Intelligent AI-driven network equipment self-inspection and maintenance system

Abstract

The invention relates to the technical field of equipment maintenance, in particular to an intelligent AI-driven network equipment self-inspection and maintenance system. Comprising a data acquisition and monitoring module, a data analysis and prediction module, an intelligent fault diagnosis module, a maintenance plan optimization module, a work order management and execution module, an inventory and spare part management module, a data visualization and report module and a data backup module. Compared with the prior art in which the inventory level is determined by adopting a fixed inventory quantity or a prediction method based on simple statistics, the method provided by the invention neglects the volatility of a purchase cycle and the dynamic change of the actual demand of spare parts, so that the risk of inventory overstock or shortage is relatively high; according to the scheme, multiple factors such as volatility of a historical purchase cycle, daily maintenance requirements, availability of spare parts and reliability of a supply chain are comprehensively considered, and the safe inventory quantity is obtained through accurate calculation, so that the flexibility and accuracy of inventory management are improved, the inventory cost is effectively reduced, and the timeliness and high efficiency of maintenance work are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment maintenance, and in particular to an intelligent AI-driven network equipment self-checking and maintenance system. Background Art

[0002] The spare parts inventory management required for network equipment maintenance usually relies on a fixed safety stock quantity or a forecasting method based on simple statistics. These methods have exposed many shortcomings in practical applications. First, the fixed safety stock quantity is difficult to adapt to the dynamic changes in spare parts demand. Due to the complexity of network equipment and the diversity of maintenance needs, the demand for spare parts often shows large fluctuations. The fixed inventory quantity may not be able to meet the spare parts demand during the peak period, resulting in obstruction of maintenance work, or causing inventory backlogs and increasing inventory costs.

[0003] Secondly, although the prediction method based on simple statistics takes into account historical data to a certain extent, it often ignores the volatility of the procurement cycle and the changes in spare parts availability. The volatility of the procurement cycle may lead to the instability of spare parts supply, while the changes in spare parts availability may affect the timely procurement of spare parts. These factors may have a negative impact on the timeliness and efficiency of maintenance work.

[0004] In addition, spare parts inventory management in existing technologies also faces challenges from changes in the supply chain environment. With changes in the market environment and the increase in uncertainties in the supply chain, spare parts inventory management needs to be more flexible and accurate. However, traditional inventory management methods often have difficulty adapting to such changes, resulting in low inventory management efficiency and failure to meet the needs of maintenance work. Summary of the invention

[0005] In order to overcome the problems raised in the above background technology, the present invention proposes an intelligent AI-driven network equipment self-inspection and maintenance system.

[0006] The technical solution of the present invention is: an intelligent AI-driven network equipment self-checking and maintenance system, comprising: The data collection and monitoring module is used to collect the operating status data of network devices in real time through the Internet of Things technology, and monitor the network status, traffic and number of connections of network devices; Data analysis and prediction module, which is used to use big data analysis and machine learning algorithms to conduct in-depth mining and analysis of the collected data of network devices; Intelligent fault diagnosis module, which combines historical fault data and AI algorithms to make preliminary diagnosis of abnormal status of network equipment based on the data results of data analysis and prediction module; Maintenance plan optimization module, which is used to automatically generate and optimize maintenance plans based on the historical maintenance records and diagnostic data of the equipment; Work order management and execution module, used to realize online creation, allocation, tracking and closed-loop management of maintenance tasks; Inventory and spare parts management module, which is used to monitor spare parts inventory in real time, and intelligently generate procurement plans based on maintenance needs and forecast data to optimize spare parts inventory management; The data visualization and reporting module is used to integrate the system's work records and data and present them in a visual form.

[0007] Preferably, the data acquisition and monitoring module includes: A11: Sensor network, including temperature sensors, humidity sensors, pressure sensors, displacement sensors and light sensors, which are used to monitor and measure the physical quantities of network devices and convert the collected physical quantities into electrical signals; A12: Data collector, used to receive signals from the sensor network and convert the received signals into a usable data format; A13: Surveillance camera, used to capture and record network devices and surrounding environment images in real time; A14: Data storage device, used to store the digital signals converted by the data collector.

[0008] Preferably, the data analysis and prediction module includes the following steps when using big data analysis and machine learning algorithms to perform in-depth mining and analysis on the collected data of network devices: S11: Data extraction, extracting historical operation data of network devices from the database, including performance indicators, device logs, and fault records of network devices; S12: Data cleaning and preprocessing, applying big data analysis and machine learning algorithms to clean the data and normalize, standardize and discretize the data; S13: Feature extraction, extracting useful features from the cleaned and preprocessed data, where the algorithms used include principal component analysis, trend analysis algorithm, and seasonal analysis algorithm; S14: Model construction: first build a prediction model and set the model parameters, then use historical data to train the model; S15: Model application, inputting the data collected by the data acquisition and monitoring module into the trained model, and the model outputs the prediction results.

[0009] Preferably, the intelligent fault diagnosis module combines historical fault data and AI algorithms to make a preliminary diagnosis of the abnormal state of the network device according to the data results of the data analysis and prediction module, including the following steps: S21: Data reception, receiving the output results from the data analysis and prediction module, including real-time operation status data of network devices, abnormal status indications, and possible fault warning information, and preprocessing the output results; S22: Fault feature matching: Combine historical fault data and use AI algorithms to perform feature matching on the current status of the device; S23: Based on the result of feature matching, AI algorithm is applied to perform fault diagnosis and prediction, and the fault type corresponding to the current state is identified. The fault types include hardware fault, software fault and network fault. S24: Fault analysis: after identifying the fault type, the cause of the fault is analyzed in one step, wherein the fault causes include device aging, configuration errors and network attacks.

[0010] Preferably, when the intelligent fault diagnosis module uses the AI ​​algorithm to perform feature matching on the current state of the device, the AI ​​algorithm used is a deep learning algorithm, wherein the network structure of the deep learning algorithm includes an input layer, a hidden layer and an output layer, wherein the mapping from the input layer to the hidden layer is: ; The mapping from hidden layer to output layer is: ; Among them, x is the input data, is the activation value of the hidden layer, y is the output value, and are the weight matrices from the input layer to the hidden layer and from the hidden layer to the output layer, respectively. and are the bias terms of the hidden layer and the output layer, respectively. is the activation function, and T represents the transpose operation of the matrix.

[0011] Preferably, the work order management and execution module includes the following steps when realizing online creation, allocation, tracking and closed-loop management of maintenance tasks: S31: receiving the maintenance plan transmitted by the maintenance plan optimization module; S32: Creating a maintenance task online, wherein the maintenance task includes a task description, a task priority, and a task executor; S33: Allocate maintenance tasks to corresponding technicians and teams; S34: Tracking the execution status of the maintenance task, the execution status of the maintenance task includes the task progress and task results, wherein the task progress is recorded by image recording; S35: Implement closed-loop management of maintenance tasks, including task completion confirmation and task feedback.

[0012] As a preferred embodiment, the inventory and spare parts management module monitors the spare parts inventory in real time, and intelligently generates a procurement plan based on maintenance needs and forecast data to optimize spare parts inventory management, including the following steps: S41: real-time monitoring of spare parts inventory, wherein the spare parts inventory includes spare parts types, quantities and storage locations; S42: Forecast maintenance needs and intelligently generate procurement plans based on maintenance needs; S43: Send the purchase plan to the relevant staff for review to ensure that the purchased spare parts meet the actual needs; S44: Purchase spare parts according to the purchase plan and track the purchase progress in real time.

[0013] Preferably, when predicting maintenance needs, the following steps are included: S51: collecting historical data, wherein the collected historical data includes spare parts types, spare parts replacement time, spare parts replacement frequency and spare parts procurement cycle; S52: Calculate the replacement frequency of spare parts within a procurement cycle based on historical data, and then identify the highest value of the replacement frequency of various types of spare parts in all procurement cycles, wherein a moving average method is used for prediction; S53: Determine safety stock: Determine the safety stock level of various spare parts based on the historical maximum replacement frequency and supply chain reliability during the procurement cycle; S54: Calculate the required quantity of spare parts, add the daily maintenance demand and the safety stock, and obtain the total required quantity of each type of spare parts.

[0014] As a preferred method, when predicting maintenance needs, the principle formula is: ; Where D is the number of spare parts required, M is the highest value of the replacement frequency of various spare parts in all procurement cycles, S is the safety stock, and S is calculated by the following formula: ; Where n is the number of historical procurement cycles, is the mean of the historical purchasing cycle, K is a constant coefficient, and the value range of K is 0-0.2. is the value of the i-th historical procurement cycle.

[0015] As a preference, a data backup module is also included, which is responsible for backing up and storing data in the network device, wherein the data backup module includes: A11: Backup library, used to regularly back up and store data in network devices; A12: A temporary cache library is used to back up the vulnerable and important parts of the network equipment in real time, and transfer the data to the backup library regularly.

[0016] Beneficial effects of the present invention: 1. Compared with the existing technology, which usually adopts a fixed safety stock quantity or a simple statistical forecasting method to determine the inventory level, these methods often ignore the volatility of the procurement cycle and the dynamic changes in the actual demand for spare parts, resulting in a high risk of inventory backlog or shortage, and it is difficult to adapt to changes in the supply chain environment. This solution adopts a more advanced calculation method, which comprehensively considers multiple factors such as the volatility of the historical procurement cycle, daily maintenance needs, spare parts availability and supply chain reliability, and obtains the safety stock quantity through accurate calculation. This method not only improves the flexibility and accuracy of inventory management, effectively reduces inventory costs, but also ensures the timeliness and efficiency of maintenance work, significantly improving the overall operational efficiency; 2. Compared with the existing technology, which often relies on a single data storage medium or a simple backup strategy, lacks real-time and flexibility, and is difficult to cope with the high-frequency changes and sudden failures of data in network devices, resulting in long data recovery time and high risk of loss. This solution innovatively adopts a data backup module, in which the backup library is responsible for regularly backing up and storing the data in the network equipment to ensure the long-term preservation and traceability of the data; while the temporary cache library focuses on real-time backup of the vulnerable parts and important parts of the network equipment, and regularly transfers the data to the backup library, realizing rapid data recovery and efficient management. This solution not only significantly improves the real-time and reliability of data backup, but also greatly shortens the data recovery time, reduces the risk of data loss, and provides a strong guarantee for the data security of network devices; 3. By integrating modules such as data collection and monitoring, data analysis and prediction, intelligent fault diagnosis, maintenance plan optimization, work order management and execution, inventory and spare parts management, as well as data visualization and reporting, comprehensive real-time monitoring of network equipment status, accurate fault prediction and rapid response, intelligent optimization and efficient execution of maintenance plans, and intelligent management of spare parts inventory are achieved, which significantly improves the efficiency and quality of network equipment operation and maintenance, reduces operation and maintenance costs, and enhances the stability and security of the network system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Shown is a schematic diagram of the structure of the intelligent AI-driven network equipment self-checking and maintenance system of the present invention; Figure 2 What is shown is a workflow diagram of the inventory and spare parts management module in the intelligent AI-driven network equipment self-inspection and maintenance system of the present invention. DETAILED DESCRIPTION

[0018] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0019] See also Figure 1-2 The present invention provides an embodiment: an intelligent AI-driven network equipment self-checking and maintenance system, comprising: The data collection and monitoring module is used to collect the operating status data of network devices in real time through the Internet of Things technology, and monitor the network status, traffic and number of connections of network devices; Data analysis and prediction module, which is used to use big data analysis and machine learning algorithms to conduct in-depth mining and analysis of the collected data of network devices; Intelligent fault diagnosis module, which combines historical fault data and AI algorithms to make preliminary diagnosis of abnormal status of network equipment based on the data results of data analysis and prediction module; Maintenance plan optimization module, which is used to automatically generate and optimize maintenance plans based on the historical maintenance records and diagnostic data of the equipment; Work order management and execution module, used to realize online creation, allocation, tracking and closed-loop management of maintenance tasks; Inventory and spare parts management module, which is used to monitor spare parts inventory in real time, and intelligently generate procurement plans based on maintenance needs and forecast data to optimize spare parts inventory management; The data visualization and reporting module is used to integrate the system's work records and data and present them in a visual form.

[0020] As described above, the present invention realizes comprehensive real-time monitoring of network equipment status, accurate fault prediction and rapid response, intelligent optimization and efficient execution of maintenance plans, and intelligent management of spare parts inventory by integrating modules such as data collection and monitoring, data analysis and prediction, intelligent fault diagnosis, maintenance plan optimization, work order management and execution, inventory and spare parts management, and data visualization and reporting, which significantly improves the efficiency and quality of network equipment operation and maintenance, reduces operation and maintenance costs, and enhances the stability and security of the network system.

[0021] Preferably, the data acquisition and monitoring module includes: A11: Sensor network, including temperature sensors, humidity sensors, pressure sensors, displacement sensors and light sensors, which are used to monitor and measure the physical quantities of network devices and convert the collected physical quantities into electrical signals; A12: Data collector, used to receive signals from the sensor network and convert the received signals into a usable data format; A13: Surveillance camera, used to capture and record network devices and surrounding environment images in real time; A14: Data storage device, used to store the digital signals converted by the data collector.

[0022] As described above, the present invention integrates a sensor network (including multiple physical quantity sensors), a data collector, a monitoring camera and a data storage device, which can comprehensively monitor the physical environmental parameters of network equipment (such as temperature, humidity, pressure, displacement, and light), capture images of the equipment and its surrounding environment in real time, and efficiently convert these data into a usable format and store them securely, thereby realizing accurate monitoring of the status of network equipment and environmental adaptability assessment, effectively improving the intelligence level of operation and maintenance management and fault prevention capabilities.

[0023] Preferably, the data analysis and prediction module includes the following steps when using big data analysis and machine learning algorithms to perform in-depth mining and analysis on the collected data of network devices: S11: Data extraction, extracting historical operation data of network devices from the database, including performance indicators, device logs, and fault records of network devices; S12: Data cleaning and preprocessing, applying big data analysis and machine learning algorithms to clean the data and normalize, standardize and discretize the data; S13: Feature extraction, extracting useful features from the cleaned and preprocessed data, where the algorithms used include principal component analysis, trend analysis algorithm, and seasonal analysis algorithm; S14: Model construction: first build a prediction model and set the model parameters, then use historical data to train the model; S15: Model application, inputting the data collected by the data acquisition and monitoring module into the trained model, and the model outputs the prediction results.

[0024] As described above, the present invention uses systematic steps, including data extraction, cleaning and preprocessing, feature extraction, model building and application, to deeply mine the historical operation data of network equipment using big data analysis and machine learning algorithms, thereby effectively improving the accuracy and efficiency of data processing, and can accurately predict the performance trends and potential failures of network equipment, providing strong data support for the preventive maintenance and optimization management of network equipment, and significantly enhancing the intelligence level of operation and maintenance decision-making.

[0025] Preferably, the intelligent fault diagnosis module combines historical fault data and AI algorithms to make a preliminary diagnosis of the abnormal state of the network device according to the data results of the data analysis and prediction module, including the following steps: S21: Data reception, receiving the output results from the data analysis and prediction module, including real-time operation status data of network devices, abnormal status indications, and possible fault warning information, and preprocessing the output results; S22: Fault feature matching: Combine historical fault data and use AI algorithms to perform feature matching on the current status of the device; S23: Based on the result of feature matching, AI algorithm is applied to perform fault diagnosis and prediction, and the fault type corresponding to the current state is identified. The fault types include hardware fault, software fault and network fault. S24: Fault analysis: after identifying the fault type, the cause of the fault is analyzed in one step, wherein the fault causes include device aging, configuration errors and network attacks.

[0026] Preferably, when the intelligent fault diagnosis module uses the AI ​​algorithm to perform feature matching on the current state of the device, the AI ​​algorithm used is a deep learning algorithm, wherein the network structure of the deep learning algorithm includes an input layer, a hidden layer and an output layer, wherein the mapping from the input layer to the hidden layer is: ; The mapping from hidden layer to output layer is: ; Among them, x is the input data, is the activation value of the hidden layer, y is the output value, and are the weight matrices from the input layer to the hidden layer and from the hidden layer to the output layer, respectively. and are the bias terms of the hidden layer and the output layer, respectively. is the activation function, and T represents the transpose operation of the matrix.

[0027] As described above, the present invention receives and pre-processes the output results of the data analysis and prediction module, combines historical fault data, and uses advanced AI algorithms to perform fault feature matching and diagnostic prediction. It can quickly identify the hardware, software and network fault types of network equipment, and further analyze the causes of the faults, such as equipment aging, configuration errors or network attacks, thereby achieving accurate preliminary diagnosis of abnormal states of network equipment, significantly improving the efficiency and accuracy of troubleshooting, and providing strong support for rapid response and repair.

[0028] Preferably, the work order management and execution module includes the following steps when realizing online creation, allocation, tracking and closed-loop management of maintenance tasks: S31: receiving the maintenance plan transmitted by the maintenance plan optimization module; S32: Creating a maintenance task online, wherein the maintenance task includes a task description, a task priority, and a task executor; S33: Allocate maintenance tasks to corresponding technicians and teams; S34: Tracking the execution status of the maintenance task, the execution status of the maintenance task includes the task progress and task results, wherein the task progress is recorded by image recording; S35: Implement closed-loop management of maintenance tasks, including task completion confirmation and task feedback.

[0029] As described above, the present invention realizes the creation, precise allocation, execution tracking and closed-loop management of online maintenance tasks by receiving maintenance plans, and uses images to record task progress to ensure task transparency and execution efficiency. At the same time, through the task completion confirmation and feedback mechanism, the management and control of the maintenance process is strengthened, which significantly improves the response speed, execution quality and overall operation and maintenance efficiency of maintenance work.

[0030] As a preferred embodiment, the inventory and spare parts management module monitors the spare parts inventory in real time, and intelligently generates a procurement plan based on maintenance needs and forecast data to optimize spare parts inventory management, including the following steps: S41: real-time monitoring of spare parts inventory, wherein the spare parts inventory includes spare parts types, quantities and storage locations; S42: Forecast maintenance needs and intelligently generate procurement plans based on maintenance needs; S43: Send the purchase plan to the relevant staff for review to ensure that the purchased spare parts meet the actual needs; S44: Purchase spare parts according to the purchase plan and track the purchase progress in real time.

[0031] As described above, the present invention monitors the types, quantities and storage locations of spare parts in real time, accurately predicts maintenance needs and intelligently generates procurement plans, executes procurement after review by staff, and tracks procurement progress in real time, thereby achieving efficient management of spare parts inventory and optimization of procurement processes, ensuring the timeliness and accuracy of spare parts supply, effectively reducing inventory costs, and improving the intelligence level of spare parts management.

[0032] Preferably, when predicting maintenance needs, the following steps are included: S51: collecting historical data, wherein the collected historical data includes spare parts types, spare parts replacement time, spare parts replacement frequency and spare parts procurement cycle; S52: Calculate the replacement frequency of spare parts within a procurement cycle based on historical data, and then identify the highest value of the replacement frequency of various types of spare parts in all procurement cycles, wherein a moving average method is used for prediction; S53: Determine safety stock: Determine the safety stock level of various spare parts based on the historical maximum replacement frequency and supply chain reliability during the procurement cycle; S54: Calculate the required quantity of spare parts, add the daily maintenance demand and the safety stock, and obtain the total required quantity of each type of spare parts.

[0033] As a preferred method, when predicting maintenance needs, the principle formula is: ; Where D is the number of spare parts required, M is the highest value of the replacement frequency of various spare parts in all procurement cycles, S is the safety stock, and S is calculated by the following formula: ; Where n is the number of historical procurement cycles, is the mean of the historical purchasing cycle, K is a constant coefficient, and the value range of K is 0-0.2. is the value of the i-th historical procurement cycle.

[0034] As mentioned above, compared with the existing technology, which usually adopts a fixed safety stock quantity or a simple statistical-based forecasting method to determine the inventory level, these methods often ignore the volatility of the procurement cycle and the dynamic changes in the actual demand for spare parts, resulting in a high risk of inventory backlog or shortage, and it is difficult to adapt to changes in the supply chain environment. This solution adopts a more advanced calculation method, which comprehensively considers multiple factors such as the volatility of historical procurement cycles, daily maintenance needs, spare parts availability and supply chain reliability, and obtains the safety stock quantity through precise calculation. This method not only improves the flexibility and accuracy of inventory management, effectively reduces inventory costs, but also ensures the timeliness and efficiency of maintenance work, and significantly improves the overall operational efficiency.

[0035] As a preference, a data backup module is also included, which is responsible for backing up and storing data in the network device, wherein the data backup module includes: A11: Backup library, used to regularly back up and store data in network devices; A12: A temporary cache library is used to back up the vulnerable and important parts of the network equipment in real time, and transfer the data to the backup library regularly.

[0036] As described above, compared with the existing technology, which often relies on a single data storage medium or a simple backup strategy, lacks real-time and flexibility, and is difficult to cope with high-frequency changes and sudden failures of data in network devices, resulting in long data recovery time and high risk of loss, this solution innovatively adopts a data backup module, in which the backup library is responsible for regularly backing up and storing data in network devices to ensure long-term preservation and traceability of data; while the temporary cache library focuses on real-time backup of data in the prone to failure parts and important parts of network devices, and regularly transfers data to the backup library, realizing rapid data recovery and efficient management. This solution not only significantly improves the real-time and reliability of data backup, but also greatly shortens data recovery time, reduces the risk of data loss, and provides strong protection for the data security of network devices.

[0037] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge scope of those skilled in the art without departing from the spirit of the present invention.

Claims

1. Intelligent AI-driven network equipment self-checking and maintenance system; its features are: Included are: The data collection and monitoring module is used to collect the operating status data of network devices in real time through the Internet of Things technology, and monitor the network status, traffic and number of connections of network devices; Data analysis and prediction module, which is used to use big data analysis and machine learning algorithms to conduct in-depth mining and analysis of the collected data of network devices; Intelligent fault diagnosis module, which combines historical fault data and AI algorithms to make preliminary diagnosis of abnormal status of network equipment based on the data results of data analysis and prediction module; Maintenance plan optimization module, which is used to automatically generate and optimize maintenance plans based on the historical maintenance records and diagnostic data of the equipment; Work order management and execution module, used to realize online creation, allocation, tracking and closed-loop management of maintenance tasks; Inventory and spare parts management module, which is used to monitor spare parts inventory in real time, and intelligently generate procurement plans based on maintenance needs and forecast data to optimize spare parts inventory management; The data visualization and reporting module is used to integrate the system's work records and data and present them in a visual form.

2. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 1, characterized in that: The data acquisition and monitoring modules include: A11: Sensor network, including temperature sensors, humidity sensors, pressure sensors, displacement sensors and light sensors, which are used to monitor and measure the physical quantities of network devices and convert the collected physical quantities into electrical signals; A12: Data collector, used to receive signals from the sensor network and convert the received signals into a usable data format; A13: Surveillance camera, used to capture and record network devices and surrounding environment images in real time; A14: Data storage device, used to store the digital signals converted by the data collector.

3. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 2, characterized in that: The data analysis and prediction module uses big data analysis and machine learning algorithms to conduct in-depth mining and analysis of the collected data from network devices, including the following steps: S11: Data extraction, extracting historical operation data of network devices from the database, including performance indicators, device logs, and fault records of network devices; S12: Data cleaning and preprocessing, applying big data analysis and machine learning algorithms to clean the data and normalize, standardize and discretize the data; S13: Feature extraction, extracting useful features from the cleaned and preprocessed data, where the algorithms used include principal component analysis, trend analysis algorithm, and seasonal analysis algorithm; S14: Model construction: first build a prediction model and set the model parameters, then use historical data to train the model; S15: Model application, inputting the data collected by the data acquisition and monitoring module into the trained model, and the model outputs the prediction results.

4. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 3, characterized in that: The intelligent fault diagnosis module combines historical fault data and AI algorithms to make a preliminary diagnosis of abnormal status of network devices based on the data results of the data analysis and prediction module, including the following steps: S21: Data reception, receiving the output results from the data analysis and prediction module, including real-time operation status data of network devices, abnormal status indications, and possible fault warning information, and preprocessing the output results; S22: Fault feature matching: Combine historical fault data and use AI algorithms to perform feature matching on the current status of the device; S23: Based on the result of feature matching, AI algorithm is applied to perform fault diagnosis and prediction, and the fault type corresponding to the current state is identified. The fault types include hardware fault, software fault and network fault. S24: Fault analysis: after identifying the fault type, the cause of the fault is analyzed in one step, wherein the fault causes include device aging, configuration errors and network attacks.

5. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 4 is characterized in that: When the intelligent fault diagnosis module uses the AI ​​algorithm to perform feature matching on the current state of the equipment, the AI ​​algorithm used is a deep learning algorithm, where the network structure of the deep learning algorithm includes an input layer, a hidden layer, and an output layer, where the mapping from the input layer to the hidden layer is: ； The mapping from hidden layer to output layer is: ； Among them, x is the input data, is the activation value of the hidden layer, y is the output value, and are the weight matrices from the input layer to the hidden layer and from the hidden layer to the output layer, respectively. and are the bias terms of the hidden layer and the output layer, respectively. is the activation function, and T represents the transpose operation of the matrix.

6. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 5, characterized in that: The work order management and execution module includes the following steps when implementing online creation, allocation, tracking and closed-loop management of maintenance tasks: S31: receiving the maintenance plan transmitted by the maintenance plan optimization module; S32: Creating a maintenance task online, wherein the maintenance task includes a task description, a task priority, and a task executor; S33: Allocate maintenance tasks to corresponding technicians and teams; S34: Tracking the execution status of the maintenance task, the execution status of the maintenance task includes the task progress and task results, wherein the task progress is recorded by image recording; S35: Implement closed-loop management of maintenance tasks, including task completion confirmation and task feedback.

7. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 6, characterized in that: The inventory and spare parts management module monitors the spare parts inventory in real time, and intelligently generates procurement plans based on maintenance needs and forecast data to optimize spare parts inventory management. The following steps are included: S41: real-time monitoring of spare parts inventory, wherein the spare parts inventory includes spare parts types, quantities and storage locations; S42: Forecast maintenance needs and intelligently generate procurement plans based on maintenance needs; S43: Send the purchase plan to the relevant staff for review to ensure that the purchased spare parts meet the actual needs; S44: Purchase spare parts according to the purchase plan and track the purchase progress in real time.

8. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 7, characterized in that: When forecasting maintenance needs, the following steps are involved: S51: collecting historical data, wherein the collected historical data includes spare parts types, spare parts replacement time, spare parts replacement frequency and spare parts procurement cycle; S52: Calculate the replacement frequency of spare parts within a procurement cycle based on historical data, and then identify the highest value of the replacement frequency of various types of spare parts in all procurement cycles, wherein a moving average method is used for prediction; S53: Determine safety stock: Determine the safety stock level of various spare parts based on the historical maximum replacement frequency and supply chain reliability during the procurement cycle; S54: Calculate the required quantity of spare parts, add the daily maintenance demand and the safety stock, and obtain the total required quantity of each type of spare parts.

9. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 8, characterized in that: When predicting maintenance needs, the principle formula is: ； Where D is the number of spare parts required, M is the highest value of the replacement frequency of various spare parts in all procurement cycles, S is the safety stock, and S is calculated by the following formula: ； Where n is the number of historical procurement cycles, is the mean of the historical purchasing cycle, K is a constant coefficient, and the value range of K is 0-0.

2. is the value of the i-th historical procurement cycle.

10. The intelligent AI-driven network equipment self-checking and maintenance system according to claim 9, characterized in that: It also includes a data backup module, which is responsible for the backup and storage of data in the network device, wherein the data backup module includes: A11: Backup library, used to regularly back up and store data in network devices; A12: A temporary cache library is used to back up the vulnerable and important parts of the network equipment in real time, and transfer the data to the backup library regularly.

Images