Power grid fault chain identification method and system

Abstract

The invention relates to the field of power systems, in particular to a power grid fault chain identification method and system, and the method comprises the steps: obtaining alarm information and WAMS synchronous measurement data in a power grid; acquiring a rule base and a power grid topology model; based on the power grid topology model and the WAMS synchronous measurement data, correlation analysis is carried out on fault events among the multiple fault events generating the alarm information through a rule base, correlation analysis comprises at least one of a time type, a space type and an electrical type, and the causal relationship among the fault events is recorded in the rule base.

Description

Technical Field

[0001] The present invention relates to the field of power systems, and more specifically, to a method and system for identifying chain reactions of power grid faults. Background Art

[0002] As the scale of power systems continues to expand and the grid structure becomes increasingly complex, the risks faced by grid operations are also gradually increasing. Especially in large-scale interconnected grids, local faults may cause large-scale power outages through chain reactions in the grid, causing serious impacts on the society and economy.

[0003] Traditional fault identification and analysis methods mainly rely on manual experience and offline post-fault analysis. These methods usually monitor the operating parameters of the power system, such as voltage, current and frequency, and combine the experience of the dispatcher to make preliminary judgments and handle faults. Some regions have also adopted fault monitoring methods based on SCADA (Supervisory Control and Data Acquisition System) and EMS (Energy Management System), but these systems have certain limitations in the real-time and accuracy of data.

[0004] In recent years, with the development of wide area measurement systems (WAMS), the dynamic monitoring capabilities of power systems have been significantly improved. WAMS uses synchronized phasor measurement units (PMUs) to obtain real-time data from each node of the power grid, providing a new means for fault identification. Some research institutions have proposed disturbance identification and low-frequency oscillation detection methods based on WAMS data, aiming to promptly detect abnormal events in the power grid.

[0005] Although the above methods have improved the efficiency of fault identification to a certain extent, they still have the following shortcomings: Lack of real-time performance: Traditional fault analysis methods often cannot meet the real-time requirements of the power grid and cannot capture the chain reaction and propagation path of faults in a timely manner; Lack of correlation analysis: Existing methods mainly focus on the detection of single fault events, lack in-depth analysis of the temporal, spatial and electrical correlations between faults, and cannot accurately identify chain failures; Insufficient data utilization: Although WAMS provides rich real-time data, existing methods do not make high use of these data and fail to fully mine the correlation information in the data; Incomplete rule base: The lack of a systematic fault causal relationship rule base limits the accuracy and reliability of fault analysis. Summary of the invention

[0006] The purpose of the present invention is to overcome the shortcomings of the prior art and propose a method and system for identifying chain failures in power grids. By establishing a chain failure experience knowledge base and combining it with real-time power grid operation data, the chain failures are analyzed and identified in multiple dimensions, and a chain failure analysis report is output in a timely manner. The method can improve the real-time and accuracy of chain failure identification, provide effective decision-making support for power grid operators, prevent the expansion and spread of accidents, and improve the safety and reliability of the power grid.

[0007] According to one aspect of the present invention, a method for identifying chain reactions of power grid faults is provided, comprising: obtaining alarm information and WAMS synchronous measurement data within the power grid; obtaining a rule base and a power grid topology model; based on the power grid topology model and the WAMS synchronous measurement data, using the rule base to perform correlation analysis on fault events between multiple fault events that generate alarm information, wherein the correlation analysis may include the time dimension, spatial location or electrical characteristics of the event, and analyzing at least one type of correlation relationship, and the rule base records the causal relationship between the fault events.

[0008] Furthermore, the method further includes: in the case of time type analysis, taking a specific fault event that generates alarm information as a starting point, searching forward and backward for fault events before and after the specific fault event according to a predetermined time window to form a fault event set.

[0009] Furthermore, the method further includes: in the case of spatial type analysis, taking the node where a specific fault event among multiple fault events that generate alarm information occurs as the fault node; analyzing the distance and impact degree between the fault node and other nodes in the power grid topology model.

[0010] Furthermore, the method further includes: in the case of electrical type analysis, using WAMS synchronous measurement data to calculate changes in power grid flow before and after the fault and identify electrical correlations.

[0011] Furthermore, the method further includes: further includes: visually displaying the results of the correlation analysis.

[0012] According to another aspect of the present invention, there is provided a power grid fault chain identification system, comprising: a data interface module: used to obtain alarm information and WAMS synchronous measurement data within the power grid; a model and rule management module: used to obtain a rule base and a power grid topology model; a fault analysis module: based on the power grid topology model and WAMS synchronous measurement data, using the rule base to perform correlation analysis on fault events between multiple fault events that generate alarm information, wherein the correlation analysis includes at least one of a time type, a space type and an electrical type, and the rule base records the causal relationship between the fault events.

[0013] Furthermore, in the case of the time type, the fault analysis module takes a specific fault event among multiple fault events that generate alarm information as a starting point, searches for fault events before and after the specific fault event according to a predetermined time window, and forms a fault event set.

[0014] Furthermore, in the case of a spatial type, the fault analysis module takes the node where a specific fault event among multiple fault events generating alarm information occurs as the fault node, and analyzes the distance and impact degree between the fault node and other nodes in the topology model.

[0015] Furthermore, in the case of electrical type, the fault analysis module uses WAMS synchronous measurement data to calculate the changes in power flow before and after the fault.

[0016] Furthermore, the system further includes a visualization module, which visualizes the results of the correlation analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 2 is a schematic diagram showing the structure of a power grid fault chain identification system according to an embodiment of the present invention.

[0018] Figure 2 is a schematic diagram showing a fault chain reasoning rule path according to an embodiment of the present invention.

[0019] Figure 3 is a logic diagram illustrating the processing of cascading failure disturbance identification alarm information flow according to an embodiment of the present invention.

[0020] Figure 4 The following is a logic diagram showing a commutation failure alarm caused by a DC short-circuit fault in a short area according to an embodiment of the present invention.

[0021] Figure 5 is a logic diagram showing that a commutation failure causes a transient low voltage and a transient overvoltage in a power grid at the opposite end according to an embodiment of the present invention.

[0022] Figures 6 to 8 1 is a flow chart showing a method for identifying power grid fault cascading according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] Embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0024] Figure 1 2 is a schematic diagram showing the structure of a power grid fault chain identification system according to an embodiment of the present invention.

[0025] Reference Figure 1The power grid fault chain identification system (hereinafter referred to as “system”) 100 of the present invention may include a data interface module 110 , a model and rule management module 120 , and a fault analysis module 130 .

[0026] The data interface module 110 may be configured to be responsible for data interaction with WAMS, SCADA and other business systems to obtain real-time measurement data and alarm information. Therefore, the data interface module 110 may obtain alarm information in the power grid and WAMS synchronous measurement data.

[0027] The real-time measurement data (also referred to as “WAMS synchronous measurement data”) may include voltage, current, frequency, power angle, and active power of a node, but the present invention is not limited thereto.

[0028] The alarm information may include alarm information output by online disturbance identification of the entire network, low-frequency oscillation alarm information, alarm information output by related business applications (such as protection signal applications), etc., but the invention is not limited to this.

[0029] The model and rule management module 120 may be configured to maintain the power grid topology model and the causal rule base, and support dynamic update and expansion of the power grid topology model and the causal rule base.

[0030] The power grid topology model may include the location, direction, distance, etc. of each node, but the present invention is not limited thereto.

[0031] The causal relationship between fault events may be recorded in the causal rule base, which may be established based on historical experience and expert knowledge to describe the possible causal relationship between various events.

[0032] Therefore, the rule base can be used to perform causal reasoning on fault events and identify the chain relationship and propagation path of faults.

[0033] Figure 2 The fault chain reasoning rule path according to the present invention is shown, which shows the causal relationship between fault events and the reasoning path based on the rule base. In order to facilitate the description and matching of rules, various fault events and risk states are encoded, such as unit tripping (E1), line tripping (E2), transformer tripping (E3), etc. However, the reasoning rules of the present invention are not limited to Figure 2 The content shown can be expanded and adjusted according to actual application requirements to cover more types of fault events and causal relationships.

[0034] Return to reference Figure 1 The fault analysis module 130 may be configured to perform accident correlation analysis, risk state identification, and causal reasoning to generate analysis results.

[0035] First, the fault analysis module 130 can perform correlation analysis on the fault events between the multiple fault events that generate alarm information based on the power grid topology model and the WAMS synchronous measurement data using a rule base.

[0036] The correlation analysis may include at least one of time correlation analysis, space correlation analysis and electrical correlation analysis.

[0037] In the case of time correlation analysis, the fault analysis module 130 can start from a major disturbance event and calculate the time interval according to [TT forward , T+T back ] time window, search for related accidents before and after, and construct the initial accident set, where T is the time point of the major disturbance event, T forward To search forward for the length of time, search for the preceding triggering event of the current major disturbance event within this time period, such as T forward It can be 30 minutes, but the present invention is not limited thereto; T back To search backwards for the time length, since the data time stamps from different data acquisition devices may have errors in the second level, we search for trigger events that may trigger the current major disturbance event at a similar time within this time period, such as T back It may be 1 second, but the present invention is not limited thereto.

[0038] In the case of spatial correlation analysis, the fault analysis module 130 can use the distribution characteristics of the voltage mutation amount to determine the scope of the fault. Specifically, the fault analysis module 130 can analyze the distance and impact degree between the fault point and other nodes according to the grid topology.

[0039] In the case of electrical correlation analysis, the fault analysis module 130 can use WAMS data to calculate the changes in power flow before and after the fault, and identify power flow transfer and possible overload lines caused by the fault.

[0040] Secondly, the fault analysis module 130 can identify the risk status of equipment in the power grid.

[0041] Specifically, the fault analysis module 130 monitors the operating status of the equipment based on real-time data and power grid topology, and identifies possible risk states.

[0042] The risk status may include the following categories, but the present invention is not limited thereto.

[0043] Voltage over-limit: voltage exceeds the set threshold range; frequency over-limit: frequency deviates from the rated value beyond the allowable range; power angle over-limit: the generator power angle is close to an unstable state; power over-limit: the power of the line or transformer exceeds the rated capacity; oscillation state: abnormal oscillation phenomena such as ultra-low frequency, low frequency, sub / super synchronous oscillation are detected.

[0044] In the present invention, cascading failures can be analyzed in the following manner: Analyzing cascading failures may include alarm information flow processing, which includes: queue management, which is used to enter the alarm information into the processing queue in chronological order to form a first-in-first-out (FIFO) information flow; causal relationship judgment: for each alarm information, based on the correlation of time, space and electrical type, determine its causal relationship with the previous event; event merging and updating: based on the judgment results, update the cascading failure event table, merge related events, and form a complete cascading failure chain, such as Figure 3 shown.

[0045] Analyzing cascading failures may further include generating an analysis report, which includes collating the analysis results and generating a cascading failure analysis report, including the cause of the failure, the development process, the scope of impact, and the recommended treatment measures.

[0046] In the present invention, the specific implementation of the cascading failure judgment rule is as follows: DC fluctuation fault identification: By monitoring the changes in the AC side voltage of the converter station and the nearby area voltage, it is determined whether the fault occurs on the AC side or the DC side, and the fault conduction direction is determined.

[0047] Commutation failure caused by DC short circuit: Set specific judgment conditions, such as voltage drop amplitude, time window and topological connection, to identify commutation failure events caused by short circuit faults (such as Figure 4 As shown in FIG. 1 , it describes the judgment conditions and logic of commutation failure caused by DC short-circuit fault in the vicinity).

[0048] Phase-commutation failure causes abnormal voltage in the opposite power grid: Analyze the fluctuation characteristics of the voltage in the opposite converter station after the phase-commutation failure, and determine the impact of the fault on the opposite power grid (such as Figure 5 As shown in Figure 2, it describes the impact of commutation failure on the opposite power grid and the judgment logic of fault conduction).

[0049] Optionally, 100 may include a visualization display module 140 .

[0050] The visualization display module 140 may be configured to display the cascading failure composition, causal relationship and analysis report, such as a cascading failure composition table and a cascading failure event table, through a human-computer interaction interface. That is, the results of the correlation analysis may be visualized.

[0051] Figures 6 to 8 1 is a flow chart showing a method for identifying power grid fault cascading according to an embodiment of the present invention.

[0052] Reference Figure 6 , the power grid fault chain identification method according to the embodiment of the present invention may include: acquiring alarm information and WAMS synchronous measurement data in the power grid (S601).

[0053] As mentioned above, the alarm information may include alarm information output by online disturbance identification of the entire network, low-frequency oscillation alarm information, alarm information output by related business applications (such as protection signal applications), etc., but the invention is not limited to this.

[0054] As mentioned above, the WAMS synchronous measurement data may include the voltage, current, frequency, power angle and active power of the node, but the present invention is not limited thereto.

[0055] The method for identifying power grid fault cascading according to the embodiment of the present invention may further include: acquiring a rule base and a power grid topology model (S603).

[0056] As mentioned above, the causal relationship between fault events may be recorded in the causal rule base, which may be established based on historical experience and expert knowledge to describe the possible causal relationship between various events.

[0057] Therefore, the rule base can be used to perform causal reasoning on fault events and identify the chain relationship and propagation path of faults.

[0058] As mentioned above, the power grid topology model may include the location, direction, distance, etc. of each node, but the present invention is not limited thereto.

[0059] The method for identifying power grid fault chains according to an embodiment of the present invention may further include: based on the power grid topology model and WAMS synchronous measurement data, using a rule base to perform correlation analysis on fault events between multiple fault events that generate alarm information, wherein the correlation analysis includes at least one of a time type, a space type, and an electrical type (S605).

[0060] In the case of time correlation analysis, the major disturbance event can be used as the starting point, and related accidents can be searched forward and backward according to the time window of [T-predetermined time, T+predetermined time] to construct an initial accident set, where T is the time point of the major disturbance event. For example, the predetermined time can be 30 minutes, but the present invention is not limited to this.

[0061] In the case of spatial correlation analysis, the distribution characteristics of voltage mutation can be used to determine the scope of the fault. Specifically, the distance and impact degree between the fault point and other nodes can be analyzed based on the grid topology.

[0062] In the case of electrical correlation analysis, WAMS data can be used to calculate the changes in grid power flow before and after the fault, and identify power flow transfer and possible overloaded lines caused by the fault.

[0063] Figure 7 and Figure 8 A schematic flow chart of a method for identifying power grid fault cascading faults according to an embodiment of the present invention is also shown.

[0064] exist Figure 7 and Figure 8 In the figure, the processes of data acquisition, correlation analysis and result output are shown. Since they have been described above, they will not be repeated here.

[0065] exist Figure 8 In the table, “TimeRelated” indicates time correlation, “TopoRelated” indicates spatial correlation, “EventTypeRelated” indicates electrical correlation, “Y” indicates “yes”, “N” indicates “no”, and “True” indicates “true”.

[0066] The present invention comprehensively analyzes the correlation of fault events from the three dimensions of time, space and electricity, and accurately identifies the chain reaction of faults; fully utilizes the synchronous measurement data of WAMS, combined with the power grid topology model, to achieve real-time monitoring and analysis of faults; establishes a detailed fault event coding and causal rule library to support the reasoning and identification of complex faults; designs an efficient alarm information flow processing logic, which can quickly locate related events in massive data and generate a chain of chain faults; not only pays attention to explicit fault events, but also can identify the continuous risk status of equipment to prevent potential accident hazards.

[0067] Compared with the prior art, the present invention has the following significant advantages: improving the real-time performance of fault identification: through real-time data acquisition and automated analysis process, it is able to quickly identify the chain reaction of a fault after it occurs, thereby reducing the response time of fault handling; improving the accuracy of fault analysis: utilizing multi-dimensional correlation analysis and a complete causal rule base to accurately identify the cause and propagation path of the fault, thereby reducing the probability of misjudgment and missed judgment; enhancing the safety of power grid operation: by identifying the risk status and potential faults of equipment, preventing further expansion of accidents, and ensuring the stable operation of the power grid; improving data utilization efficiency: deeply mining the real-time data value of WAMS and other systems, and improving the efficiency of data processing and analysis; assisting decision support: providing detailed chain failure analysis reports, providing scientific decision-making basis for power grid dispatching and operation personnel, and optimizing the fault handling strategy of the power grid.

[0068] This specification includes many details of specific embodiments, but it should be understood that these details do not limit any inventive concepts or what is claimed in the specification, but rather describe features of some exemplary embodiments.

[0069] Features described in the specification in the context of each exemplary embodiment may be implemented as a combination in one or more single exemplary embodiments. Conversely, various features described in the specification in the context of a single exemplary embodiment may be implemented as multiple exemplary embodiments individually or in appropriate sub-combinations.

[0070] Furthermore, features may be operated in a particular combination and may initially be described as claimed in the combination, but in certain cases one or more features may be excluded from the claimed combination and the claimed combination may be changed into a sub-combination or a modification of the sub-combination.

[0071] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that the operations be performed in a particular order or sequentially to achieve desirable results, nor should it be understood as requiring that all operations be performed.

[0072] While the inventive concept has been described in conjunction with what are presently considered to be practical exemplary embodiments, it should be understood that the inventive concept is not limited to such exemplary embodiments. Rather, the inventive concept is intended to cover various modifications and equivalent arrangements within the spirit and scope of the appended claims.

Claims

1. A method for identifying chain failures in a power grid, characterized in that: include: Obtain alarm information in the power grid and WAMS synchronous measurement data; Obtain the rule base and power grid topology model; Based on the power grid topology model and the WAMS synchronous measurement data, the rule base is used to perform correlation analysis on the fault events between the multiple fault events that generate the alarm information, The correlation analysis includes at least one of a time type, a space type and an electrical type, and the rule base records the causal relationship between the fault events.

2. The power grid fault chain identification method according to claim 1, characterized in that: Further including: In the case of the time type, a specific fault event among the multiple fault events that generate the alarm information is taken as a starting point, and fault events before and after the specific fault event are searched according to a predetermined time window to form a fault event set.

3. The power grid fault chain identification method according to claim 1, characterized in that: Further including: In the case of the space type, a node where a specific fault event among multiple fault events generating the alarm information occurs is taken as a fault node; The distance and impact degree between the faulty node and other nodes in the power grid topology model are analyzed.

4. The power grid fault chain identification method according to claim 1, characterized in that: Further including: In the case of the electrical type, the WAMS synchronous measurement data is used to calculate the changes in power flow before and after the fault and identify electrical correlations.

5. The method for identifying power grid fault cascading failures according to any one of claims 1 to 4, characterized in that: Further including: The results of the correlation analysis are visualized.

6. A power grid fault chain identification system, characterized in that: include: Data interface module: used to obtain alarm information in the power grid and WAMS synchronous measurement data; Model and rule management module: used to obtain the rule base and power grid topology model; Fault analysis module: based on the grid topology model and the WAMS synchronous measurement data, using the rule base to perform correlation analysis on the fault events between the multiple fault events that generate the alarm information, The correlation analysis includes at least one of a time type, a space type and an electrical type, and the rule base records the causal relationship between the fault events.

7. The power grid fault chain identification system according to claim 6, characterized in that: In the case of the time type, the fault analysis module takes a specific fault event among the multiple fault events that generate the alarm information as a starting point, searches for fault events before and after the specific fault event according to a predetermined time window, and forms a fault event set.

8. The power grid fault chain identification system according to claim 6, characterized in that: In the case of the spatial type, the fault analysis module takes the node where a specific fault event among the multiple fault events that generate the alarm information occurs as the fault node, and analyzes the distance and impact degree between the fault node and other nodes in the power grid topology model.

9. The power grid fault chain identification system according to claim 6, characterized in that: In the case of the electrical type, the fault analysis module uses the WAMS synchronous measurement data to calculate the changes in the power flow before and after the fault.

10. The power grid fault chain identification system according to any one of claims 6 to 9, characterized in that: The system further includes a visualization module, which visualizes the result of the correlation analysis.

Images