efficiency. It can effectively capture long-term dependencies in time series data and is widely used in tasks such as natural language processing and time series prediction.

[0028] As the sequence length increases, the traditional gated recurrent unit needs to process the input of each time step, which will bring higher computational complexity and the model may lose important long-term dependency information. Therefore, the present invention creatively introduces a jump mechanism.

[0029] In a possible implementation, S3 specifically includes sub-steps S301 to S305: S301: Construct historical traffic time series data based on historical traffic data.

[0030] S302: Using a perceptron-based skip network, calculate the skip probability of the historical traffic time series data input into the gated recurrent unit: ; Among them, S t represents the hidden state of the perceptron at time t, RELU represents the RELU activation function, and W 1 represents the weight matrix of the first layer of perceptron, x t represents the historical traffic time series data at time t, b 1 represents the bias term of the first layer of perceptron, π t represents the skip probability of the historical traffic time series data at time t, Softmax represents the Softmax activation function, and W 2 represents the weight matrix of the second layer perceptron, b 2 Represents the bias term of the second layer perceptron.

[0031] S303: Perform skip processing on the historical traffic time series data according to the skip probability of the historical traffic time series data at each moment, and input the data into the gated cycle unit.

[0032] In the present invention, the skip network can effectively and selectively skip those unimportant input data. This skip mechanism reduces the computational burden of each time step, thereby improving computational efficiency. For large-scale or long-time series of traffic data, the skip network can help avoid unnecessary calculations, especially when faced with a large amount of historical traffic data, it can automatically select the data that is most critical to prediction for processing.

[0033] S304: Extract the hidden state in the historical traffic time series data through the gated recurrent unit: ; Among them, z t represents the output vector of the update gate at time t, σ() represents the activation function, and W uz represents the weight matrix between the input layer and the update gate, X t represents the traffic time series data at time t, W hz represents the self-connection weight matrix of the update gate between time t and time t-1, h t-1 represents the hidden state vector at time t-1, b z represents the bias term of the update gate, r t represents the output vector of the reset gate at time t, W ur represents the weight matrix between the input layer and the reset gate, W hr represents the self-connection weight matrix between the reset gate at time t and time t-1, b r represents the bias term of the reset gate, c t represents the output vector of the candidate layer at time t, W uc Represents the weight matrix between the input layer and the candidate layer, W hc represents the self-connection weight matrix between the candidate layer at time t and time t-1, b c represents the bias term of the candidate layer, e represents the element-by-element product operation, and h t represents the hidden state vector at time t.

[0034] In the present invention, the gated recurrent unit can effectively capture the long-term dependencies in time series data and control the flow of information through mechanisms such as update gates and reset gates, so that the system can better process complex time series data and extract useful hidden states from them, thereby helping to perform accurate traffic prediction and trend analysis.

[0035] S305: Predict the traffic data based on the hidden state: ; in, represents the traffic data prediction value at time t+1, σ() represents the activation function, and W R represents the prediction weight matrix, bR Represents the bias term of the prediction layer.

[0036] In the present invention, by combining the perceptron-based skip network with the gated recurrent unit, the efficiency, accuracy and flexibility of traffic data prediction can be effectively improved. The skip network optimizes the calculation process, enhances the adaptability and selectivity of the model, can more efficiently capture the key features in the time series data, and reduce unnecessary calculations, thereby improving the accuracy of traffic prediction and low-load period analysis, and further improving the intelligence level and stability of the communication management machine upgrade process.

[0037] In a possible implementation, the skip-gated recurrent unit is optimized by using a Seagull optimization algorithm.

[0038] Specifically, the inverse of the cross entropy loss function is used as the fitness function of the Seagull optimization algorithm.

[0039] Initialize the seagull individuals. Each seagull individual represents a feasible XX set. Each seagull individual consists of multiple dimensional components, and each component represents one; In the global search phase, avoid collisions and move towards the optimal individual: ; ; ; in, represents the position of the i-th seagull individual after the global search phase at the t-th iteration, represents the position of the i-th seagull after anti-collision processing at the t-th iteration, A represents the control factor, represents the position of the i-th seagull at the t-th iteration, represents the displacement of the i-th seagull individual towards the optimal individual at the t-th iteration, B represents the search balance factor, represents the optimal individual position at the tth iteration; In the present invention, the position of the individual is adjusted so that it moves in a reasonable direction of its current position, while ensuring that the individual does not deviate excessively when approaching the global optimal individual, thereby avoiding collision and search confusion.

[0040] ; Among them, t represents the current number of iterations, T represents the maximum number of iterations, and f c Indicates linearly decreasing frequency; In the present invention, the control factor is gradually reduced as the number of iterations increases. In the early stage, a larger control factor can make the search process have a larger step, helping the algorithm to explore a wider solution space, and as the iterations proceed, the control factor is gradually reduced, making the search process more refined, helping the algorithm to find the optimal solution more accurately.

[0041] ; Among them, r 1 Represents a random number between 0 and 1; In the local search phase, a random number r is generated 2 , according to the random number r 2 , choose between the spiral search strategy and the bracketing strategy in parallel, and move in a spiral motion: ; ; ; ; ; in, represents the position of the i-th seagull after the spiral motion at the t-th iteration, x represents the spiral flight coefficient in the x direction, y represents the spiral flight coefficient in the y direction, z represents the spiral flight coefficient in the z direction, r represents the radius of the spiral flight trajectory, θ represents a random number between 0 and 2π, u and v represent spiral constants, and e represents a natural constant;

[0042] In the present invention, by introducing the spiral search strategy and the encirclement strategy in the local search stage, and selecting between the two according to the random number r2, the diversity and flexibility of the search process can be effectively enhanced. The spiral motion mode allows the individual to explore near the current position, while the encirclement strategy ensures that the individual can perform a detailed search around the current optimal solution. The use of random numbers to switch between the two strategies ensures that the search direction has appropriate randomness and controllability, further improving the efficiency and accuracy of the search. This strategy improves the adaptability of the algorithm, making the optimization process more efficient and having a stronger global exploration capability.

[0043] Perform mutation operations on each seagull individual: ; in, represents the position of the ith seagull individual after mutation at the tth iteration, P r represents a random individual, ω represents an adaptive scaling factor; In the present invention, the mutation operation enables each seagull individual to perform appropriate random deviations based on the current optimal solution to explore new solution spaces. This dynamic adjustment method helps the algorithm maintain search stability while enhancing the ability to jump out of the local optimal solution. By adjusting the adaptive scaling factor, the variation amplitude can be flexibly adjusted according to the current search state, further improving the accuracy and efficiency of the search.

[0044] ; Among them, ω max represents the maximum scale factor, ω min Indicates the minimum scale factor; In the present invention, the adaptive scaling factor can realize the dynamic adjustment of the scaling factor, so that the search process has different exploration characteristics at different stages. In the early stage, a larger scaling factor can promote a wider search, increase the diversity of exploration, and prevent the algorithm from converging prematurely; and as the iteration proceeds, the scaling factor gradually decreases, so that the search process is more focused on the fine optimization of the optimal solution. This method helps to balance the relationship between global search and local optimization, improves search efficiency, avoids stagnation at the local optimal solution, and enhances the ability to find the global optimal solution.

[0045] Determine whether the fitness value of the position after mutation is greater than the fitness value of the position before mutation; if so, use the position after mutation to replace the position before mutation; otherwise, keep the position before mutation unchanged; Update the fitness value of each seagull individual and the global optimal individual; Determine whether the current number of iterations has reached the maximum number of iterations; if so, output the seagull individual representative with the highest current fitness; otherwise, return to continue iterating.

[0046] In the present invention, the jump gated recurrent unit (GRU) is optimized by the Seagull optimization algorithm. By dynamically adjusting the parameters, the adaptability of the model is enhanced, the effect of traffic prediction is improved, the convergence process is accelerated, and the risks of overfitting and over-dependence on initial conditions are avoided, thereby improving the stability and accuracy of the entire prediction process.

[0047] S4: Analyze the communication load and determine the low-load period based on the predicted traffic data.

[0048] In a possible implementation, S4 specifically includes: determining the portion of the predicted flow data that is less than the flow threshold as a low-load period.

[0049] Among them, those skilled in the art can set the size of the flow threshold according to actual conditions, and the present invention does not limit it.

[0050] In the present invention, the advantage of determining the portion of the predicted traffic data that is less than the traffic threshold as a low-load period is that it can accurately identify the period when the network load is relatively light, thereby providing the best opportunity for system upgrade.

[0051] S5: Use the Z-Score threshold check to determine whether the low-load period is a real low-load period. If so, proceed to the next step to execute the upgrade plan. Otherwise, postpone the upgrade plan.

[0052] In a possible implementation, S5 specifically includes sub-steps S501 to S503: S501: Calculate the mean and standard deviation of historical traffic data.

[0053] S502: For the predicted traffic data at each time point during the low-load period, calculate the Z-Score: ; Among them, Z t It represents the Z-Score of the predicted traffic data at time t. Y t represents the predicted traffic data at time t, μ represents the mean of historical traffic data, and σ represents the standard deviation of historical traffic data.

[0054] S503: Determine whether the Z-Score of the predicted traffic data at each time point in the low-load period is less than the dynamic threshold. If so, determine that the low-load period is a real low-load period, and proceed to the next step to execute the upgrade plan. Otherwise, determine that the low-load period is a false low-load period, and postpone the upgrade plan.

[0055] In the present invention, Z-Score verification measures the deviation between predicted traffic and historical traffic data to ensure that upgrades are performed only during low-load periods when traffic data is stable and in line with expectations, thereby avoiding upgrade operations during abnormal traffic conditions. In this way, the system can ensure the accuracy and reliability of the upgrade plan, reduce upgrade failures or service interruptions caused by traffic fluctuations, emergencies, or short-term traffic changes, and further improve system stability and user experience.

[0056] Optionally, the dynamic threshold is specifically: ;

[0057] Among them, Z th represents the dynamic threshold, Z 0 represents the basic threshold, k represents the adjustment coefficient, and C represents the fluctuation coefficient.

[0058] ; Among them, μ represents the mean of historical traffic data, and σ represents the standard deviation of historical traffic data.

[0059] In the present invention, a dynamic threshold is used, which can automatically adjust the judgment criteria according to the fluctuation of historical traffic data. This method can flexibly adjust the threshold according to the degree of fluctuation of actual traffic, avoiding premature or erroneous upgrades during periods of large traffic fluctuations. When the traffic fluctuates greatly, the threshold will be reduced accordingly, making the system more sensitive to traffic changes; conversely, the threshold will be increased to reduce unnecessary upgrade operations. This adaptive threshold adjustment improves the accuracy and stability of the upgrade timing, avoids upgrading during periods of abnormal traffic, and thus ensures high availability of the system and high efficiency of the upgrade.

[0060] Optionally, postponing the upgrade plan in S5 specifically includes postponing the upgrade plan to the next low-load period for execution.

[0061] In the present invention, the upgrade plan is postponed to the next low-load period, which can ensure that the upgrade is performed during the period when the network load is light, thereby avoiding interference with normal business operations. By postponing the upgrade, the system can choose to perform the upgrade operation during the period when the traffic is stable and low, reducing the impact on users and ensuring that the service is not interrupted or the performance is degraded.

[0062] S6: Unzip the upgrade file and install it to the standby system partition, and perform a pre-startup check and test on the standby system.

[0063] S7: When the backup system runs stably, the system is booted to the backup system partition through the hardware switching mechanism to perform system upgrade.

[0064] In a possible implementation, S7 specifically includes sub-steps S701 to S708: S701: Monitor the health status of the backup system.

[0065] S702: Based on data consistency verification, confirm that the standby system has synchronized data.

[0066] Optionally, S702 specifically includes: performing data consistency verification by verifying the hash value to confirm that the standby system has synchronized the data.

[0067] It should be noted that monitoring the health status of the standby system and verifying data consistency ensures that the standby system is ready and synchronized with the primary system before switching, thereby avoiding upgrade failures due to data inconsistency or system instability.

[0068] S703: Verify the startup environment of the standby system.

[0069] It should be noted that startup environment verification and boot program update ensure that the backup system can take over the task smoothly, while real-time monitoring of system resources and rollback mechanism provide fault detection and recovery guarantee for the upgrade process, ensuring the security of the upgrade process.

[0070] S704: Update the boot program and switch to the backup system partition.

[0071] S705: Start the backup system partition.

[0072] S706: Install the upgrade file.

[0073] S707: During the upgrade process, the usage of system resources is continuously monitored. If the system detects an abnormality or upgrade failure, the rollback mechanism is immediately activated to restore the system to the state before the upgrade.

[0074] It should be noted that if an exception or failure occurs, the system can be quickly rolled back and restored to the stable state before the upgrade, minimizing the impact on users.

[0075] S708: After the upgrade is completed, perform system verification.

[0076] It should be noted that the system verification after the upgrade is completed ensures the normal operation of the new system and guarantees the reliability and performance of the system after a successful upgrade.

[0077] From the above description, it can be seen that the communication management machine upgrade method provided by the present invention, by using a backup system partition and adopting a hardware switching mechanism, has almost no impact on the user during the upgrade process, and the system can switch seamlessly, avoiding downtime or service interruption that may be caused during the upgrade process, ensuring the continuous operation of the communication management machine, enhancing the stability of the service, and improving the user experience.

[0078] Reference Manual Attached Figure 2 , showing a schematic structural diagram of a communication management machine upgrade system provided by an embodiment of the present invention.

[0079] The embodiment of the present invention provides a communication management machine upgrade system 20, comprising: Processor 201; The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201, the above-mentioned communication management machine upgrade method is implemented.

[0080] From the above description, it can be seen that the communication management machine upgrade method provided by the present invention, by using a backup system partition and adopting a hardware switching mechanism, has almost no impact on the user during the upgrade process, and the system can switch seamlessly, avoiding downtime or service interruption that may be caused during the upgrade process, ensuring the continuous operation of the communication management machine, enhancing the stability of the service, and improving the user experience.

[0081] Unless otherwise specifically stated, the relative arrangement of the parts and steps described in these embodiments, numerical expressions and numerical values ​​do not limit the scope of the present invention. At the same time, it should be understood that, for ease of description, the sizes of the various parts shown in the accompanying drawings are not drawn according to the actual proportional relationship. The technology, method and equipment known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technology, method and equipment should be regarded as a part of the authorization specification. In all examples shown and discussed here, any specific value should be interpreted as being merely exemplary, rather than as a limitation. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once a certain item is defined in an accompanying drawing, it does not need to be further discussed in subsequent drawings.

[0082] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations), and the spatially relative descriptions used here are interpreted accordingly.

[0083] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, perpendicular, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the devices or elements referred to must have a specific direction or be constructed and operated in a specific direction. Therefore, they cannot be understood as limiting the scope of protection of the present invention. The directional words "inside and outside" refer to the inside and outside relative to the contours of each component itself.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A communication management machine upgrade method, characterized in that: include: S1: Obtain available upgrade files from the upgrade file server and download them to local storage; S2: Get historical traffic data; S3: predicting the traffic data using a jump gated recurrent unit according to the historical traffic data; S4: Analyze the communication load and determine the low-load period based on the predicted traffic data; S5: Using the Z-Score threshold check, determine whether the low-load period is a real low-load period; if so, proceed to the next step and execute the upgrade plan; Otherwise, postpone the upgrade plan; S6: decompressing the upgrade file and installing it to the standby system partition, and performing a pre-startup check and test on the standby system; S7: When the backup system runs stably, the system is booted to the backup system partition through the hardware switching mechanism to perform system upgrade.

2. The communication management machine upgrade method according to claim 1, characterized in that: The S3 specifically includes: S301: constructing historical traffic time series data based on the historical traffic data; S302: Using a perceptron-based skip network, the skip probability of the historical traffic time series data being input into a gated recurrent unit is calculated: ； Among them, S t represents the hidden state of the perceptron at time t, RELU represents the RELU activation function, W1 represents the weight matrix of the first layer of perceptron, x t represents the historical traffic time series data at time t, b1 represents the bias term of the first layer perceptron, π t represents the skip probability of the historical traffic time series data at time t, Softmax represents the Softmax activation function, W2 represents the weight matrix of the second-layer perceptron, and b2 represents the bias term of the second-layer perceptron; S303: skipping the historical traffic time series data according to the skipping probability of the historical traffic time series data at each moment, and inputting the data into the gated cycle unit; S304: Extracting the hidden state in the historical traffic time series data through a gated recurrent unit: ； Among them, z t represents the output vector of the update gate at time t, σ() represents the activation function, and W uz represents the weight matrix between the input layer and the update gate, X t represents the traffic time series data at time t, W hz represents the self-connection weight matrix of the update gate between time t and time t-1, h t-1 represents the hidden state vector at time t-1, b z represents the bias term of the update gate, r t represents the output vector of the reset gate at time t, W ur represents the weight matrix between the input layer and the reset gate, W hr represents the self-connection weight matrix between the reset gate at time t and time t-1, b r represents the bias term of the reset gate, c t represents the output vector of the candidate layer at time t, W uc Represents the weight matrix between the input layer and the candidate layer, W hc represents the self-connection weight matrix between the candidate layer at time t and time t-1, b c represents the bias term of the candidate layer, e represents the element-by-element product operation, and h t represents the hidden state vector at time t; S305: Predicting the traffic data according to the hidden state: ； in, represents the traffic data prediction value at time t+1, σ() represents the activation function, and W R represents the prediction weight matrix, b R Represents the bias term of the prediction layer.

3. The communication management machine upgrade method according to claim 1, characterized in that: The skip gated recurrent unit is optimized by means of a Seagull optimization algorithm.

4. The communication management machine upgrade method according to claim 1, characterized in that: The S4 is specifically: The portion of the predicted traffic data that is smaller than the traffic threshold is determined as the low-load period.

5. The communication management machine upgrade method according to claim 1, characterized in that: The S5 specifically includes: S501: Calculate the mean and standard deviation of historical traffic data; S502: For the predicted traffic data at each time point in the low-load period, calculate the Z-Score: ； Among them, Z t It represents the Z-Score of the predicted traffic data at time t. Y t represents the predicted flow data at time t, μ represents the mean of historical flow data, and σ represents the standard deviation of historical flow data; S503: Determine whether the Z-Score of the predicted traffic data at each time point in the low-load period is less than the dynamic threshold; if so, determine that the low-load period is a real low-load period, and proceed to the next step to execute the upgrade plan; otherwise, determine that the low-load period is a false low-load period, and postpone the upgrade plan.

6. The communication management machine upgrade method according to claim 5, characterized in that: The dynamic threshold is specifically: ； Among them, Z th represents the dynamic threshold, Z0 represents the basic threshold, k represents the adjustment coefficient, and C represents the fluctuation coefficient; ； Among them, μ represents the mean of historical traffic data, and σ represents the standard deviation of historical traffic data.

7. The communication management machine upgrade method according to claim 1, characterized in that: The postponed upgrade plan in S5 is specifically as follows: Postpone the upgrade until the next low-load period.

8. The communication management machine upgrade method according to claim 1, characterized in that: The S7 specifically includes: S701: monitor the health status of the backup system; S702: Based on data consistency verification, confirm that the standby system has synchronized data; S703: Verify the startup environment of the standby system; S704: Update the boot program and switch to the backup system partition; S705: starting the backup system partition; S706: Install the upgrade file; S707: During the upgrade process, the usage of system resources is continuously monitored. If the system detects an abnormality or upgrade failure, a rollback mechanism is immediately initiated to restore the system to the state before the upgrade; S708: After the upgrade is completed, perform system verification.

9. The communication management machine upgrade method according to claim 1, characterized in that: The S702 is specifically as follows: Data consistency is verified by verifying the hash value to confirm that the backup system has synchronized the data.

10. A communication management machine upgrade system, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the communication management machine upgrade method according to any one of claims 1 to 9 is implemented.

Images