Recording fault reason identification system

Abstract

The invention belongs to the technical field of power grid fault identification of a power system, and particularly relates to a recording fault reason identification system which comprises a data preprocessing module, a data processing module, a fault reason identification module and an auxiliary decision module. The recording fault reason identification system can accurately diagnose the fault and timely process the fault, reduce the influence of the fault on the operation of the power grid, improve the safety and reliability of the operation of the power grid, provide powerful support for the future fault prevention and treatment by establishing a typical fault reason library, further guarantee the stable operation of the power grid, and improve the reliability of the power grid. According to the method and the system, the information of a plurality of data sources can be fused, the running state of the equipment when the fault occurs can be comprehensively known, intelligent support is provided for fault processing, the fault type time with the maximum correlation degree can be automatically output through the fault auxiliary decision-making module, the processing principle is prompted, manual intervention is reduced, and the automation level of fault processing is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system grid fault identification, and particularly to a fault cause identification system for oscillogram. Background Technique

[0002] With the increasing expansion of the power grid scale, the requirements for the power grid fault diagnosis system are also getting higher and higher. A perfect power grid fault information comprehensive analysis system needs to be able to accurately perform functions such as fault component diagnosis, post-accident data analysis, and protection action behavior evaluation. This plays a very important role in the safe and reliable operation of the power system. Therefore, a power grid fault diagnosis system based on fault oscillogram information has emerged, which uses rich oscillogram information to provide a basis for further diagnosis.

[0003] Currently, in the prior art, when a fault occurs in the power system, there is a technology that can automatically and accurately record the changes in various electrical quantities during the process before and after the fault. These electrical quantities include current, voltage, power, etc., and their changes play an important role in analyzing the fault cause, judging whether the protection action is correct, and improving the safe operation level of the power system. However, there is a lack of automatically outputting fault handling principles and recommended measures through autonomous learning, combining fault handling experience and expert knowledge base. In view of this, the present invention proposes a fault cause identification system for oscillogram. Summary of the Invention

[0004] The main object of the present invention is to provide a fault cause identification system for oscillogram, which can solve the problems raised in the above background technique.

[0005] To achieve the above object, the fault cause identification system for oscillogram proposed by the present invention includes:

[0006] A data preprocessing module, which is used to receive and store the original data recorded by the fault oscillogram device, and at the same time decompose the original data and convert it into electrical quantity characteristic values that can directly reflect the operating states of primary and secondary equipment;

[0007] A data processing module, which is used to preprocess the multi-system fault data of the power grid to ensure the data quality;

[0008] A fault cause identification module, which based on the fault historical data, establishes the corresponding relationship between the typical fault characteristic values and the fault causes, and at the same time uses artificial intelligence technology based on the neural network algorithm to construct a fault cause identification model with multiple characteristic quantities;

[0009] and an auxiliary decision-making module, which is used to extract the characteristic values of the electrical quantities of the line fault, input them into the multi-characteristic fault identification model, calculate the occurrence probability values of various fault causes, and output the fault type time with the highest correlation degree according to the occurrence probability values of the fault causes, and prompt the handling principle.

[0010] Preferably, the data preprocessing module includes a data reception and storage unit, a data decomposition and conversion unit, and a feature information unit.

[0011] Preferably, the data reception and storage unit is used to receive the original data recorded by the fault recorder in real time and store it safely and efficiently in the database for subsequent processing and analysis. The data decomposition and conversion unit decomposes the received original data in detail, extracts the key electrical quantity information, and converts it into electrical quantity characteristic values that can directly reflect the operating states of primary and secondary equipment. These electrical quantity characteristic values include current, voltage, frequency, etc. The feature information unit is used to comprehensively sort out and analyze the electrical quantity characteristic information of the fault recorder data, including time series analysis, feature value extraction, etc., providing a solid foundation for subsequent data fusion analysis and fault cause identification.

[0012] Preferably, the data processing module includes an information fusion unit, a feature extraction and analysis unit, and a wide-area fault information extraction unit.

[0013] Preferably, the information fusion unit is used for data model fusion, information fusion, and knowledge fusion technologies to integrate information from different data sources, including the fault recorder system, the protection information system, the meteorological system, etc., to form a comprehensive fault information view. The feature extraction and analysis unit is used to extract the typical characteristic values of various faults, including the magnitude of the fault current, the fault duration, the change of electrical quantities before and after the fault, etc., and analyze the nature, cause, and development trend of the line fault based on these characteristic values. The wide-area fault information extraction unit is used to pay attention to the primary and secondary fault characteristic quantity information of the substations on both sides of the fault line, the fault characteristic quantity information of other intervals within the substation area, and the power grid fault characteristic quantity information within the wide area, providing comprehensive data support for fault location, cause analysis, and handling.

[0014] Preferably, the fault cause identification module includes a rule establishment unit, a model construction unit, a model training and learning unit, and a fault cause library establishment unit.

[0015] Preferably, the rule establishment unit deeply analyzes the corresponding relationship between typical fault characteristic values and fault causes based on fault historical data, and establishes a mapping relationship between characteristic values and fault causes. The model construction unit constructs a fault cause identification model with multiple characteristic quantities based on artificial intelligence technologies such as neural network algorithms, such as convolutional neural network (CNN), recurrent neural network (RNN), etc., to achieve accurate identification of fault causes. The model training and learning unit conducts sample training and learning on the model through a large amount of historical fault data, continuously optimizing the parameters and structure of the model, and improving the accuracy and generalization ability of the model for fault cause identification. The fault cause library establishment unit is used to apply the trained model to the analysis of actual fault data, forming a typical fault cause library, providing a scientific basis for fault diagnosis and handling.

[0016] Preferably, the auxiliary decision-making module includes a fault data fusion analysis unit, an auxiliary characteristic value acquisition unit, and a fault handling principle prompt unit.

[0017] Preferably, the fault data fusion analysis unit is used to, when a fault occurs, fuse and analyze the data of the fault recording system and the protection information system in real time, extract the characteristic values of the line fault electrical quantities, and conduct correlation analysis with the data of other auxiliary systems, such as the meteorological system, the lightning location system, the wildfire warning system, etc. The auxiliary characteristic value acquisition unit is used to acquire the characteristic values related to the fault, such as meteorological conditions, lightning activity conditions, wildfire warning levels, etc., providing supplementary information for in-depth analysis of fault causes. The fault handling principle prompt unit can automatically output fault handling principles and recommended measures according to the fault probability and the fault type with the highest correlation degree, combined with fault handling experience and the expert knowledge base, providing scientific decision-making support for operation and maintenance personnel.

[0018] Preferably, after the fault data fusion analysis unit extracts the characteristic values of the line fault electrical quantities, the extracted characteristic values can be input into the multi-characteristic quantity fault identification model to calculate the occurrence probability values of various fault causes, providing strong support for accurate identification of fault causes.

[0019] The present invention provides a fault cause identification system for waveform recording. It has the following beneficial effects:

[0020] (1) This fault cause identification system for waveform recording can accurately diagnose faults and handle faults in a timely manner, while reducing the impact of faults on power grid operation, improving the safety and reliability of power grid operation, and through the establishment of a typical fault cause library, it can provide strong support for future fault prevention and treatment, further ensuring the stable operation of the power grid.

[0021] (2) The fault cause identification system for wave recording can integrate information from multiple data sources, comprehensively understand the operating state of equipment when a fault occurs, provide intelligent support for fault handling, and through the fault auxiliary decision-making module, can automatically output the fault type and time with the highest correlation degree, and prompt the handling principle, reducing manual intervention and improving the automation level of fault handling.

[0022] (3) The fault cause identification system for wave recording can comprehensively obtain fault information through data preprocessing and data fusion analysis, improve the accuracy of fault diagnosis, and through building a fault cause identification model, can quickly identify the fault cause and improve the diagnosis efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the structures shown in these drawings.

[0024] Figure 1 It is the system block diagram of the present invention;

[0025] Figure 2 It is the system flow chart of the present invention;

[0026] Figure 3 It is the block diagram of the data preprocessing module of the present invention;

[0027] Figure 4 It is the block diagram of the data processing module of the present invention;

[0028] Figure 5 It is the block diagram of the fault cause identification module of the present invention;

[0029] Figure 6 It is the block diagram of the auxiliary decision-making module of the present invention;

[0030] Figure 7 It is the system fault identification and handling flow chart of the present invention;

[0031] Figure 8 It is the schematic diagram of the processing equipment of the present invention.

[0032] The realization of the purpose, functional features and advantages of the present invention will be further described in conjunction with the embodiments with reference to the drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0034] Please refer to Figures 1-8 , the present invention proposes a fault cause identification system for oscillogram, including a data preprocessing module, a data processing module, a fault cause identification module, and an auxiliary decision-making module. Among them, the data preprocessing module plays a crucial role in the entire power system fault analysis process. Its core function is to deeply process the complex data recorded by the fault oscillograph device. This module will decompose these original fault oscillogram data one by one according to specific rules and algorithms. During the decomposition process, it will use professional power system analysis knowledge and advanced data processing technologies to accurately convert the original oscillogram data into electrical quantity characteristic values that can directly reflect the operating states of primary and secondary equipment. Through such a processing method, we can comprehensively and deeply obtain the electrical quantity characteristic information hidden in the fault oscillogram data, providing a solid data foundation for the stable operation and fault repair of the entire power system; the data processing module applies technologies such as time series data processing technology, big data analysis methods, and machine learning algorithms to preprocess the fault data of multiple power grid systems to ensure data quality. Through technologies such as data model fusion, information fusion, and knowledge fusion, it integrates the information of multiple data sources to comprehensively understand the operating states of primary and secondary equipment during the fault occurrence. At the same time, it can extract the typical characteristic values of various faults, display the nature, cause, and development trend of the line fault, extract the primary and secondary fault characteristic quantity information of the substations on both sides of the fault line, the fault characteristic quantity information of other intervals within the substation area, and the fault characteristic quantity information of the power grid within the wide area; the fault cause identification module can establish the corresponding relationship between the typical fault characteristic values and the fault causes based on the fault historical data, and use artificial intelligence technology based on the neural network algorithm to construct a fault cause identification model with multiple characteristic quantities. Through sample training and learning of the model with a large amount of historical fault data, the correlation relationship between the input characteristic quantities and the fault causes is obtained, forming a typical fault cause library, providing a basis for fault diagnosis; the auxiliary decision-making module is used to extract the characteristic values of the electrical quantities of the line fault, input them into the multi-characteristic quantity fault identification model, calculate the occurrence probability values of various fault causes, and output the fault type time with the highest correlation degree according to the occurrence probability values of the fault causes, and prompt the processing principle.

[0035] In an embodiment of the present invention, the data preprocessing module includes a data reception and storage unit, a data decomposition and conversion unit, and a feature information unit. The data reception and storage unit is used to receive in real time the original data recorded by the fault recorder and store it safely and efficiently in the database for subsequent processing and analysis. The data decomposition and conversion unit decomposes the received original data in detail, extracts key electrical quantity information, and converts it into electrical quantity characteristic values that can directly reflect the operating states of primary and secondary equipment. These electrical quantity characteristic values include current, voltage, frequency, etc. The feature information unit is used to comprehensively sort out and analyze the electrical quantity feature information of the fault recorder data, including time series analysis, feature value extraction, etc., providing a solid foundation for subsequent data fusion analysis and fault cause identification. In this way, the original data of the fault recorder can be received efficiently and stored safely, and through detailed decomposition and conversion, the electrical quantity characteristic values that can directly reflect the equipment operating state can be extracted, providing an accurate and comprehensive information basis for subsequent data fusion analysis, fault cause identification, etc., thereby improving the accuracy and efficiency of fault analysis. By constructing a fault cause identification model, the fault cause can be quickly identified and the diagnosis efficiency can be improved.

[0036] Furthermore, the data processing module includes an information fusion unit, a feature extraction and analysis unit, and a wide-area fault information extraction unit. The information fusion unit is used for data model fusion, information fusion, and knowledge fusion technologies to integrate information from different data sources, including the fault recorder system, the protection information system, the meteorological system, etc., to form a comprehensive view of fault information. The feature extraction and analysis unit is used to extract typical feature values of various faults, including the magnitude of the fault current, the fault duration, the change in electrical quantities before and after the fault, etc., and analyze the nature, cause, and development trend of the line fault based on these feature values. The wide-area fault information extraction unit is used to focus on the primary and secondary fault characteristic quantity information of the substations on both sides of the fault line, the fault characteristic quantity information of other intervals within the substation area, and the power grid fault characteristic quantity information within the wide area, providing comprehensive data support for fault location, cause analysis, and processing. In this way, the information of multiple data sources can be fused, the operating state of the equipment during the fault occurrence can be comprehensively understood, intelligent support for fault handling can be provided, and through the fault auxiliary decision-making module, the fault type time with the highest correlation can be automatically output, and the processing principle can be prompted, reducing manual intervention and improving the automation level of fault handling.

[0037] Furthermore, the fault cause identification module includes a rule establishment unit, a model construction unit, a model training and learning unit, and a fault cause database establishment unit. The rule establishment unit deeply analyzes the correspondence between typical fault characteristic values and fault causes based on historical fault data, and establishes a mapping relationship between characteristic values and fault causes. The model construction unit constructs a fault cause identification model with multiple characteristic quantities based on artificial intelligence technologies such as neural network algorithms, such as convolutional neural network (CNN), recurrent neural network (RNN), etc., to achieve accurate identification of fault causes. The model training and learning unit performs sample training and learning on the model through a large amount of historical fault data, continuously optimizes the parameters and structure of the model, and improves the accuracy and generalization ability of the model for fault cause identification. The fault cause database establishment unit is used to apply the trained model to the analysis of actual fault data, form a typical fault cause database, provide a scientific basis for fault diagnosis and handling, enable accurate fault diagnosis and timely fault handling, reduce the impact of faults on power grid operation, and improve the safety and reliability of power grid operation.

[0038] Furthermore, the auxiliary decision-making module includes a fault data fusion analysis unit, an auxiliary characteristic value acquisition unit, and a fault handling principle prompt unit. The fault data fusion analysis unit is used to, when a fault occurs, fuse and analyze the data of the fault recorder system and the protection information system in real time, extract the characteristic values of the electrical quantities of the line fault, and conduct correlation analysis with the data of other auxiliary systems, such as the meteorological system, the lightning location system, the wildfire warning system, etc. After the fault data fusion analysis unit extracts the characteristic values of the electrical quantities of the line fault, the extracted characteristic values can be input into the multi-characteristic quantity fault identification model to calculate the occurrence probability values of various fault causes, providing strong support for the accurate identification of fault causes. The auxiliary characteristic value acquisition unit is used to acquire characteristic values related to the fault, such as meteorological conditions, lightning activity conditions, wildfire warning levels, etc., providing supplementary information for the in-depth analysis of fault causes. The fault handling principle prompt unit can automatically output fault handling principles and recommended measures according to the fault probability and the fault type with the highest correlation degree, combined with fault handling experience and the expert knowledge base, providing scientific decision-making support for operation and maintenance personnel. By establishing a typical fault cause database, it can provide strong support for future fault prevention and treatment, further ensure the stable operation of the power grid, and at the same time can fuse and analyze the fault data from multiple systems in real time, including the fault recorder system, the protection information system, and auxiliary systems such as meteorology, lightning location, and wildfire warning. By extracting key characteristic values and inputting them into the multi-characteristic quantity fault identification model, accurately identify the fault cause, and at the same time, combined with auxiliary characteristic values and the expert knowledge base, automatically output scientific fault handling principles and recommended measures, providing fast and accurate decision-making support for operation and maintenance personnel, effectively improving the efficiency and accuracy of fault handling.

[0039] The above are only the preferred embodiments of the present invention, and do not thereby limit the patent scope of the present invention. Any equivalent structural transformation made by using the content of the specification and drawings of the present invention under the inventive concept of the present invention, or any direct / indirect application in other related technical fields is included in the patent protection scope of the present invention.

Claims

1. A recording fault cause identification system, characterized in that: include: A data preprocessing module, which is used to receive and store the original data recorded by the fault recording device, and decompose the original data and convert it into electrical quantity characteristic values ​​that can directly reflect the operating status of the primary and secondary equipment; A data processing module, which is used to pre-process the fault data of multiple power grid systems to ensure data quality; A fault cause identification module, which establishes a correspondence between typical fault characteristic values ​​and fault causes based on fault history data, and uses artificial intelligence technology based on a neural network algorithm to build a fault cause identification model with multiple characteristic quantities; And an auxiliary decision-making module, which is used to extract the characteristic values ​​of the electrical quantities of the line fault, input them into the multi-characteristic fault identification model, calculate the probability values ​​of various fault causes, and output the time of the fault type with the greatest correlation based on the probability values ​​of the fault causes, and suggest the processing principles.

2. A recording fault cause identification system according to claim 1, characterized in that: The data preprocessing module includes a data receiving and storing unit, a data decomposing and converting unit and a characteristic information unit.

3. A recording fault cause identification system according to claim 2, characterized in that: The data receiving and storage unit is used to receive the original data recorded by the fault recording device in real time, and store it safely and efficiently in the database for subsequent processing and analysis. The data decomposition and conversion unit performs detailed decomposition of the received original data, extracts key electrical quantity information, and converts it into electrical quantity characteristic values ​​that can directly reflect the operating status of primary and secondary equipment. The characteristic information unit is used to comprehensively organize and analyze the electrical quantity characteristic information of the fault recording data, including time series analysis, characteristic value extraction, etc., to provide a solid foundation for subsequent data fusion analysis and fault cause identification.

4. A recording fault cause identification system according to claim 1, characterized in that: The data processing module includes an information fusion unit, a feature extraction and analysis unit and a wide-area fault information extraction unit.

5. A recording fault cause identification system according to claim 4, characterized in that: The information fusion unit is used for data model fusion, information fusion and knowledge fusion technology, integrating information from different data sources to form a comprehensive fault information view. The feature extraction and analysis unit is used to extract typical characteristic values ​​of various faults, and analyze the nature, cause and development trend of line faults based on these characteristic values. The wide-area fault information extraction unit is used to focus on the primary and secondary fault characteristic quantity information of substations on both sides of the fault line, the characteristic quantity information of other interval faults in the station area, and the characteristic quantity information of power grid faults in the wide area, providing comprehensive data support for fault location, cause analysis and processing.

6. A recording fault cause identification system according to claim 1, characterized in that: The fault cause identification module includes a rule establishing unit, a model building unit, a model training and learning unit and a fault cause library establishing unit.

7. A recording fault cause identification system according to claim 6, characterized in that: The rule establishing unit deeply analyzes the correspondence between typical fault characteristic values ​​and fault causes based on fault history data, and establishes a mapping relationship between characteristic values ​​and fault causes. The model building unit builds a fault cause identification model with multiple characteristic quantities based on artificial intelligence technology of neural network algorithm to achieve accurate identification of fault causes. The model training and learning unit performs sample training and learning on the model through a large amount of historical fault data, continuously optimizes the parameters and structure of the model, and improves the accuracy and generalization ability of the model in identifying fault causes. The fault cause library establishing unit is used to apply the trained model to the analysis of actual fault data to form a typical fault cause library, providing a scientific basis for fault diagnosis and processing.

8. A recording fault cause identification system according to claim 1, characterized in that: The auxiliary decision-making module includes a fault data fusion analysis unit, an auxiliary feature value acquisition unit and a fault handling principle prompting unit.

9. A recording fault cause identification system according to claim 8, characterized in that: The fault data fusion and analysis unit is used to fuse and analyze the data of the fault recording system and the protection information system in real time when a fault occurs, extract the characteristic values ​​of the line fault electrical quantities, and perform correlation analysis with the data of other auxiliary systems. The auxiliary characteristic value acquisition unit is used to obtain characteristic values ​​related to the fault to provide supplementary information for in-depth analysis of the cause of the fault. The fault handling principle prompting unit can automatically output fault handling principles and recommended measures based on the fault probability and the time of the fault type with the highest correlation, combined with fault handling experience and expert knowledge base, to provide scientific decision support for operation and maintenance personnel.

10. A recording fault cause identification system according to claim 9, characterized in that: After the fault data fusion analysis unit extracts the characteristic values ​​of the line fault electrical quantities, the extracted characteristic values ​​can be input into the multi-characteristic fault identification model to calculate the occurrence probability values ​​of various fault causes, providing strong support for the accurate identification of the fault causes.

Images