Electric power communication network fault prediction and diagnosis system

Abstract

The invention discloses an electric power communication network fault prediction and diagnosis system. The system comprises a data acquisition module, a fault diagnosis module and a prediction analysis module. Wherein the data acquisition module is used for acquiring monitoring data in real time and then managing the transmission rate and priority of the data by using a sliding window protocol; the fault diagnosis module is used for detecting abnormal data deviating from a normal range through an adaptive data analysis algorithm; and the prediction analysis module is used for predicting faults occurring in the system through a dynamic prediction algorithm. According to the invention, a sliding window protocol is adopted to optimize data transmission, the depth and breadth of fault diagnosis are improved through an adaptive data analysis algorithm, and the prediction accuracy is enhanced by using a dynamic prediction algorithm, so that the system response speed is accelerated, the fault diagnosis coverage is expanded, and the prediction precision is improved. And the safe and stable operation capability of the electric power communication network is obviously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of communication network fault monitoring, and more particularly to a power communication network fault prediction and diagnosis system. Background Art

[0002] The power communication network is an important infrastructure to ensure the stable operation of the power system and the reliability of information transmission. The fault prediction and diagnosis system technology is of great significance to ensure the reliability of the power communication system.

[0003] At present, with the development of technologies such as big data, cloud computing, and artificial intelligence, the power communication network fault prediction and diagnosis system has gradually shifted from the traditional rule-based method to the data-driven method. The advancement of science and technology has shown great advantages in processing large amounts of complex data and discovering implicit relationships between data, which has significantly improved the prediction and diagnosis of power communication network faults;

[0004] However, the prediction and diagnosis system of power communication network faults still has problems such as slow system response speed, insufficient fault diagnosis and low prediction accuracy.

[0005] Therefore, how to further improve the above problems and enhance the fault prediction and diagnosis effects is an issue that technical personnel in this field need to solve urgently. Summary of the invention

[0006] In view of this, the present invention provides a power communication network fault prediction and diagnosis system to solve the problems of poor system environmental adaptability, low data processing capability and poor system monitoring efficiency in the prior art.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A power communication network fault prediction and diagnosis system includes a data acquisition module, a fault diagnosis module and a prediction analysis module; wherein:

[0009] The data acquisition module is used to collect monitoring data in real time and then manage the transmission rate and priority of the data using a sliding window protocol;

[0010] The fault diagnosis module is used to detect abnormal data that deviates from a normal range through an adaptive data analysis algorithm;

[0011] The prediction analysis module is used to predict system failures through a dynamic prediction algorithm.

[0012] Preferably, detecting abnormal data that deviates from a normal range by an adaptive data analysis algorithm comprises the following steps:

[0013] Construct a fault diagnosis model based on autoencoder,

[0014] Extract features from raw data through time series analysis and recursive feature elimination to train the fault diagnosis model;

[0015] The trained fault diagnosis model is used to detect anomalies in real-time data. When the real-time data differs significantly from the model expectations, the data is marked as abnormal. The formula of the anomaly detection function is:

[0016]

[0017] In the formula, Y represents the anomaly detection function, λ represents the monitoring factor of the fault diagnosis model, δ represents the real-time data to be detected in the power communication network, ρ represents the expected value of the fault diagnosis model, n represents the order of termination of the real-time data, and tan 2 (δ+ρ) represents the evaluation function detection coefficient.

[0018] As a preference, the trained fault diagnosis model is evaluated, and the evaluation function is:

[0019]

[0020] In the formula, P represents the evaluation function, κ represents the error factor, α i represents the predicted value of the power communication network, β i represents the real value of the power communication network, i represents the data type predicted by the power communication network, Represents the function evaluation coefficients.

[0021] Preferably, the dynamic prediction algorithm comprises the following steps:

[0022] Build a dynamic prediction model based on time series ARIMA and perform dynamic data prediction after training;

[0023] The prediction result verification function is used to analyze and verify the prediction result; the formula of the prediction result verification function is expressed as:

[0024]

[0025] Where, T c represents the prediction result verification function, θ represents the error value of the model data, ε i Represents the predicted value output by the dynamic prediction model, μ i represents the true value of the power communication network; i represents the order in which the dynamic prediction model is verified;

[0026] Set monitoring thresholds for data anomalies and identify potential faults based on the monitoring thresholds and prediction results.

[0027] As a preferred embodiment, the dynamic prediction model is adjusted according to the feedback of the real-time status data value of the power communication network, and the feedback function expression is:

[0028]

[0029] In the formula, F represents the feedback function, ω represents the function comparison factor, represents the prediction result value of the dynamic prediction model, ξ represents the real-time status data value of the power communication network, Represents the feedback function coefficient.

[0030] Preferably, the data acquisition module comprises a sensor monitoring unit, a data processing unit, a data flow control unit and a communication unit connected in sequence; wherein the data flow control unit is used to manage the transmission rate and priority of the data using a sliding window protocol after real-time acquisition of monitoring data;

[0031] The sensor monitoring unit is used to deploy IoT sensors on nodes and devices of the power communication network to collect network status parameters;

[0032] A data processing unit, used to process the collected data by cleaning, filtering and format conversion;

[0033] The communication unit uses Wi-Fi wireless communication to achieve secure data transmission.

[0034] Preferably, the fault diagnosis module comprises a data anomaly detection unit, a fault identification unit and a fault location unit which are connected in sequence; wherein the data anomaly detection unit is used to detect abnormal data deviating from a normal range through an adaptive data analysis algorithm;

[0035] A fault identification unit, used to identify and classify the specific type of fault through random forest;

[0036] The fault location unit locates the faulty node through network topology and device status information.

[0037] Preferably, the prediction and analysis module comprises a real-time prediction and analysis unit, a risk assessment unit and an alarm unit which are connected in sequence; wherein the real-time prediction and analysis unit is used to predict the failure of the system through a dynamic prediction algorithm;

[0038] Risk assessment unit, used to assess the impact of failures on the network through failure frequency, historical impact data and network criticality data;

[0039] An alarm unit is used to send a fault alarm to the control terminal through a threshold trigger mechanism.

[0040] Preferably, the fault prediction and diagnosis system further comprises a control terminal connected to the data acquisition module, the fault diagnosis module and the prediction and analysis module.

[0041] The control terminal includes a system status monitoring unit, a task coordination unit and a system configuration unit;

[0042] System status monitoring unit, used to display network status data to operators through a visual interface to monitor the operating status of the system in real time;

[0043] The task coordination unit implements task scheduling and reasonable resource allocation through the distributed task scheduling framework Celery;

[0044] The system configuration unit is used to configure and modify system parameters through the configuration management database CMDB.

[0045] Preferably, the fault prediction and diagnosis system further comprises a human-computer interaction module, wherein the human-computer interaction module comprises a graphical user interface, a push notification unit and a query feedback unit;

[0046] A graphical user interface, which is used to provide users with a visual interface through a liquid crystal display to display system status, fault information and prediction results;

[0047] Push notification unit, used to push power communication network information to operation and maintenance personnel via SMS and instant messaging tools;

[0048] The query feedback unit is used to query the system detailed information through the non-relational database MongoDB.

[0049] Preferably, the fault prediction and diagnosis system further comprises a rights management module, which comprises a user identity authentication unit, a data protection unit and an access record unit;

[0050] User authentication unit, used to verify user identity through fingerprint and facial recognition;

[0051] A data protection unit is used to encrypt data using the AES data encryption standard to ensure data security during transmission and storage;

[0052] The access logging unit is used to record detailed logs of user access and operations through the log management system ELK Stack for auditing and security analysis.

[0053] It can be seen from the above technical solutions that the present invention discloses a power communication network fault prediction and diagnosis system. Compared with the prior art, the present invention manages the transmission rate and priority of data through a sliding window protocol, thereby improving the system response speed; processes and analyzes data information through an adaptive data analysis algorithm, thereby improving the depth and breadth of fault diagnosis; and continuously learns and updates the model from new data through a dynamic prediction algorithm, thereby improving the accuracy of system prediction, thereby achieving faster system response speed, expanded fault diagnosis coverage, and improved prediction accuracy, and significantly improving the safe and stable operation capability of the power communication network, providing a strong guarantee for the reliability of the power system and the continuity of information transmission.

[0054] Other advantages, objectives and features of the present invention will be described in part in the following description; and in part, will be apparent to those skilled in the art based on an examination of the following; or, may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0056] Figure 1 It is a structural schematic diagram of the data acquisition module of the present invention;

[0057] Figure 2 It is a structural schematic diagram of the fault diagnosis module of the present invention;

[0058] Figure 3 It is a structural schematic diagram of the prediction and analysis module of the present invention;

[0059] Figure 4 It is a schematic diagram of the overall structure of the present invention;

[0060] Figure 5 It is a structural schematic diagram of the control terminal of the present invention;

[0061] Figure 6 It is a structural schematic diagram of the human-computer interaction module of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] The embodiment of the present invention discloses a power communication network fault prediction and diagnosis system, which mainly includes a data acquisition module, a fault diagnosis module and a prediction and analysis module; wherein the data acquisition module is used to collect monitoring data in real time and then manage the transmission rate and priority of the data using a sliding window protocol; the fault diagnosis module is used to detect abnormal data that deviates from a normal range through an adaptive data analysis algorithm; the prediction and analysis module is used to predict faults that may occur in the system through a dynamic prediction algorithm.

[0064] In one embodiment, the data acquisition module is used to provide real-time and historical data to the system; Figure 1, including a sensor monitoring unit, a data processing unit, a data flow control unit and a communication unit, wherein the sensor monitoring unit deploys IoT sensors on nodes and devices of the power communication network to realize the collection of network status parameters, monitor the pulse of the network in real time, and capture the changes in network status parameters. These sensors can not only sense the temperature and humidity of the environment, but also monitor the operating status of the equipment to ensure the health and safety of the power communication network; the data processing unit processes the collected data through cleaning, filtering and format conversion. After receiving the raw data collected by the sensor monitoring unit, the data processing unit immediately starts data cleaning, filtering and format conversion. Data cleaning removes useless noise and errors to ensure the purity of the data; data filtering is accurate positioning, retaining only valuable information; and format conversion allows data to be saved in a unified format to ensure the efficiency and accuracy of subsequent processing; the data flow control unit manages the data transmission rate and priority through the sliding window protocol, and cleverly regulates the data transmission rate and priority. This not only avoids data congestion and ensures timely data transmission, but also adjusts the transmission priority according to the importance of the data, so that key information can always be delivered first; the communication unit uses Wi-Fi wireless communication to achieve secure data transmission. Wi-Fi wireless communication not only provides a high-speed data transmission channel, but also ensures the security of data during transmission through encryption and other means. These four units work closely together to build an efficient, safe and intelligent data acquisition and transmission system, providing solid data support for the stable operation of the power communication network. The output end of the sensor monitoring unit is connected to the input end of the data processing unit, the output end of the data processing unit is connected to the input end of the data flow control unit, and the output end of the data flow control unit is connected to the input end of the communication unit.

[0065] In one embodiment, the fault diagnosis module is used to quickly and accurately identify problems that occur in the network and provide specific directions for predictive analysis; specifically, Figure 2, including a data anomaly detection unit, a fault identification unit and a fault location unit. The data anomaly detection unit detects abnormal data and potential faults that deviate from the normal range through an adaptive data analysis algorithm. The adaptive data analysis algorithm always monitors the subtle changes in the system data flow. When the data deviates from the preset normal range, it can timely monitor the abnormal situation of the data. Through continuous analysis, the unit can quickly capture abnormal data and even identify potential warning signals before a fault occurs, providing key clues for subsequent diagnostic work; the fault identification unit identifies and classifies the specific type of fault through random forest. The random forest algorithm constructs multiple decision trees to conduct multi-dimensional evaluation and comparison of fault modes, thereby accurately identifying the type of fault, including hardware fault, software fault or network fault, etc., to ensure the accuracy and comprehensiveness of fault identification. The powerful machine learning algorithm of random forest is used to conduct in-depth analysis and classification of the specific types of faults; the fault location unit locates the faulty node through network topology and device status information. By analyzing the physical connection and logical relationship of the network and combining real-time device status data, the unit can quickly lock the source of the fault. Whether it is a node in the network or a specific hardware device, it can be quickly and accurately identified, providing a direct target for subsequent fault repair. Through the close collaboration of these three units, the fault diagnosis module can achieve comprehensive monitoring, accurate identification and rapid positioning of system faults, greatly shortening the fault handling time, improving the stability and reliability of the system, and providing a solid guarantee for the efficient operation of the industrial network. The output end of the data anomaly detection unit is connected to the input end of the fault identification unit, and the output end of the fault identification unit is connected to the input end of the fault positioning unit.

[0066] In a preferred embodiment, the working steps of the adaptive data analysis algorithm are as follows:

[0067] a) Feature selection and extraction

[0068] Extract meaningful features from raw data through time series analysis and recursive feature elimination;

[0069] b) Model training and evaluation

[0070] The historical data information is trained by a fault diagnosis model based on an autoencoder to identify abnormal data points that exceed the normal range in the power communication system, and the accuracy of the fault diagnosis model is evaluated according to the evaluation function. The formula expression of the evaluation function is:

[0071]

[0072] In formula (1), P represents the evaluation function, κ represents the error factor, and α irepresents the predicted value of the power communication network, β i represents the real value of the power communication network, i represents the data type predicted by the power communication network, represents the function evaluation coefficient;

[0073] c) Anomaly Detection

[0074] The real-time data is input into the trained fault diagnosis model, and the data is detected according to the anomaly detection function. When the real-time data differs significantly from the model expectation, the data is marked as abnormal. The formula expression of the anomaly detection function is:

[0075]

[0076] In formula (2), Y represents the anomaly detection function, λ represents the monitoring factor of the fault diagnosis model, δ represents the real-time data to be detected in the power communication network, ρ represents the expected value of the fault diagnosis model, n represents the order of termination of the real-time data, and tan 2 (δ+ρ) represents the detection coefficient of the evaluation function;

[0077] d) Model update and optimization

[0078] Updating the fault diagnosis model according to the new data to enable the adaptive data analysis algorithm to self-learn and adjust as the network environment changes;

[0079] e) Result output and feedback

[0080] The fault diagnosis result is presented to the operator through the graphical user interface so that the operator can take action.

[0081] The functions of the adaptive data analysis algorithm are: it can automatically adjust the analysis strategy and model parameters in a constantly changing environment to maintain the accuracy of the analysis results; it can identify and diagnose abnormal conditions of equipment or networks, predict potential failures, and achieve early warning; it provides real-time data analysis and decision support to ensure stable system operation. The adaptive data analysis algorithm solves the problems of high data complexity, unpredictable environmental changes, and slow resource updates. The adaptive data analysis algorithm processes large-scale, high-dimensional, and nonlinear data sets to extract valuable information; it adapts to dynamic environments such as equipment aging and network load changes in the power communication system, and continuously optimizes fault diagnosis and prediction; through intelligent analysis, it rationally allocates maintenance resources, reduces unnecessary maintenance costs, and improves system operation efficiency. Through the error factor κ, the power communication network prediction value α i , the real value of the power communication network β i and function evaluation coefficients Generate an evaluation function formula to evaluate the accuracy of the fault diagnosis model; through the fault diagnosis model monitoring factor λ, the real-time data to be detected of the power communication network δ, the expected value ρ of the fault diagnosis model and the function detection coefficient tan 2 (δ+ρ) is used to obtain the anomaly detection function formula and detect the data. At the same time, the differences between the adaptive data analysis algorithm and the traditional data analysis algorithm are shown in Table 1:

[0082] Table 1

[0083] In summary, the adaptive data analysis algorithm is superior to the traditional algorithm in both detection efficiency and learning efficiency. It can be seen that the adaptive data analysis algorithm is the best choice for the present invention.

[0084] In one embodiment, the predictive analysis module is used to warn of potential problems in advance, implement preventive maintenance, and reduce the occurrence of failures; Figure 3 , including a real-time prediction and analysis unit, a risk assessment unit and an alarm unit. The real-time prediction and analysis unit predicts the faults that may occur in the system through a dynamic prediction and analysis algorithm, and can gain insight into any abnormalities or faults that may occur in the system, ensuring that the problem is discovered in time for processing; the risk assessment unit evaluates the impact of the fault on the network through the fault frequency, historical impact data and network criticality data. The risk assessment unit not only manages the frequency of the fault, but also analyzes the historical impact data and network criticality data in depth, so as to evaluate the impact of each predicted fault on the entire network, providing decision makers with a quantitative risk perspective, so that each response can be timely; the alarm unit sends a fault alarm to the control terminal through a threshold trigger mechanism. Once the risk assessment unit determines that the potential impact of the fault exceeds the preset warning line, the alarm unit will act quickly and send a timely and accurate fault alarm to the control terminal. This mechanism ensures that the information is transmitted efficiently and accurately, wins valuable response time for the operation and maintenance personnel, and effectively avoids the further expansion of potential risks. Through the close cooperation of these three units, the prediction and analysis module can not only predict faults, assess risks, but also provide immediate warnings, forming a complete set of fault prediction and response mechanisms, providing a solid guarantee for the stable operation of the network. The output end of the real-time prediction and analysis unit is connected to the input end of the risk assessment unit, and the output end of the risk assessment unit is connected to the input end of the alarm unit.

[0085] In a preferred embodiment, the steps for implementing the dynamic prediction algorithm are as follows:

[0086] a) Data collection and preprocessing

[0087] Continuously collect network operation data, including equipment status and performance indicators, and standardize and normalize the collected data to ensure data quality in preparation for model input;

[0088] b) Detection threshold setting and feature extraction

[0089] Analyze historical data to determine the normal data range, set the monitoring threshold for data anomalies, and perform feature selection to eliminate irrelevant and redundant features;

[0090] c) Model selection and updating

[0091] Input historical data into the dynamic forecasting model based on time series ARIMA for model training, and continuously optimize it to adapt to data changes;

[0092] d) Dynamic data prediction

[0093] Inputting real-time data into the dynamic prediction model to perform real-time prediction to generate future trends and abnormal probabilities of power communication network data;

[0094] e) Analysis and verification of prediction results

[0095] The prediction result verification function is used to analyze and verify the prediction data information output by the model. The formula of the prediction result verification function is expressed as:

[0096]

[0097] In formula (3), T c represents the prediction result verification function, θ represents the error value of the model data, ε i Represents the predicted value output by the dynamic prediction model, μ i represents the true value of the power communication network; i represents the order in which the dynamic prediction model is verified;

[0098] f) Model parameter adjustment

[0099] The model prediction results are compared with the actual situation of the power communication network according to the feedback function to adjust the model parameters through feedback. The formula expression of the feedback function is:

[0100]

[0101] In formula (4), F represents the feedback function, ω represents the function comparison factor, represents the prediction result value of the dynamic prediction model, ξ represents the real-time status data value of the power communication network, represents the coefficient of the feedback function;

[0102] g) Threshold setting and continuous updating

[0103] An abnormality detection threshold is set according to the prediction results of the dynamic prediction model to identify potential faults, and the above steps are repeated to achieve continuous learning and optimization of the dynamic prediction model.

[0104] The role of the dynamic prediction algorithm is to monitor and predict the operating status of the power communication network in real time, including equipment status, performance indicators, etc., to achieve accurate prediction of the future trend of the network; by setting the detection threshold, the algorithm can identify potential abnormal conditions in the network, issue early warnings, and reduce the probability of failure; through the prediction results, the allocation and use of network resources can be optimized to improve network operation efficiency and stability. The dynamic prediction algorithm solves the complexity and uncertainty of the power communication network, identifies possible failures in advance, reduces the impact of sudden failures on the network, reasonably allocates network resources according to the prediction results, improves resource utilization, and uses the error value θ of the model data and the prediction value ε output by the dynamic prediction model. i and the true value μ of the power communication network i Generate a prediction result verification function to analyze and verify the prediction data information output by the model; use the function comparison factor ω and the dynamic prediction model to predict the result value Real-time state data value ξ and function coefficient of power communication network The feedback function formula is obtained, and the model prediction results are compared with the actual situation of the power communication network to adjust the model parameters. At the same time, the differences between the dynamic prediction algorithm and the traditional algorithm are shown in Table 2:

[0105] Table 2

[0106]

[0107] In summary, the dynamic prediction algorithm is superior to the traditional algorithm in terms of prediction accuracy. Therefore, it can be seen that the dynamic prediction algorithm is the best choice for the present invention.

[0108] In order to further optimize the above technical solution, the fault prediction and diagnosis system of the present application also includes a control terminal for ensuring that the system runs efficiently and orderly and responds to various operating instructions and status changes in a timely manner.

[0109] like Figure 4 The output end of the control terminal is electrically connected to the input ends of the data acquisition module, the fault diagnosis module, the prediction and analysis module, the human-computer interaction module and the authority management module; the output end of the data acquisition module is electrically connected to the input end of the fault diagnosis module; the output end of the fault diagnosis module is electrically connected to the input end of the prediction and analysis module; the output end of the prediction and analysis module is electrically connected to the input end of the human-computer interaction module; the output end of the human-computer interaction module is electrically connected to the input end of the authority management module.

[0110] Further, if Figure 5 , the control terminal includes a system status monitoring unit, a task coordination unit and a system configuration unit, which together build an efficient and intelligent management platform. The system status monitoring unit displays network status data to the operator through a visual interface, providing the operator with a real-time updated system operation status screen, including key indicators such as network status data, resource utilization, and system health index, so that the operator can quickly identify potential problems and achieve a comprehensive grasp of the system status. This unit not only provides real-time monitoring, but also has a historical data backtracking function to help analyze system performance trends and provide data support for decision-making to monitor the operating status of the system in real time; the task coordination unit implements task scheduling and reasonable allocation of resources through the distributed task scheduling framework Celery. The task coordination unit can automatically identify task priorities and dynamically adjust execution strategies according to system loads to ensure that high-priority tasks are processed in a timely manner while avoiding resource waste. Through intelligent algorithms, the unit can also predict task execution time, optimize resource allocation, and improve overall operating efficiency. In addition, it also supports cross-node task scheduling, which enhances the scalability and fault tolerance of the system; the system configuration unit configures and modifies system parameters through the configuration management database CMDB. Operators can easily adjust system parameters in this unit, including resource thresholds, network configurations, security policies, etc., without restarting the system, which greatly improves the flexibility and response speed of the system. CMDB also records every configuration change for easy tracking and auditing, ensuring the consistency and security of system configuration. In summary, the control terminal not only realizes real-time monitoring of system status through the collaborative work of the three major units, ensuring efficient task scheduling and reasonable resource allocation, but also provides flexible system configuration management, providing operators with a powerful and intuitive control interface, and significantly improving the convenience and intelligence of system management.

[0111] In order to further optimize the above technical solution, the fault prediction and diagnosis system of the present application also includes a human-computer interaction module to enhance the operability and information transparency of the system and improve the response efficiency;

[0112] Among them, Figure 6The human-computer interaction module includes a graphical user interface, a push notification unit and a query feedback unit. The graphical user interface provides a visual interface for users through a liquid crystal display to display system status, fault information and prediction results, so as to ensure that operation and maintenance personnel can quickly locate key information and make timely responses. In addition, the graphical user interface also supports personalized customization. Users can adjust the priority of information display according to their needs to ensure that important information is highlighted and improve operational efficiency; the push notification unit pushes power communication network information to operation and maintenance personnel through text messages and instant messaging tools. Whether it is a change in system status or an instant report of a fault, it can be delivered to the operation and maintenance personnel at the first time to ensure that they can respond quickly, effectively shorten the fault handling time, and ensure the stable operation of the power communication network; the query feedback unit queries the system detailed information through the non-relational database MongoDB, and the system provides fast and accurate data retrieval services. With its flexible document storage method and efficient query performance, MongoDB can quickly process a large number of complex queries and provide real-time query services for system detailed information for operation and maintenance personnel. Whether it is troubleshooting or in-depth analysis of the operating status, operation and maintenance personnel can quickly obtain the required information through the query feedback unit to provide strong support for decision-making. Through the close collaboration of these three units, the human-computer interaction module not only improves the operation and maintenance efficiency of the power communication network, but also provides a more friendly and efficient user experience for operation and maintenance personnel through intelligent interaction design. It is an indispensable intelligent center of the power communication network system.

[0113] In order to further optimize the above technical solution, the fault prediction and diagnosis system of the present application also includes a rights management module to enhance the operability and information transparency of the system and improve the response efficiency;

[0114] Among them, the authority management module includes a user identity authentication unit, a data protection unit and an access record unit. The user identity authentication unit verifies the user identity through fingerprint and facial recognition, providing high-precision protection for the confirmation of the user identity. This not only greatly reduces the security risks that may be brought by traditional passwords, such as password leakage or forgetting, but also improves the user experience, making the identity authentication process safe and convenient; the data protection unit encrypts the data through the AES data encryption standard to ensure the security of the data during transmission and storage. This encryption process covers the two stages of data transmission and storage. Whether the data is dynamic or static, it can be equally protected. Through the encryption of the AES encryption algorithm, even if the data is intercepted during transmission, the attacker cannot interpret its content, ensuring the confidentiality and integrity of the data; the access record unit records detailed logs of user access and operations through the log management system ELK Stack for auditing and security analysis. The ELK Stack can not only collect, store and analyze log data in real time, but also display it in the form of intuitive charts and reports, providing strong data support for security auditing and troubleshooting. By analyzing the access logs, abnormal behaviors such as multiple failed login attempts or unauthorized data access can be discovered in a timely manner, so that measures can be taken quickly to prevent potential security threats. In addition, the retention of access records also provides a reliable basis for subsequent compliance checks and responsibility tracing, ensuring the security and compliance of the system. In summary, the permission management module builds a comprehensive, multi-level security protection system through the three mechanisms of user identity authentication, data encryption, and access records. This not only protects the security of data, but also maintains the stable operation of the system, providing users with a safe and reliable service experience.

[0115] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0116] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A power communication network fault prediction and diagnosis system, characterized in that: It includes data acquisition module, fault diagnosis module and prediction analysis module; among them, The data acquisition module is used to collect monitoring data in real time and then manage the transmission rate and priority of the data using a sliding window protocol; The fault diagnosis module is used to detect abnormal data that deviates from a normal range through an adaptive data analysis algorithm; The prediction analysis module is used to predict system failures through a dynamic prediction algorithm.

2. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: Detecting abnormal data that deviates from the normal range through an adaptive data analysis algorithm includes: Construct a fault diagnosis model based on autoencoder, Extract features from raw data through time series analysis and recursive feature elimination to train the fault diagnosis model; The trained fault diagnosis model is used to detect anomalies in real-time data. When the real-time data differs significantly from the model expectations, the data is marked as abnormal. The formula of the anomaly detection function is: In the formula, Y represents the anomaly detection function, λ represents the monitoring factor of the fault diagnosis model, δ represents the real-time data to be detected in the power communication network, ρ represents the expected value of the fault diagnosis model, n represents the order of termination of the real-time data, and tan 2 (δ+ρ) represents the evaluation function detection coefficient.

3. The power communication network fault prediction and diagnosis system according to claim 2, characterized in that: The trained fault diagnosis model is evaluated, and the evaluation function is: In the formula, P represents the evaluation function, κ represents the error factor, α i represents the predicted value of the power communication network, β i represents the real value of the power communication network, i represents the data type predicted by the power communication network, Represents the function evaluation coefficients.

4. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: Dynamic prediction algorithm, the steps include: Build a dynamic prediction model based on time series ARIMA and perform dynamic data prediction after training; The prediction result verification function is used to analyze and verify the prediction result; the formula of the prediction result verification function is expressed as: Where, T c represents the prediction result verification function, θ represents the error value of the model data, ε i Represents the predicted value output by the dynamic prediction model, μ i represents the true value of the power communication network; i represents the order in which the dynamic prediction model is verified; Set monitoring thresholds for data anomalies and identify potential faults based on the monitoring thresholds and prediction results.

5. The power communication network fault prediction and diagnosis system according to claim 4, characterized in that: The dynamic prediction model is adjusted according to the feedback of the real-time status data value of the power communication network. The feedback function expression is: In the formula, F represents the feedback function, ω represents the function comparison factor, represents the prediction result value of the dynamic prediction model, ξ represents the real-time status data value of the power communication network, Represents the feedback function coefficient.

6. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: The data acquisition module includes a sensor monitoring unit, a data processing unit, a data flow control unit and a communication unit which are connected in sequence; wherein the data flow control unit is used to manage the transmission rate and priority of the data using a sliding window protocol after real-time acquisition of monitoring data.

7. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: The fault diagnosis module includes a data anomaly detection unit, a fault identification unit and a fault location unit which are connected in sequence; wherein the data anomaly detection unit is used to detect abnormal data deviating from a normal range through an adaptive data analysis algorithm.

8. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: The prediction and analysis module includes a real-time prediction and analysis unit, a risk assessment unit and an alarm unit which are connected in sequence; wherein the real-time prediction and analysis unit is used to predict system failures through a dynamic prediction algorithm.

9. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: It also includes a control terminal connected to the data acquisition module, the fault diagnosis module and the prediction and analysis module. The control terminal includes a system status monitoring unit, a task coordination unit and a system configuration unit; System status monitoring unit, used to display network status data to operators through a visual interface to monitor the operating status of the system in real time; The task coordination unit implements task scheduling and reasonable resource allocation through the distributed task scheduling framework Celery; The system configuration unit is used to configure and modify system parameters through the configuration management database CMDB.

10. The power communication network fault prediction and diagnosis system according to claim 1, characterized in that: It also includes a human-computer interaction module, which includes a graphical user interface, a push notification unit and a query feedback unit; A graphical user interface, which is used to provide users with a visual interface through a liquid crystal display to display system status, fault information and prediction results; Push notification unit, used to push power communication network information to operation and maintenance personnel via SMS and instant messaging tools; The query feedback unit is used to query the system detailed information through the non-relational database MongoDB.

Images