Novel ground fault studying and judging system and method

Abstract

The invention provides a novel ground fault studying and judging system and method, and relates to the technical field of power system fault detection.The novel ground fault studying and judging system obtains various parameters of a power system in real time, preprocesses the parameters, extracts multiple characteristic factors related to a ground fault from the preprocessed data, and judges the ground fault according to the extracted characteristic factors. Meanwhile, comprehensive analysis is carried out in combination with historical fault records and environmental factors, a high-performance algorithm server is utilized to carry out real-time study and judgment on the extracted feature factors, and a possibility evaluation result of the grounding fault is generated; the research and judgment result is transmitted to a mobile application program or a monitoring system in real time through a message queue system, meanwhile, a json file is automatically generated and packaged according to the research and judgment result, and when the research and judgment result confirms that a grounding fault exists, warning information is pushed to related personnel or systems through short message service so that necessary measures can be taken in time for fault processing. According to the method, real-time monitoring, accurate research and judgment and quick response of the grounding fault of the power system are realized through an intelligent means.

Description

Technical Field

[0001] The invention belongs to the technical field of power system fault detection, and specifically proposes a novel ground fault analysis system and method. Background Art

[0002] In the power system, ground fault is a common fault type, which may cause serious consequences such as equipment damage and power outage. Therefore, rapid and accurate analysis of ground fault is of great significance to ensure the stable operation of the power system.

[0003] Traditional ground fault analysis methods mainly rely on manual inspections and fault indicator indications. However, these methods have many shortcomings. For example, manual inspections require a lot of manpower and time, and are easily affected by factors such as weather and terrain, resulting in low inspection efficiency and accuracy. Although fault indicators can indicate the fault location, they can often only provide qualitative information and cannot provide detailed fault characteristics and parameters, thus limiting the accuracy and reliability of the analysis.

[0004] With the improvement of the intelligence level of power systems, more and more intelligent equipment and technologies are being used in ground fault analysis. However, there are still some problems with the existing intelligent analysis methods. For example, some methods only focus on the single characteristics of fault current or voltage, while ignoring other related factors, resulting in inaccurate analysis results. In addition, although some methods consider multiple characteristics, the analysis process is complicated and the calculation amount is large, making it difficult to achieve real-time analysis and the accuracy of the analysis results remains to be discussed. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to propose a new ground fault analysis system and method, aiming to improve the accuracy and efficiency of ground fault analysis, reduce the false alarm and missed alarm rates, and provide strong guarantee for the stable operation of the power system.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is as follows: a new ground fault analysis method, comprising the following steps: 1) The terminal detects 3U0 or 3I0 over-limit events and initiates analysis On the terminal equipment of the power system, a 3U0 (zero-sequence voltage) or 3I0 (zero-sequence current) over-limit detection module is set. When 3U0 or 3I0 is detected to exceed the preset threshold, it is considered that a ground fault may have occurred. At this time, the analysis and judgment process is started and the starting time point T0 is recorded.

[0007] 2) Report to the cloud master after the fault lasts for a set period of time After starting the analysis and judgment process, if the fault continues for a set time (such as 5 seconds, which can be set) and is not eliminated, the terminal device will report the fault event and the starting time point T0 to the cloud master station. As the monitoring and management center of the power system, the cloud master station is responsible for receiving, processing and analyzing fault information from the terminal equipment.

[0008] 3) The APP filters alarms within the same time window and obtains a list of related FXU (feeder terminal unit) devices The cloud master sends the received fault information to the APP (mobile application). After receiving the alarm information, the APP first filters multiple alarm information within the same time window by substation. If there are multiple alarm information in the same substation, these alarm information are merged to reduce redundancy. Then, according to the device number in the alarm information, a list of all FXU (feeder terminal unit) devices in the same bus section under the substation to which the device belongs is obtained. FXU is an intelligent device in the power system used to monitor and control the operating status of the feeder.

[0009] 4) APP releases the message of retrieving recorded waveforms After obtaining the list of related FXU devices, the APP publishes a message to retrieve the recorded waveform. The message uses the starting time point T0 as the reference timestamp and sets the number of cycles before and after (configurable as needed). The number of cycles refers to the number of waves contained in a complete sine wave cycle. By setting the number of cycles before and after, it can be ensured that the recorded waveform data contains complete waveform information before and after the fault occurs.

[0010] 5) The terminal receives the subscription message for recording and uploads the recording file After receiving the waveform subscription message, the terminal device retrieves the corresponding waveform data according to the requirements in the message and generates a waveform file. The waveform file is a file containing waveform data, which can be used for subsequent analysis and judgment. Then, the terminal device uploads the waveform file to the file server for subsequent processing and analysis.

[0011] 6) The terminal publishes a message that the recording file upload is complete and carries the file URL After the recording file is uploaded, the terminal device publishes a recording file upload completion message, which carries the URL (Uniform Resource Locator) of the file server. This message is used to notify the APP and other related systems that the recording file has been uploaded and is available for download and processing.

[0012] 7) The APP collects recorded file information and downloads the file After receiving the recording file upload completion message, the APP collects the recording file messages of all relevant substations and downloads the corresponding recording files. After the download is completed, the APP can process and analyze the recording files to extract fault feature information.

[0013] 8) APP packages the json file and uploads it to the file server After downloading and processing all relevant substation recording files, the APP needs to generate a json file containing the substation topology information. This file describes the connection relationship and operating status of each device in the substation. Then, the APP uploads the packaged json file to the file server for subsequent analysis.

[0014] 9) APP notifies the algorithm server to conduct research and judgment After the json file is uploaded, the APP notifies the algorithm server to conduct analysis. The algorithm server is a high-performance computing device or cluster in the power system, which is used to analyze and process the received data in real time. After receiving the analysis request, the algorithm server analyzes and judges the json file according to the preset algorithm and model to determine the fault location and type and other information.

[0015] 10) The algorithm server notifies the APP of the analysis results through the Kafka message queue After completing the analysis, the algorithm server sends the analysis results to the APP through the Kafka message queue. Kafka is a distributed stream processing platform that can be used to build real-time data pipelines and stream applications. Through the Kafka message queue, the algorithm server can send the analysis results to the APP in real time for subsequent processing and display.

[0016] 11) The APP saves the analysis results and sends warning SMS messages based on the analysis results After receiving the analysis results, the APP saves them to the database for subsequent query and analysis. At the same time, based on the fault location and type in the analysis results, the APP can push the corresponding alarm text message to the relevant personnel or system to remind them to take necessary measures to deal with the fault. For the arc suppression coil grounding system, if the analysis result is within the FXU protection zone, the APP will push a permanent grounding fault alarm text message; for the ungrounded system, the APP can retrieve the recording at T0+30s and push the alarm text message accordingly to further confirm the fault situation.

[0017] The present invention also provides a novel ground fault analysis and judgment system, comprising: A data acquisition module, for receiving real-time data from the power system, including but not limited to current, voltage, frequency and phase angle; Data preprocessing module, used to preprocess real-time data, including data cleaning, outlier removal and data standardization; A feature extraction module is used to extract feature factors to determine the possibility of ground faults. The feature factors include but are not limited to current imbalance, voltage fluctuation rate, historical fault records and environmental factors; The analysis module uses a high-performance algorithm server, combines the characteristic factors, and adopts machine learning or deep learning algorithms to conduct real-time analysis of ground faults and generate analysis results; The message queue module realizes data synchronization and status update between the algorithm server and the APP or monitoring system. It is also used to transmit the analysis results to the mobile application (APP) or monitoring system in real time through the message queue system, such as Kafka; The alarm processing module is used to filter the alarm information according to the preset rules in the APP or monitoring system according to the analysis results, and generate a ground fault alarm list; the preset rules include but are not limited to the alarm level, alarm frequency and alarm type; The user interaction module allows users to obtain the FXU (feeder terminal unit) device list through the APP, and automatically generates and packages a json file based on the analysis results. The file contains the fault point location information and its topological structure; The SMS service module pushes alarm information to relevant personnel or systems via SMS service when the analysis results confirm the existence of a ground fault, so that necessary measures can be taken in time.

[0018] The post-processing module is used to perform statistical analysis, trend prediction and fault mode identification on the research and judgment results.

[0019] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention improves the accuracy and reliability of the judgment by comprehensively considering multiple characteristic factors (such as 3U0, 3I0, recording data, etc.). At the same time, by acquiring and processing data in real time and using a high-performance algorithm server for judgment, real-time detection and rapid response to ground faults are achieved; (2) The present invention reduces redundant information and improves processing efficiency by filtering alarm information within the same time window through APP and obtaining a list of related FXU devices. At the same time, by packaging json files and uploading them to a file server, it provides rich topology information support for subsequent research and analysis; (3) The present invention realizes real-time communication and data transmission between the algorithm server and the APP through the Kafka message queue, which improves the scalability and flexibility of the system. At the same time, by pushing alarm text messages, it reminds relevant personnel or systems to take necessary measures to deal with faults in a timely manner, further improving the security and stability of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 The figure is a flow chart of a novel ground fault analysis method in an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The specific implementation modes of the present invention are described in detail below with reference to the accompanying drawings.

[0022] Example 1 like Figure 1 As shown, this embodiment is described by taking the grounding fault analysis of a certain power system as an example.

[0023] First, a 3U0 or 3I0 over-limit detection module is set on the terminal device of the power system. When it is detected that 3U0 or 3I0 exceeds the preset threshold, it is considered that a ground fault may have occurred and the judgment process is started (step S1). At this time, the starting time point T0 is recorded and the relevant information is saved.

[0024] Then, if the fault persists for a set time (such as 5 seconds) and is not eliminated, the terminal device reports the fault event and the starting time point T0 to the cloud master (step S2). The cloud master receives and processes this information and sends it to the APP.

[0025] After receiving the alarm information, the APP first filters multiple alarm information within the same time window by substation (step S3). Then, according to the device number in the alarm information, it obtains a list of all FXU devices in the same bus section of the substation to which the device belongs (step S4).

[0026] Next, the APP issues a message to retrieve the recorded wave (step S5). The message uses the starting time point T0 as the reference timestamp and sets the number of cycles before and after (e.g., 5 cycles before and after). Then, the terminal device is waited for uploading the recorded wave file.

[0027] After receiving the subscription message for retrieving the wave recording, the terminal device retrieves the corresponding wave recording data according to the requirements and generates a wave recording file (step S6). Then the wave recording file is uploaded to the file server (step S7). After completion, a wave recording file upload completion message is published with the URL of the file server (step S8).

[0028] After receiving the recording file upload completion message, the APP collects the recording file messages of all relevant substations and downloads the corresponding recording files (step S9). The recording files are then processed and analyzed to extract fault feature information.

[0029] After the processing is completed, the APP generates a json file containing the substation topology information (step S10), and then uploads the packaged json file to the file server (step S11).

[0030] Then the APP notifies the algorithm server to conduct analysis (step S12). After receiving the analysis request, the algorithm server analyzes and analyzes the recorded data and json file according to the preset algorithm and model to determine the fault location and type and other information (step S13).

[0031] After the analysis is completed, the algorithm server sends the analysis results to the APP through the Kafka message queue (step S14). After receiving the analysis results, the APP saves them to the database (step S15). At the same time, according to the fault location and type in the analysis results, the corresponding alarm SMS is pushed to the relevant personnel or system (step S16).

[0032] In this embodiment, the accuracy and reliability of the judgment are improved by comprehensively considering multiple characteristic factors; real-time detection and rapid response to ground faults are achieved by acquiring and processing data in real time and using a high-performance algorithm server for judgment; processing efficiency is improved by filtering alarm information through APP, obtaining FXU device lists, packaging json files, etc.; real-time communication and data transmission between the algorithm server and APP are achieved through the Kafka message queue, which improves the scalability and flexibility of the system; and relevant personnel or systems are reminded to take necessary measures to handle faults in a timely manner by pushing alarm text messages, which further improves the security and stability of the system.

[0033] In the specific implementation process, the present invention can be further optimized and improved according to actual needs. For example, more characteristic factors can be introduced for research and judgment to improve the accuracy and reliability of research and judgment; algorithms and models can be optimized to improve the speed and efficiency of research and judgment; more intelligent technologies and means, such as machine learning and deep learning, can be introduced to further improve the intelligence level and automation of research and judgment.

[0034] In addition, the present invention can also be combined with other power system intelligent technologies, such as smart grids, smart substations, etc., to achieve more comprehensive and efficient power system fault monitoring and analysis. For example, the analysis and judgment system and method of the present invention can be integrated into the monitoring system of the smart grid to achieve real-time monitoring and rapid response to power system faults; the analysis and judgment results of the present invention can be integrated and analyzed with the information of other intelligent devices to further improve the accuracy and reliability of the analysis and judgment.

[0035] In summary, the present invention proposes a new ground fault analysis method, which improves the accuracy and efficiency of the analysis by comprehensively considering multiple characteristic factors, acquiring and processing data in real time, using high-performance algorithm servers for analysis, filtering alarm information through APP, obtaining FXU equipment lists, packaging json files, etc., reduces the false alarm and missed alarm rates, and provides a strong guarantee for the stable operation of the power system. At the same time, the present invention also has the advantages of scalability, flexibility and high intelligence level, and can be combined with other power system intelligent technologies to achieve more comprehensive and efficient power system fault monitoring and analysis.

[0036] The above description is only a preferred embodiment of the present invention and does not constitute a limitation on the protection scope of the present invention. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

1. A new ground fault analysis method, characterized in that: The following steps are involved: Step 1: On the terminal equipment of the power system, a 3U0 or 3I0 over-limit detection module is set. When it is detected that 3U0 or 3I0 exceeds the preset threshold, the analysis and judgment process is started and the starting time point T0 is recorded; Step 2: After starting the analysis process, if the fault continues for the set time and is not eliminated, the terminal device reports the fault event and the starting time point T0 to the cloud master station; Step 3: The cloud master sends the received fault information to the APP, which filters the alarms within the same time window and obtains the list of related FXU devices; Step 4: APP publishes the message of retrieving the recorded waveform; Step 5: The terminal device receives the recording message and uploads the recording file to the file server; Step 6: After the recording file is uploaded, the terminal device publishes a recording file upload completion message, which carries the URL of the file server; Step 7: After receiving the recording file upload completion message, the APP collects the recording file messages of all relevant substations and downloads the corresponding recording files; Step 8: After downloading and processing the recording files of all relevant substations, the APP first generates a json file containing the substation topology information, and then uploads the packaged json file to the file server for subsequent analysis; Step 9: After the JSON file is uploaded, the APP notifies the algorithm server to conduct analysis; Step 10: The algorithm server notifies the APP of the analysis results through the Kafka message queue; Step 11: The APP saves the analysis results and pushes an alarm SMS based on the analysis results.

2. A new ground fault analysis method according to claim 1, characterized in that: In step 2, the cloud master station is the monitoring and management center of the power system, responsible for receiving, processing and analyzing fault information from terminal devices.

3. A new ground fault analysis method according to claim 1, characterized in that: In the step three, the cloud master sends the received fault information to the APP. After receiving the alarm information, the APP first filters multiple alarm information within the same time window by substation. If there are multiple alarm information in the same substation, these alarm information are merged to reduce redundancy; then, a list of all FXU devices in the same bus section under the substation to which the device belongs is obtained according to the device number in the alarm information; FXU is an intelligent device in the power system, which is used to monitor and control the operating status of the feeder.

4. A new ground fault analysis method according to claim 1, characterized in that: In the step 4, after obtaining the list of relevant FXU devices, the APP publishes a message to retrieve the recorded waveform. The recorded waveform message uses the starting time point T0 as the reference timestamp and sets the number of cycles before and after. The number of cycles refers to the number of waves contained in a complete sinusoidal wave cycle. By setting the number of cycles before and after, it can be ensured that the recorded waveform data contains complete waveform information before and after the fault occurs.

5. A new ground fault analysis method according to claim 1, characterized in that: In step five, after receiving the recording message, the terminal device retrieves the corresponding recording data according to the requirements in the message and generates a recording file; the recording file is a file containing waveform data, which is used for subsequent analysis and judgment; then, the terminal device uploads the recording file to the file server for subsequent processing and analysis.

6. A new ground fault analysis method according to claim 1, characterized in that: In step nine, the algorithm server is a high-performance computing device or cluster in the power system, which is used to perform real-time analysis and processing on the received data. After receiving the analysis request, the algorithm server analyzes and analyzes the json file according to the preset algorithm and model to determine the fault location and type information.

7. A new ground fault analysis method according to claim 1, characterized in that: In the step 10, after completing the analysis, the algorithm server sends the analysis result to the APP through the Kafka message queue; through the Kafka message queue, the algorithm server can send the analysis result to the APP in real time for subsequent processing and display.

8. A new ground fault analysis method according to claim 1, characterized in that: In the step eleven, after receiving the analysis result, the APP saves it to the database for subsequent query and analysis. At the same time, according to the fault location and type information in the analysis result, the APP pushes the corresponding alarm text message to the relevant personnel or system to remind them to take necessary measures to deal with the fault; for the arc suppression coil grounding system, if the analysis result is within the FXU protection zone, the APP pushes a permanent grounding fault alarm text message; for the ungrounded system, the APP can retrieve the recording at T0+30s and push an alarm text message accordingly to further confirm the fault situation.

9. A new type of ground fault analysis system, characterized in that: include: A data acquisition module for receiving real-time data from the power system, including current, voltage, frequency and phase angle; Data preprocessing module, used to preprocess real-time data, including data cleaning, outlier removal and data standardization; A feature extraction module is used to extract feature factors and determine the possibility of ground faults. The feature factors include current imbalance, voltage fluctuation rate, historical fault records and environmental factors. The analysis module uses a high-performance algorithm server, combines the characteristic factors, and adopts machine learning or deep learning algorithms to conduct real-time analysis of ground faults and generate analysis results; The message queue module is used to synchronize data and update status between the algorithm server and the APP or monitoring system, and pass the judgment results through the message queue system; The alarm processing module is used in the APP or monitoring system to filter the alarm information according to the preset rules based on the analysis results and generate a ground fault alarm list; the preset rules include the alarm level, alarm frequency and alarm type; The user interaction module allows users to obtain the FXU device list through the APP, and automatically generates and packages a json file based on the analysis results. The file contains the fault point location information and its topological structure; SMS service module: when the analysis results confirm that there is a ground fault, the SMS service will push the alarm information to the relevant personnel or system so that necessary measures can be taken in time; The post-processing module is used to perform statistical analysis, trend prediction and fault mode identification on the research and judgment results.

Images