Intelligent substation relay protection system hidden fault automatic diagnosis platform construction method

Abstract

The invention discloses an automatic diagnosis platform construction method for hidden faults of an intelligent substation relay protection system, and belongs to the technical field of relay protection. According to the intelligent substation relay protection system, the problem that hidden faults are difficult to diagnose by means of regular planned maintenance in the prior art is solved, data of the intelligent substation relay protection system are collected and processed through the deployed monitoring equipment, and the processed data are transmitted to the data center to be analyzed; an encryption transmission technology and a redundancy transmission technology are adopted for data transmission, transmission safety and reliability are improved, a data center analyzes data collected by monitoring equipment through a data mining technology, hidden faults of the intelligent substation relay protection system are automatically diagnosed through analysis, and then the intelligent substation relay protection system is monitored. And the processed collected data, the analyzed data and the analysis result are displayed through a visual data display page, and an identity verification unit is established on the automatic diagnosis platform, so that the safety of the automatic diagnosis platform is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of relay protection, and in particular to a method for constructing an automatic diagnosis platform for hidden faults of a relay protection system of an intelligent substation. Background Art

[0002] Relay protection is an important guarantee for the safe operation of the power system. It requires that the primary system can be well adapted in any state, and can quickly and reliably remove the fault or send a signal within the predetermined protection range when the primary system fails or is abnormal. For the relay protection system to function correctly, it must ensure that all links are in normal working condition.

[0003] Relay protection system is an important means to ensure the safe operation of power system. However, hidden faults may cause false operation or refusal of protection, which may lead to serious power accidents. Hidden faults of relay protection are relatively hidden. These faults will not appear when the power system is operating normally, posing a threat to the safe and stable operation of the power system. At present, most of them only rely on regular planned maintenance to ensure the operation of relay protection system, and it is difficult to diagnose the relay protection system. Therefore, it does not meet the existing needs. In this regard, we propose a method for constructing an automatic diagnosis platform for hidden faults of relay protection system in smart substation. Summary of the invention

[0004] The purpose of the present invention is to provide a method for constructing an automatic diagnosis platform for hidden faults of a relay protection system of an intelligent substation. The method collects and processes data of the relay protection system of the intelligent substation by deploying monitoring equipment, and transmits the processed data to a data center for analysis. The data center uses data mining technology to analyze the data collected by the monitoring equipment, and automatically diagnoses the hidden faults of the relay protection system of the intelligent substation through analysis. The processed collected data, the analyzed data and the analysis results are displayed through a visual data display page, and the automatic diagnosis platform establishes an identity authentication unit to ensure the security of the automatic diagnosis platform, thereby solving the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for constructing an automatic diagnosis platform for hidden faults of a relay protection system of an intelligent substation, comprising the following steps:

[0006] Step 1: Determine the area of ​​the smart substation relay protection system and deploy monitoring equipment in the area;

[0007] Step 2: Collect and process the data of the relay protection system of the smart substation through the deployed monitoring equipment, and transmit the processed data to the data center for analysis;

[0008] The monitoring equipment collects and processes the data of the relay protection system of the smart substation, which specifically includes the following steps:

[0009] Data collection: Monitor the relay protection system of the smart substation in the area and collect data about the relay protection system of the smart substation through monitoring;

[0010] Data processing: Process the collected data, including data cleaning, data conversion and data integration;

[0011] Data transmission: The processed data is transmitted through encryption transmission technology and redundant transmission technology. Encryption transmission technology is used to protect data to prevent data leakage and tampering, and redundant transmission technology is used to improve the reliability of data transmission and ensure that data is transmitted to the data center in a timely manner;

[0012] When processing the collected data, the following steps are also included:

[0013] Real-time monitoring: real-time monitoring of the data processing time corresponding to each batch of data processing data;

[0014] Data volume acquisition: when the data processing time exceeds a preset data processing time threshold, the data volume of a batch of data processing corresponding to the data processing time exceeding the preset data processing time threshold is extracted;

[0015] Data volume comparison: comparing the data volume of the data processed with a preset data volume threshold;

[0016] Acquiring a first data processing duration evaluation parameter: when the amount of data processed does not exceed a preset data amount threshold, obtaining a first data processing duration evaluation parameter using the data processing duration corresponding to each batch of data processed;

[0017] The first data processing time evaluation parameter is obtained by the following formula:

[0018]

[0019] Among them, R 01 represents the first data processing time evaluation parameter; m represents the number of data processing times where the data processing time does not exceed the preset data processing time reference value; n represents the number of data processing times where the data processing time exceeds the preset data processing time reference value; T j represents the data processing time of the jth data processing among the data processing times that the data processing time does not exceed the preset data processing time reference value; T i represents the data processing time of the i-th data processing among the data processing times whose data processing time exceeds the preset data processing time reference value; T c Indicates the preset data processing time reference value; C crepresents the maximum theoretical data volume that can be processed by the data processing corresponding to the preset data processing time reference value; C j The data processing amount of the jth data processing among the data processing times whose data processing duration does not exceed the preset data processing duration reference value; C i Indicates the data processing amount of the i-th data processing among the data processing times that do not exceed the preset data processing time reference value during data processing;

[0020] First abnormality alarm: when the first data processing time evaluation parameter does not exceed the preset first parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued;

[0021] Data processing self-check evaluation: when the amount of data processed exceeds a preset data amount threshold, the data processing status of the data processing is evaluated using the amount of data processed in combination with the data processing time;

[0022] Data processing self-examination and evaluation, including the following steps:

[0023] Data information retrieval: when the data volume of the data processing exceeds a preset data volume threshold, the data volume and data processing duration of the data processing exceeding a preset data processing duration threshold are retrieved;

[0024] Time difference data acquisition: acquiring time difference data between the data processing time and the preset data processing time threshold according to the data processing time exceeding the preset data processing time threshold and the preset data processing time threshold;

[0025] Acquisition of the second data processing duration evaluation parameter: the second data processing duration evaluation parameter is acquired by using the data volume and data processing duration of the data processing exceeding the preset data processing duration threshold value in combination with the time difference data between the data processing duration and the preset data processing duration threshold value; wherein the second data processing duration evaluation parameter is acquired by the following formula:

[0026]

[0027] Among them, R 02 represents the second data processing time evaluation parameter; C m and T m respectively represent the data volume and data processing time of data processing exceeding the preset data processing time threshold; T y Indicates the preset data processing time threshold; C y Indicates the maximum theoretical amount of data that can be processed by the data processing corresponding to the preset data processing time threshold;

[0028] Second abnormality alarm: when the second data processing time evaluation parameter does not exceed the preset second parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued;

[0029] Step 3: The data center collects historical fault data and extracts features through data mining technology, establishes a fault diagnosis model and trains it, and transmits the processed data in step 2 to the trained fault diagnosis model. The fault diagnosis model analyzes the data and realizes the automatic diagnosis function of the fault diagnosis model.

[0030] Step 4: Design an intuitive and clear visual data display page;

[0031] The processed collected data, analyzed data and analysis results are displayed to users through a visual data display page, and an early warning page is established based on the analysis results. For hidden faults in the relay protection system of the smart substation, early warning notifications are issued to staff through the early warning page;

[0032] Step 5: Establish an identity authentication unit to ensure the security of the automatic diagnosis platform.

[0033] Furthermore, the relay protection system area of ​​the smart substation is determined, and the following processes are specifically performed:

[0034] Determine the relay protection system area of ​​the smart substation;

[0035] Through actual visits, determine the distribution and environment of the substation protection system in the area;

[0036] According to the distribution and environment of the substation relay protection system, analyze the best location for monitoring the substation relay protection system;

[0037] Among them, the distribution situation refers to the distribution density of the substation relay protection system;

[0038] Environmental conditions include ambient temperature, ambient humidity, and electromagnetic interference;

[0039] Deploy monitoring equipment at the analyzed optimal regulatory locations.

[0040] Furthermore, there is one or more relay protection system areas in the smart substation, and corresponding monitoring equipment is deployed in each relay protection system area of ​​the smart substation. Each relay protection system area of ​​the smart substation collects data through the corresponding monitoring equipment, and the number of monitoring equipment is reasonably deployed according to the actual situation of each relay protection system area of ​​the smart substation.

[0041] Furthermore, the data center specifically performs the following process:

[0042] Collect data: Collect historical fault data of the relay protection system of the smart substation;

[0043] Feature extraction: Use data mining technology to mine historical fault data, and then extract features from historical fault data to extract features related to hidden faults in the relay protection system of smart substations;

[0044] Among them, data mining technology uses cluster analysis method to mine historical fault data. The cluster analysis method groups historical fault data into subsets with similar characteristics and classifies detailed historical fault data into one category, thereby revealing the structure and pattern information hidden in the historical fault data;

[0045] Data division: Data division is performed on the historical fault data after feature extraction, and the historical fault data is divided into a training set and a test set;

[0046] Model establishment: In the process of model establishment, a fault diagnosis model is selected according to the relay protection system of the intelligent substation, wherein the fault diagnosis model is established by using an artificial neural network or a decision tree;

[0047] Model training: The fault diagnosis model is trained using the training set. The training set uses the extracted features to train the fault diagnosis model, selects features related to hidden faults of the smart substation relay protection system from the features, and trains the fault diagnosis model based on the features;

[0048] Model evaluation: Use the test set to evaluate the fault diagnosis model to assess its diagnostic accuracy, robustness, and generalization ability.

[0049] Model deployment: Deploy and apply the trained and evaluated fault diagnosis model, transfer the processed data in step 2 to the fault diagnosis model, and the fault diagnosis model analyzes the data. Once a fault is detected, an early warning message is immediately issued, and the early warning page established in step 4 sends an early warning notification to the staff.

[0050] Furthermore, the fault diagnosis model needs to be regularly updated and maintained, specifically:

[0051] Model update: The fault diagnosis model is updated regularly according to the changes in the operating environment of the smart substation relay protection system;

[0052] Model maintenance: The fault diagnosis model needs to be maintained during the application process. The fault diagnosis model needs to be monitored, abnormal conditions of the fault diagnosis model need to be discovered in time, and response measures need to be taken in time for the abnormal conditions.

[0053] Furthermore, the identity verification unit is used to ensure the security of the automatic diagnosis platform, and specifically includes the following steps:

[0054] Identity registration: Users register an account to access the automatic diagnosis platform. The account registration is completed after authorization. When registering an account, users need to enter basic information, including personal information, mobile phone number, and ID card number, and confirm the user's identity through SMS verification;

[0055] Account authorization: The staff will review the registered account based on the basic user information entered when registering the account. The staff will decide whether to grant access rights to the registered account based on the basic user information;

[0056] Abnormal warning: When an account logs in multiple times but fails, the account is judged to be abnormal and a warning message is sent to the automatic diagnosis platform. The automatic diagnosis platform handles the abnormal account.

[0057] Furthermore, the staff will decide whether to grant access rights to the registered account based on the user's basic information, which can be divided into the following situations:

[0058] If it is decided not to grant access rights to the registered account, the account registration will fail and the account will not be able to log in to the automatic diagnosis platform for access;

[0059] If it is decided to grant access rights to the registered account, the account registration is successful, and the user uses the successfully registered account to log in to the automatic diagnosis platform for access.

[0060] Furthermore, when an account is abnormal, a warning message is sent to the automatic diagnosis platform, and the automatic diagnosis platform processes the abnormal account, specifically including the following steps:

[0061] The automatic diagnosis platform obtains abnormal accounts based on warning information;

[0062] The automatic diagnosis platform retrieves the basic information entered when the abnormal account was registered, and determines the user who owns the abnormal account based on the basic information;

[0063] Send a text message notification to the user who owns the abnormal account, and the automatic diagnosis platform bans the abnormal account;

[0064] The user submits an account abnormality description to the automatic diagnosis platform based on the SMS notification. After the automatic diagnosis platform reviews the submitted account abnormality description, it decides whether to lift the ban on the user's abnormal account.

[0065] Compared with the prior art, the present invention has the following beneficial effects:

[0066] 1. After determining the area of ​​the relay protection system of the smart substation, the present invention deploys reasonable monitoring equipment in the area. The monitoring equipment can realize the collection and processing of the data of the relay protection system of the smart substation, and the processed data is transmitted using encryption transmission technology and redundant transmission technology to improve the security and reliability of data transmission. The data center analyzes the transmitted data. The data center uses data mining technology to perform cluster analysis on a large amount of pre-stored historical data, classifies similar faults into one category, and establishes a corresponding diagnostic model. The data collected by the current monitoring equipment is input into the diagnostic model for analysis. By bringing in the diagnostic model to analyze and determine whether there is a hidden fault problem in the current smart substation relay protection system, the above-mentioned processed collected data and analyzed data and analysis results will be displayed to the user on a visualized data display page, and can issue an early warning notification to the staff through the early warning page for the hidden faults in the relay protection system of the smart substation.

[0067] 2. The present invention provides users with account registration and verification functions by setting up an identity authentication unit, and the staff authorizes the registered account. At the same time, abnormal monitoring is performed when the account logs in, thereby effectively ensuring the security of the automatic diagnosis platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 A flow chart of a method for constructing an automatic diagnosis platform for hidden faults of a relay protection system of an intelligent substation according to the present invention;

[0069] Figure 2 A schematic diagram of the structure of the automatic diagnosis platform for hidden faults of the intelligent substation relay protection system of the present invention. DETAILED DESCRIPTION

[0070] 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.

[0071] In order to solve the technical problem that most of the existing relay protection systems rely on regular planned maintenance to ensure their operation and are difficult to diagnose, please refer to Figure 1-Figure 2 , this embodiment provides the following technical solutions:

[0072] The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation includes the following steps:

[0073] Step 1: Determine the area of ​​the smart substation relay protection system and deploy monitoring equipment in the area;

[0074] Step 2: Collect and process the data of the relay protection system of the smart substation through the deployed monitoring equipment, and transmit the processed data to the data center for analysis;

[0075] Step 3: The data center collects historical fault data and extracts features through data mining technology, establishes a fault diagnosis model and trains it, and transmits the processed data in step 2 to the trained fault diagnosis model. The fault diagnosis model analyzes the data and realizes the automatic diagnosis function of the fault diagnosis model.

[0076] Step 4: Design an intuitive and clear visual data display page;

[0077] Step 5: Establish an identity authentication unit to ensure the security of the automatic diagnosis platform.

[0078] Determine the relay protection system area of ​​the smart substation and perform the following procedures:

[0079] Determine the relay protection system area of ​​the smart substation;

[0080] Through actual visits, determine the distribution and environment of the substation protection system in the area;

[0081] According to the distribution and environment of the substation relay protection system, analyze the best location for monitoring the substation relay protection system;

[0082] The distribution situation refers to the density of the substation relay protection system, which determines the deployment density of the monitoring equipment. During the deployment of the monitoring equipment, if the substation relay protection system is too densely distributed, the deployment locations of the monitoring equipment may interfere with each other, affecting the accuracy of the data collected by the monitoring equipment.

[0083] Environmental conditions include ambient temperature, ambient humidity and electromagnetic interference. Fluctuations in ambient temperature and humidity may cause unstable data collected by monitoring equipment. Electromagnetic interference may cause disorder in the control system of monitoring equipment, affecting the performance of monitoring equipment and the accuracy of data collected by monitoring equipment.

[0084] Deploy monitoring equipment at the analyzed optimal locations for supervision;

[0085] Among them, there is one or more relay protection system areas of the smart substation, and corresponding monitoring equipment is deployed in each relay protection system area of ​​the smart substation. Each relay protection system area of ​​the smart substation collects data through the corresponding monitoring equipment, and the number of monitoring equipment is reasonably deployed according to the actual situation of each relay protection system area of ​​the smart substation.

[0086] The technical effects of the above content are as follows: the role of the intelligent substation relay protection system in the power system is vital. The substation relay protection system is distributed at various key nodes of the power system, so that it can monitor parameters such as voltage, current, frequency, and detect and judge faults in the power system to achieve rapid isolation of faults. By determining the area of ​​the intelligent substation relay protection system and determining the deployment location of the monitoring equipment according to the distribution and environmental conditions of the substation relay protection system, the reliability of the monitoring equipment and the accuracy of the collected data can be improved. The equipment can monitor the relay protection system of the smart substation in real time to ensure the safe and stable operation of the relay protection system of the smart substation. For example, monitoring equipment A is deployed in area A of the relay protection system of the smart substation, and monitoring equipment B is deployed in area B of the relay protection system of the smart substation. In this way, the deployed monitoring equipment can conduct targeted monitoring of the area to ensure the uniqueness of the data collected by the monitoring equipment. If the data collected by the subsequent monitoring equipment A is analyzed and hidden faults are found, the hidden fault can be determined in area A of the relay protection system of the smart substation, so that the staff can quickly grasp the fault area and then promptly troubleshoot the hidden fault.

[0087] The monitoring equipment collects and processes the data of the relay protection system of the smart substation, which specifically includes the following steps:

[0088] Data collection: Monitor the relay protection system of the smart substation in the area and collect data about the relay protection system of the smart substation through monitoring;

[0089] Data processing: Process the collected data, including data cleaning, data conversion and data integration;

[0090] Data transmission: The processed data is transmitted through encryption transmission technology and redundant transmission technology. Encryption transmission technology is used to protect data to prevent data leakage and tampering, and redundant transmission technology is used to improve the reliability of data transmission and ensure that data is transmitted to the data center in a timely manner.

[0091] The technical effects of the above content are: first, data about the relay protection system of the smart substation is collected, and then the collected data is cleaned, converted and integrated to improve the reliability of the data, thereby improving the accuracy of subsequent data analysis. Finally, the processed data is transmitted to the data center. During the transmission process, encryption transmission technology and redundant transmission technology are used respectively. Encryption transmission technology can protect data during transmission to prevent data leakage and tampering, thereby improving the security of data transmission. Redundant transmission technology is a technology that uses one or more sets of additional drives to store copies of data. In this way, the reliability of data transmission is improved, thereby ensuring that data is transmitted to the data center in a timely manner. The monitoring equipment can realize the collection, processing and transmission of data of the relay protection system of the smart substation through the three steps of data collection, data processing and data transmission. The data collected by the monitoring equipment provides data support for the construction of the automatic diagnosis platform.

[0092] Specifically, when processing the collected data, the following steps are also included:

[0093] Real-time monitoring: real-time monitoring of the data processing time corresponding to each batch of data processing data;

[0094] Data volume acquisition: when the data processing time exceeds a preset data processing time threshold, the data volume of a batch of data processing corresponding to the data processing time exceeding the preset data processing time threshold is extracted;

[0095] Data volume comparison: comparing the data volume of the data processed with a preset data volume threshold;

[0096] Acquiring a first data processing duration evaluation parameter: when the amount of data processed does not exceed a preset data amount threshold, obtaining a first data processing duration evaluation parameter using the data processing duration corresponding to each batch of data processed;

[0097] The first data processing time evaluation parameter is obtained by the following formula:

[0098]

[0099] Among them, R 01 represents the first data processing time evaluation parameter; m represents the number of data processing times where the data processing time does not exceed the preset data processing time reference value; n represents the number of data processing times where the data processing time exceeds the preset data processing time reference value; T j represents the data processing time of the jth data processing among the data processing times that the data processing time does not exceed the preset data processing time reference value; T irepresents the data processing time of the i-th data processing among the data processing times whose data processing time exceeds the preset data processing time reference value; T c Indicates the preset data processing time reference value; C c represents the maximum theoretical data volume that can be processed by the data processing corresponding to the preset data processing time reference value; C j The data processing amount of the jth data processing among the data processing times whose data processing duration does not exceed the preset data processing duration reference value; C i Indicates the data processing amount of the i-th data processing among the data processing times that do not exceed the preset data processing time reference value during data processing;

[0100] First abnormality alarm: when the first data processing time evaluation parameter does not exceed the preset first parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued;

[0101] Data processing self-check evaluation: When the data volume of the data processing exceeds a preset data volume threshold, the data processing status of the data processing is evaluated using the data volume of the data processing combined with the data processing time.

[0102] The technical effect of the above technical solution is: through the real-time monitoring link, the time consumption anomalies in the data processing process can be discovered in time, that is, when the data processing time exceeds the preset threshold, further analysis and evaluation processes are triggered. This real-time monitoring mechanism helps to quickly respond to and deal with potential performance issues, and reduce the business impact caused by processing delays. When an abnormal data processing time is detected, through the data volume acquisition link and the data volume comparison link, it can be accurately identified whether the increase in time consumption is due to excessive data volume or insufficient performance of the data processing step itself. This distinction helps with subsequent precise optimization measures.

[0103] The first data processing time evaluation parameter is obtained through a complex formula calculation, which comprehensively considers the data processing time and data volume in different situations and generates a comprehensive evaluation parameter (R 01 ). This parameter not only takes into account the duration of a single processing, but also incorporates the comparison between historical processing duration and theoretical processing capacity, providing a more comprehensive and scientific basis for the performance evaluation of data processing. When the evaluation parameter does not meet the preset threshold, the first abnormal alarm link will trigger an abnormal alarm, indicating that there is an abnormality in data processing. This instant feedback mechanism helps operation and maintenance personnel to quickly intervene, troubleshoot and solve problems, and reduce downtime.

[0104] When the data volume exceeds the preset threshold, the data processing self-check evaluation stage will further use the data volume and processing time to evaluate the status of the data processing step. This self-check mechanism helps to discover potential performance bottlenecks or improper configuration problems in advance, providing data support for system tuning.

[0105] The entire technical solution significantly improves the stability and reliability of the data processing step through the organic combination of real-time monitoring, precise positioning, dynamic assessment, abnormal alarm and self-test assessment, ensuring the continuity and efficiency of the business system.

[0106] In summary, through a series of refined designs and implementations, this technical solution has effectively improved the performance monitoring, problem diagnosis, and system tuning capabilities of the data processing step, providing a strong guarantee for the stable operation of the business system.

[0107] Specifically, the data processing self-examination and evaluation includes the following steps:

[0108] Data information retrieval: when the data volume of the data processing exceeds a preset data volume threshold, the data volume and data processing duration of the data processing exceeding a preset data processing duration threshold are retrieved;

[0109] Time difference data acquisition: acquiring time difference data between the data processing time and the preset data processing time threshold according to the data processing time exceeding the preset data processing time threshold and the preset data processing time threshold;

[0110] Acquisition of the second data processing duration evaluation parameter: the second data processing duration evaluation parameter is acquired by using the data volume and data processing duration of the data processing exceeding the preset data processing duration threshold value in combination with the time difference data between the data processing duration and the preset data processing duration threshold value; wherein the second data processing duration evaluation parameter is acquired by the following formula:

[0111]

[0112] Among them, R 02 represents the second data processing time evaluation parameter; C m and T m respectively represent the data volume and data processing time of data processing exceeding the preset data processing time threshold; T y Indicates the preset data processing time threshold; C y Indicates the maximum theoretical amount of data that can be processed by the data processing corresponding to the preset data processing time threshold;

[0113] Second abnormality alarm: when the second data processing time evaluation parameter does not exceed the preset second parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued.

[0114] The technical effect of the above technical solution is that when the amount of data processed exceeds the preset threshold, the technical solution not only considers the amount of data itself, but also deeply analyzes the difference between the data processing time and the preset threshold (i.e., time difference data). This in-depth analysis enables the system to more accurately determine the performance of the data processing step when processing large amounts of data, thereby more accurately identifying potential problems.

[0115] Through the acquisition formula of the second data processing time evaluation parameter, the system can combine the amount of data exceeding the preset time threshold, the actual processing time, the preset processing time threshold and the theoretical processing capacity (C y ) to comprehensively evaluate the performance of the data processing step. This method of dynamically adjusting the evaluation parameters makes the evaluation results closer to the actual operation situation and improves the accuracy and reliability of the evaluation.

[0116] When the second data processing time evaluation parameter does not reach the preset second parameter threshold, the system can immediately determine that there is an abnormality in data processing and trigger an abnormality alarm. This real-time detection and alarm mechanism helps operation and maintenance personnel to quickly respond to and handle potential problems, reducing system downtime and business losses.

[0117] Through in-depth evaluation of the performance of the data processing step, the system can allocate resources and adjust the load more reasonably to optimize the overall performance. For example, when a data processing performance bottleneck is found, task allocation can be dynamically adjusted to transfer some tasks to other data processing steps with better performance, thereby improving overall processing efficiency and stability.

[0118] This technical solution provides strong support for system maintenance and expansion by providing detailed performance evaluation data and real-time anomaly detection functions. Operation and maintenance personnel can quickly locate problems based on evaluation results and alarm information and take corresponding optimization measures. At the same time, the system is also easier to expand and upgrade according to business needs.

[0119] By ensuring the stability and efficiency of data processing, this technical solution can improve user experience. Whether it is online services, data analysis or other business scenarios that rely on data processing, users can enjoy faster and more reliable service response and data processing capabilities.

[0120] In summary, this technical solution significantly improves the performance evaluation accuracy of the data processing step and the overall stability, maintainability and scalability of the system through in-depth data analysis, dynamic performance evaluation and real-time anomaly detection and alarm mechanism.

[0121] The content of data processing is to process the data collected by the monitoring equipment, and then conduct the next step of data processing quality evaluation and monitoring. Data processing can improve the stability and reliability of data processing through the organic combination of real-time monitoring, precise positioning, dynamic evaluation, abnormal alarm and self-test evaluation. The self-test evaluation link can further improve the performance evaluation accuracy of data processing and the overall stability, maintainability and scalability of the system through in-depth data analysis, dynamic performance evaluation and real-time abnormal detection and alarm mechanism. Data processing is the basis for subsequent data analysis in the data center. When the data collected by the monitoring equipment is transmitted to the data center for analysis after a series of data processing, it can realize the automatic diagnosis of hidden faults in the relay protection system of the smart substation.

[0122] The data center specifically performs the following processes:

[0123] Collect data: Collect historical fault data of the relay protection system of the smart substation;

[0124] Feature extraction: Use data mining technology to mine historical fault data, and then extract features from historical fault data to extract features related to hidden faults in the relay protection system of smart substations;

[0125] Among them, data mining technology uses cluster analysis method to mine historical fault data. The cluster analysis method groups historical fault data into subsets with similar characteristics and classifies detailed historical fault data into one category, thereby revealing the structure and pattern information hidden in the historical fault data;

[0126] Data division: Data division is performed on the historical fault data after feature extraction, and the historical fault data is divided into a training set and a test set;

[0127] Model establishment: In the process of model establishment, a fault diagnosis model is selected according to the relay protection system of the intelligent substation, wherein the fault diagnosis model is established by using an artificial neural network or a decision tree;

[0128] Model training: The fault diagnosis model is trained using the training set. The training set uses the extracted features to train the fault diagnosis model, selects features related to hidden faults of the smart substation relay protection system from the features, and trains the fault diagnosis model based on the features;

[0129] Model evaluation: Use the test set to evaluate the fault diagnosis model to assess its diagnostic accuracy, robustness, and generalization ability.

[0130] Model deployment: Deploy and apply the trained and evaluated fault diagnosis model, transfer the processed data in step 2 to the fault diagnosis model, and the fault diagnosis model analyzes the data. Once a fault is detected, an early warning message is immediately issued, and the early warning page established in step 4 sends an early warning notification to the staff.

[0131] The technical effect of the above content is: the data center collects a large amount of historical fault data, and the collected historical fault data needs to be representative. Based on the representative historical fault data, the data mining technology uses cluster analysis method to mine the historical fault data. The cluster analysis method can be used as an independent tool to obtain the distribution status of historical fault data, observe the characteristics of each cluster of historical fault data, and classify similar faults into one category according to the characteristics of historical fault data, so as to reveal the hidden structure and pattern information in the historical fault data, so as to extract the characteristics related to the hidden faults of the intelligent substation relay protection system, and divide the historical fault data after feature extraction into training set and test set. The training set is used for model training, and the test set is used for model evaluation. The fault diagnosis model can adopt a variety of methods, such as artificial neural network or decision tree. Method, so that a suitable fault diagnosis model can be established according to the intelligent substation relay protection system. Then, the training set uses the extracted features to train the fault diagnosis model. The diagnostic accuracy and generalization ability of the fault diagnosis model can be improved through training. The fault diagnosis model is evaluated using the test set to evaluate the diagnostic accuracy, robustness and generalization ability of the fault diagnosis model. The accuracy and credibility of the fault diagnosis model can be understood through evaluation. Finally, the trained and evaluated fault diagnosis model is deployed and applied, and the processed data in step 2 is transmitted to the fault diagnosis model. The fault diagnosis model analyzes the data. Once a fault is detected, an early warning message is immediately issued, and the early warning page established in step 4 sends an early warning notification to the staff, thereby realizing the automatic diagnosis function of hidden faults, which can effectively reduce the occurrence of safety accidents.

[0132] The fault diagnosis model needs to be updated and maintained regularly, specifically:

[0133] Model update: According to the changes in the operating environment of the smart substation relay protection system, the fault diagnosis model is updated regularly. The operating environment and model update are interrelated processes. The operating environment provides the basis for the operation of the fault diagnosis model, and the model update is carried out in the operating environment to improve the performance and adaptability of the fault diagnosis model.

[0134] Model maintenance: The fault diagnosis model needs to be maintained during the application process. The fault diagnosis model needs to be monitored, abnormal conditions of the fault diagnosis model need to be discovered in time, and response measures need to be taken in time for the abnormal conditions.

[0135] The technical effects of the above content are: by regularly updating the fault diagnosis model, the performance and adaptability of the fault diagnosis model can be improved, and the performance of the fault diagnosis model can be improved through model maintenance. Model monitoring is to monitor the performance of the fault diagnosis model in real time after the model is deployed to ensure the stability and reliability of the fault diagnosis model in practical applications. The main tasks of model monitoring include: monitoring the input data distribution of the fault diagnosis model to detect potential data anomalies, monitoring the output results of the fault diagnosis model to evaluate the prediction performance of the fault diagnosis model, and monitoring the operating status of the fault diagnosis model to ensure the stable operation of the fault diagnosis model. Model maintenance is to regularly maintain and optimize the fault diagnosis model on the basis of model monitoring to ensure the performance of the fault diagnosis model in practical applications. The main tasks of model maintenance include: online learning of the fault diagnosis model to adapt to changes in data distribution, optimizing the fault diagnosis model to improve the generalization performance of the fault diagnosis model, and versioning the fault diagnosis model to facilitate the rollback and update of the fault diagnosis model. Model monitoring and maintenance are two complementary processes. Model monitoring provides data support for model maintenance, and model maintenance is carried out on the basis of model monitoring. Only through effective model monitoring can the fault diagnosis model be effectively maintained.

[0136] Design an intuitive and clear visual data display page to present the processed collected data, analyzed data and analysis results to users through the visual data display page, and establish an early warning page based on the analysis results. For hidden faults in the relay protection system of the smart substation, send early warning notifications to the staff through the early warning page.

[0137] The technical effect of the above content is: an intuitive and clear visual data display page is designed for the automatic diagnosis platform, and the processed collected data, analyzed data and analysis results can be displayed to the user through the designed visual data display page. The user can intuitively understand the operation status and potential fault risks of the smart substation relay protection system through the data displayed on the visual data display page, and for the hidden faults in the smart substation relay protection system, an early warning notification is issued to the staff through the early warning page, so that corresponding measures can be taken in time to ensure the safe and stable operation of the smart substation relay protection system.

[0138] The identity verification unit is used to ensure the security of the automatic diagnosis platform, which includes the following steps:

[0139] Identity registration: Users register an account to access the automatic diagnosis platform. The account registration is completed after authorization. When registering an account, users need to enter basic information, including personal information, mobile phone number, and ID card number, and confirm the user's identity through SMS verification;

[0140] Account authorization: The staff will review the registered account based on the basic user information entered when registering the account. The staff will decide whether to grant access rights to the registered account based on the basic user information;

[0141] Among them, the staff decides whether to grant access rights to the registered account based on the user's basic information, which is divided into the following situations:

[0142] If it is decided not to grant access rights to the registered account, the account registration will fail and the account will not be able to log in to the automatic diagnosis platform for access;

[0143] If it is decided to grant the registered account access rights, the account registration is successful, and the user uses the successfully registered account to log in to the automatic diagnosis platform for access;

[0144] Abnormal warning: When an account logs in multiple times but fails, the account is judged to be abnormal and a warning message is sent to the automatic diagnosis platform, which then handles the abnormal account.

[0145] When an account is abnormal, a warning message is sent to the automatic diagnosis platform, and the automatic diagnosis platform processes the abnormal account, specifically including the following steps:

[0146] The automatic diagnosis platform obtains abnormal accounts based on warning information;

[0147] The automatic diagnosis platform retrieves the basic information entered when the abnormal account was registered, and determines the user who owns the abnormal account based on the basic information;

[0148] Send a text message notification to the user who owns the abnormal account, and the automatic diagnosis platform bans the abnormal account;

[0149] The user submits an account abnormality description to the automatic diagnosis platform based on the SMS notification. After the automatic diagnosis platform reviews the submitted account abnormality description, it decides whether to lift the ban on the user's abnormal account.

[0150] The technical effect of the above content is: the setting of the identity authentication unit can ensure the security of the automatic diagnosis platform. Users access the account of the automatic diagnosis platform through identity registration. When registering an account, it is necessary to enter the user's basic information and confirm the user's identity through SMS verification to ensure the security of the account. After entering the user's basic information, the registered account needs to be authorized. The registration of the account is completed only after authorization. When authorizing the account, the user's basic information entered in the registered account will be reviewed to decide whether to grant the registered account access rights. If it is decided not to grant the registered account access rights, the account registration will fail and the user cannot log in to the automatic diagnosis platform for access. If it is decided to grant the registered account access rights, the account registration will be successful. Then, the user can use the successfully registered account to log in to the automatic diagnosis platform for access. Through account registration and identity authentication, the risk of virtual users can be greatly reduced. The number of registered fake users can be increased, thereby improving the security of access to the automatic diagnosis platform. When a user logs in to the automatic diagnosis platform, the account is monitored. If an account logs in multiple times but fails, the account will be determined to be abnormal and a warning message will be sent to the automatic diagnosis platform at the same time. Then, the automatic diagnosis platform obtains the abnormal account based on the warning message and retrieves the basic information entered when the abnormal account is registered. The user who owns the abnormal account can be determined based on the retrieved basic information, and then a text message notification is sent to the user who owns the abnormal account, and the automatic diagnosis platform bans the abnormal account. After receiving the text message notification, the user can submit an account abnormality description to the automatic diagnosis platform. The automatic diagnosis platform will decide whether to lift the ban on the user's abnormal account based on the submitted account abnormality description. By monitoring account logins, the security of access to the automatic diagnosis platform can be further improved.

[0151] Working principle: The deployed monitoring equipment can realize the collection and processing of the relay protection system data of the smart substation, and the processed data is transmitted using encryption transmission technology and redundant transmission technology, which can improve the security and reliability of data transmission. The data center analyzes the transmitted data and uses data mining technology to cluster the large amount of historical fault data collected in advance, classifying similar faults into one category, thereby building different types of fault diagnosis models, and then bringing the processed data into the fault diagnosis model for analysis, so that the hidden faults of the relay protection system of the smart substation can be automatically diagnosed and classified, so as to make timely judgments on the faults based on the analysis results, which can effectively reduce safety risks. The occurrence of a full accident, and the above-mentioned processed collected data, analyzed data and analysis results will be displayed to users on a visual data display page, so that users can intuitively understand the operation status and potential fault risks of the smart substation relay protection system. In addition, it can also send early warning notifications to staff through the early warning page for hidden faults in the smart substation relay protection system, so that corresponding measures can be taken in time to ensure the safe and stable operation of the smart substation relay protection system. At the same time, through the set identity authentication unit, users are provided with account registration and verification functions, and the staff authorizes the registered account and performs abnormal monitoring when the account logs in, thereby effectively ensuring the security of the automatic diagnosis platform.

[0152] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0153] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for constructing an automatic diagnosis platform for hidden faults in a relay protection system of an intelligent substation, characterized in that: The following steps are involved: Step 1: Determine the area of ​​the smart substation relay protection system and deploy monitoring equipment in the area; Step 2: Collect and process the data of the relay protection system of the smart substation through the deployed monitoring equipment, and transmit the processed data to the data center for analysis; The monitoring equipment collects and processes the data of the relay protection system of the smart substation, which specifically includes the following steps: Data collection: Monitor the relay protection system of the smart substation in the area and collect data about the relay protection system of the smart substation through monitoring; Data processing: Process the collected data, including data cleaning, data conversion and data integration; Data transmission: The processed data is transmitted through encryption transmission technology and redundant transmission technology. Encryption transmission technology is used to protect data to prevent data leakage and tampering, and redundant transmission technology is used to improve the reliability of data transmission and ensure that data is transmitted to the data center in a timely manner. When processing the collected data, the following steps are also included: Real-time monitoring: real-time monitoring of the data processing time corresponding to each batch of data processing data; Data volume acquisition: when the data processing time exceeds a preset data processing time threshold, the data volume of a batch of data processing corresponding to the data processing time exceeding the preset data processing time threshold is extracted; Data volume comparison: comparing the data volume of the data processed with a preset data volume threshold; Acquiring a first data processing duration evaluation parameter: when the amount of data processed does not exceed a preset data amount threshold, obtaining a first data processing duration evaluation parameter using the data processing duration corresponding to each batch of data processed; The first data processing time evaluation parameter is obtained by the following formula: Among them, R 01 represents the first data processing time evaluation parameter; m represents the number of data processing times where the data processing time does not exceed the preset data processing time reference value; n represents the number of data processing times where the data processing time exceeds the preset data processing time reference value; T j represents the data processing time of the jth data processing among the data processing times that the data processing time does not exceed the preset data processing time reference value; T i represents the data processing time of the i-th data processing among the data processing times whose data processing time exceeds the preset data processing time reference value; T c Indicates the preset data processing time reference value; C c represents the maximum theoretical data volume that can be processed by the data processing corresponding to the preset data processing time reference value; C j The data processing amount of the jth data processing among the data processing times whose data processing duration does not exceed the preset data processing duration reference value; C i Indicates the data processing amount of the i-th data processing among the data processing times that do not exceed the preset data processing time reference value during data processing; First abnormality alarm: when the first data processing time evaluation parameter does not exceed the preset first parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued; Data processing self-check evaluation: when the amount of data processed exceeds a preset data amount threshold, the data processing status of the data processing is evaluated using the amount of data processed in combination with the data processing time; Data processing self-examination and evaluation, including the following steps: Data information retrieval: when the data volume of the data processing exceeds a preset data volume threshold, the data volume and data processing duration of the data processing exceeding a preset data processing duration threshold are retrieved; Time difference data acquisition: acquiring time difference data between the data processing time and the preset data processing time threshold according to the data processing time exceeding the preset data processing time threshold and the preset data processing time threshold; Acquisition of the second data processing duration evaluation parameter: the second data processing duration evaluation parameter is acquired by using the data volume and data processing duration of the data processing exceeding the preset data processing duration threshold value in combination with the time difference data between the data processing duration and the preset data processing duration threshold value; wherein the second data processing duration evaluation parameter is acquired by the following formula: Among them, R 02 represents the second data processing time evaluation parameter; C m and T m respectively represent the data volume and data processing time of data processing exceeding the preset data processing time threshold; T y Indicates the preset data processing time threshold; C y Indicates the maximum theoretical amount of data that can be processed by the data processing corresponding to the preset data processing time threshold; Second abnormality alarm: when the second data processing time evaluation parameter does not exceed the preset second parameter threshold, it is determined that there is a data processing abnormality in the data processing, and an abnormality alarm is issued; Step 3: The data center collects historical fault data and extracts features through data mining technology, establishes a fault diagnosis model and trains it, and transmits the processed data in step 2 to the trained fault diagnosis model. The fault diagnosis model analyzes the data and realizes the automatic diagnosis function of the fault diagnosis model. Step 4: Design an intuitive and clear visual data display page; The processed collected data, analyzed data and analysis results are displayed to users through a visual data display page, and an early warning page is established based on the analysis results. For hidden faults in the relay protection system of the smart substation, early warning notifications are issued to staff through the early warning page; Step 5: Establish an identity authentication unit to ensure the security of the automatic diagnosis platform.

2. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 1 is characterized in that: Determine the relay protection system area of ​​the smart substation and perform the following procedures: Determine the relay protection system area of ​​the smart substation; Through actual visits, determine the distribution and environment of the substation protection system in the area; According to the distribution and environment of the substation relay protection system, analyze the best location for monitoring the substation relay protection system; Among them, the distribution situation refers to the distribution density of the substation relay protection system; Environmental conditions include ambient temperature, ambient humidity, and electromagnetic interference; Deploy monitoring equipment at the analyzed optimal regulatory locations.

3. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 1 is characterized in that: There is one or more relay protection system areas in a smart substation. Corresponding monitoring equipment is deployed in each relay protection system area of ​​the smart substation. Each relay protection system area of ​​the smart substation collects data through the corresponding monitoring equipment. The number of monitoring equipment is reasonably deployed according to the actual situation of each relay protection system area of ​​the smart substation.

4. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 1 is characterized in that: The data center specifically performs the following process: Collect data: Collect historical fault data of the relay protection system of the smart substation; Feature extraction: Use data mining technology to mine historical fault data, and then extract features from historical fault data to extract features related to hidden faults in the relay protection system of smart substations; Among them, data mining technology uses cluster analysis method to mine historical fault data. The cluster analysis method groups historical fault data into subsets with similar characteristics and classifies detailed historical fault data into one category, thereby revealing the structure and pattern information hidden in the historical fault data; Data division: Data division is performed on the historical fault data after feature extraction, and the historical fault data is divided into a training set and a test set; Model establishment: In the process of model establishment, a fault diagnosis model is selected according to the relay protection system of the intelligent substation, wherein the fault diagnosis model is established by using an artificial neural network or a decision tree; Model training: The fault diagnosis model is trained using the training set. The training set uses the extracted features to train the fault diagnosis model, selects features related to hidden faults of the smart substation relay protection system from the features, and trains the fault diagnosis model based on the features; Model evaluation: Use the test set to evaluate the fault diagnosis model to assess its diagnostic accuracy, robustness, and generalization ability. Model deployment: Deploy and apply the trained and evaluated fault diagnosis model, transfer the processed data in step 2 to the fault diagnosis model, and the fault diagnosis model analyzes the data. Once a fault is detected, an early warning message is immediately issued, and the early warning page established in step 4 sends an early warning notification to the staff.

5. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 1 is characterized in that: The fault diagnosis model needs to be updated and maintained regularly, specifically: Model update: The fault diagnosis model is updated regularly according to the changes in the operating environment of the smart substation relay protection system; Model maintenance: The fault diagnosis model needs to be maintained during the application process. The fault diagnosis model needs to be monitored, abnormal conditions of the fault diagnosis model need to be discovered in time, and response measures need to be taken in time for the abnormal conditions.

6. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 1 is characterized in that: The identity verification unit is used to ensure the security of the automatic diagnosis platform, and specifically includes the following steps: Identity registration: Users register an account to access the automatic diagnosis platform. The account registration is completed after authorization. When registering an account, users need to enter basic information, including personal information, mobile phone number, and ID card number, and confirm the user's identity through SMS verification; Account authorization: The staff will review the registered account based on the basic user information entered when registering the account. The staff will decide whether to grant access rights to the registered account based on the basic user information; Abnormal warning: When an account logs in multiple times but fails, the account is judged to be abnormal and a warning message is sent to the automatic diagnosis platform. The automatic diagnosis platform handles the abnormal account.

7. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 6 is characterized in that: The staff will decide whether to grant access rights to the registered account based on the user's basic information, which can be divided into the following situations: If it is decided not to grant access rights to the registered account, the account registration will fail and the account will not be able to log in to the automatic diagnosis platform for access; If it is decided to grant access rights to the registered account, the account registration is successful, and the user uses the successfully registered account to log in to the automatic diagnosis platform for access.

8. The method for constructing a hidden fault automatic diagnosis platform for a relay protection system of an intelligent substation according to claim 6 is characterized in that: When an account is abnormal, a warning message is sent to the automatic diagnosis platform, and the automatic diagnosis platform handles the abnormal account, which specifically includes the following steps: The automatic diagnosis platform obtains abnormal accounts based on warning information; The automatic diagnosis platform retrieves the basic information entered when the abnormal account was registered, and determines the user who owns the abnormal account based on the basic information; Send a text message notification to the user who owns the abnormal account, and the automatic diagnosis platform bans the abnormal account; The user submits an account abnormality description to the automatic diagnosis platform based on the SMS notification. After the automatic diagnosis platform reviews the submitted account abnormality description, it decides whether to lift the ban on the user's abnormal account.

Images