Voltage abnormal fault tracing evaluation method and system

Abstract

The application is suitable for the field of electric power automation technology, and provides a voltage abnormal fault tracing evaluation method and system.The method comprises the following steps: acquiring power grid electric parameter time sequence data and electric power equipment physical state data, and integrating the data into a multi-dimensional data pool; combining the actual line topology of the power grid and the equipment association logic to construct a fault propagation dynamic model to simulate a cascading reaction; analyzing the abnormal time of the electric parameter identification, and screening a candidate starting point set; calculating the physical state matching degree to sort the preliminary fault starting point; and determining the final starting point through forward fitting verification and reverse exclusion verification.The application realizes accurate and efficient tracing of the fault starting point, guarantees the safe operation of the power grid, and is suitable for the field of electric power system fault diagnosis.

Description

Technical Field

[0001] This invention belongs to the field of power automation technology, and in particular relates to a method and system for tracing and evaluating the source of voltage anomalies. Background Technology

[0002] As power systems develop towards intelligence and large-scale operation, the coverage of power grids continues to expand, and the number and types of power equipment increase significantly. The cascading effects of voltage anomaly faults on power supply reliability are becoming increasingly prominent, making rapid and accurate location of the initiation point of voltage anomaly faults a core requirement for ensuring the safe operation of the power grid. Currently, power grid monitoring technology has achieved real-time acquisition of electrical parameters and the physical state of power equipment. However, existing methods for tracing the source of voltage anomalies have not fully leveraged the synergistic value of multi-dimensional data, still relying on traditional experience-based judgments or single-data-driven analysis models, which are difficult to adapt to the dynamic and interconnected nature of fault propagation in complex power grids.

[0003] Existing technical solutions mainly collect time-series data of power grid electrical parameters, combine them with simplified power grid line topology, identify the start time of voltage anomalies at each monitoring point, and select the equipment corresponding to the monitoring point with the earliest anomaly as the candidate fault starting point. Some solutions introduce static fault simulation models to simulate fault propagation paths based on electrical parameter data, but they do not integrate physical state data of power equipment, and the models cannot dynamically determine the failure probability of equipment under electrical stress. Ultimately, most of them determine the fault starting point through single positive verification or manual experience correction. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for tracing and evaluating voltage anomaly faults, aiming to solve the technical problems existing in the prior art as identified in the background art.

[0005] This invention is implemented as follows: a voltage anomaly fault tracing and assessment method, the method comprising: Acquire time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrate them into a multi-dimensional data pool; Based on a multidimensional data pool, and combined with the actual line topology of the power grid and the correlation logic of power equipment, a dynamic model of fault propagation is constructed to dynamically simulate the chain reaction caused by voltage anomaly faults. The dynamic model of fault propagation is used to characterize the impact of each data in the multidimensional data on fault propagation. Analyze the time series data of the electrical parameters, identify the start time of voltage anomalies at each monitoring point, and screen out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate start point set; Based on the physical status data of the power equipment, the clue matching degree of each power equipment in the candidate starting point set is calculated. Based on the matching degree calculation results, all candidate starting points in the entire candidate starting point set are prioritized and the preliminary fault starting point is determined based on the priority ranking results. Based on the set of candidate fault starting points, combined with a multi-dimensional data pool and a fault propagation dynamic model, the final fault starting point is determined through forward matching verification and reverse elimination verification, thus completing the voltage anomaly fault tracing assessment.

[0006] As a further embodiment of the present invention, the acquisition of time-series data of electrical parameters and physical state data of power equipment related to voltage anomalies in the power grid specifically includes: Collect voltage and current waveform sequences at various nodes of the power grid to form the time-series data of the electrical parameters; The readings of temperature, vibration, and partial discharge signals of the power equipment are collected to form the physical state data of the power equipment; Through the data fusion processing layer, the electrical parameter time-series data and the power equipment physical status data are aligned, cleaned, and associated and stored under a unified time axis to form the multi-dimensional data pool.

[0007] As a further aspect of the present invention, the construction of a fault propagation dynamic model to dynamically simulate the chain reaction caused by voltage anomaly faults specifically includes: Parse the SCD model file and wiring diagram of the power grid to obtain the actual line topology and equipment association logic containing the connection relationships of power equipment; Based on the actual line topology and equipment association logic, a fault propagation dynamic model is established, and the electrical parameter time series data in the multi-dimensional data pool is used as the basic input to simulate the propagation process of voltage abnormality on the actual line topology. The system calls upon the physical status data of the equipment in the multidimensional data pool in real time to determine the failure probability and operating limits of the power equipment under the current electrical stress and dynamically determine the fault propagation path.

[0008] As a further aspect of the present invention, the formation of the candidate starting point set specifically includes: Transient analysis is performed on the time series data of the electrical parameters to identify the voltage mutation points of each voltage monitoring point, and the time corresponding to the voltage mutation point is recorded as the start time of the voltage anomaly at that voltage monitoring point. A global comparison of the start times of voltage anomalies at all voltage monitoring points is performed, and all monitoring points that detect the start time of voltage anomalies within a 5-millisecond time window after the zero point of the time are selected. Based on the actual line topology and equipment association logic, the direct upstream power equipment corresponding to the selected monitoring points within the 5-millisecond time window is extracted to form a candidate starting point set.

[0009] As a further aspect of the present invention, determining the initial fault starting point specifically includes: For each power device in the candidate starting point set, the degree of match between the physical state data of the power device within a preset time window before the fault occurs and the standard fault feature data of the corresponding type in the preset fault database is calculated to obtain the physical state matching degree. Based on the physical state matching degree, all candidate starting points in the candidate starting point set are sorted by priority from high to low; The candidate starting point ranked first in the priority sorting results is selected as the initial fault starting point.

[0010] As a further aspect of the present invention, the determination of the final fault origin through forward matching verification and reverse elimination verification specifically includes: The electrical parameter time series data and power equipment physical state data corresponding to the initial fault initiation point are input into the fault propagation dynamic model in the forward direction. The simulated output cascading point status and fault propagation path are compared with the actual cascading point status and the actual fault propagation path deduced from the actual line topology and equipment association logic to verify the degree of consistency. In reverse, the candidate starting points other than the initial fault starting point in the candidate starting point set are sequentially input into the fault propagation dynamic model for fault simulation. The consistency of the fault propagation path obtained from each fault simulation with the actual fault propagation path is checked, and candidate starting points that have fundamental contradictions with the simulated path and the actual path are eliminated. The initial fault origin is determined as the final fault origin if and only if the simulation results of the initial fault origin pass the consistency verification and all other candidate origins are excluded.

[0011] Another object of the present invention is to provide a voltage anomaly fault tracing and assessment system, the system comprising: The data acquisition and integration module is used to acquire time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrate them into a multi-dimensional data pool. The fault propagation dynamic simulation module is used to construct a fault propagation dynamic model based on a multi-dimensional data pool and combined with the actual line topology of the power grid and the association logic of power equipment. It dynamically simulates the chain reaction caused by voltage abnormality faults. The fault propagation dynamic model is used to characterize the impact of each data in the multi-dimensional data on fault propagation. The candidate starting point identification module is used to analyze the electrical parameter time series data, identify the start time of voltage anomalies at each monitoring point, and filter out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate starting point set. The preliminary fault origin determination module is used to calculate the clue matching degree of each power device in the candidate origin set by combining the physical state data of the power equipment, sort all candidate origins in the entire candidate origin set according to the matching degree calculation result, and determine the preliminary fault origin by combining the priority sorting result. The starting point verification module is used to determine the final fault starting point based on the set of candidate fault starting points, combined with a multi-dimensional data pool and a fault propagation dynamic model, through forward matching verification and reverse elimination verification, and to complete the voltage anomaly fault tracing assessment.

[0012] The beneficial effects of this invention are: This invention constructs a complete traceability system. A multi-dimensional data pool integrates time-series electrical parameter data and physical state data of power equipment to achieve a two-way correlation between electrical anomalies and equipment health, avoiding missed fault sources due to single data. The fault propagation dynamic model combines the actual power grid line topology with the real-time physical state of equipment to dynamically calculate the probability of equipment failure, accurately simulate the fault chain propagation path, and fit the dynamic evolution process of faults in complex power grids. The candidate starting point screening narrows the range by global comparison at anomaly times and topology back-inference of upstream equipment, and locks high-suspect equipment by combining physical state matching degree ranking, which greatly improves the accuracy of the initial starting point. The dual mechanism of forward matching verification and reverse elimination verification ensures that the final fault starting point can not only reproduce the full picture of the actual fault, but also has no other candidate equipment that can be replaced, significantly improving the uniqueness and reliability of the result.

[0013] The overall solution does not rely on human experience, enabling efficient and accurate location of the starting point of voltage anomalies, reducing the time and resource waste caused by ineffective repairs, quickly interrupting the chain reaction of faults, ensuring the safe and stable operation of the power grid, and improving power supply reliability. Attached Figure Description

[0014] Figure 1 A flowchart of the voltage anomaly fault tracing and assessment method provided in the embodiments of the present invention; Figure 2 A flowchart for obtaining time-series data of electrical parameters and physical state data of power equipment related to voltage anomalies in a power grid, provided in an embodiment of the present invention; Figure 3 A flowchart for dynamically simulating the chain reaction caused by an abnormal voltage fault, provided in an embodiment of the present invention; Figure 4 A flowchart for forming a candidate starting point set provided in an embodiment of the present invention; Figure 5A flowchart for determining the initial fault starting point provided in an embodiment of the present invention; Figure 6 A flowchart for determining the final fault starting point provided in an embodiment of the present invention; Figure 7 This is a structural block diagram of the voltage anomaly fault tracing and assessment system provided in an embodiment of the present invention. Detailed Implementation

[0015] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0016] Figure 1 The flowchart of the voltage anomaly fault tracing and assessment method provided in the embodiments of the present invention is as follows: Figure 1 As shown, the method includes: S100 acquires time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrates them into a multi-dimensional data pool. Voltage waveforms are the core basis for directly reflecting whether the voltage deviates from the rated value. They can capture dynamic processes such as minute voltage drops before a fault, sudden drops or rises at the moment of the fault, and fluctuations and recovery after the fault. Current waveforms can help determine the cause of voltage anomalies. In the physical condition data acquisition stage of power equipment, temperature data can reflect whether the equipment has increased resistance due to aging or poor contact, which in turn causes local voltage drops. Vibration data can reflect the stability of the equipment's mechanical structure. Mechanical faults may impair the equipment's operating precision and indirectly cause voltage output fluctuations. Partial discharge signals are a key precursor to the decline in the equipment's insulation performance. Insulation damage can easily lead to faults such as leakage and short circuits that directly cause voltage anomalies. Essentially, this supplements the lack of electrical parameters from the perspective of the equipment's physical health, avoiding the problem of relying solely on electrical data while ignoring the fault causes inherent in the equipment itself.

[0017] S200, based on a multi-dimensional data pool, combines the actual line topology of the power grid with the association logic of power equipment to construct a dynamic model of fault propagation, and dynamically simulates the chain reaction caused by voltage abnormality faults. The dynamic model of fault propagation is used to characterize the impact of each data in the multi-dimensional data on fault propagation. The SCD model file contains the configuration parameters, communication addresses, and electrical connection relationships of all primary equipment in the power grid, while the wiring diagram visually presents the physical wiring methods between the equipment. The combination of the two can accurately extract the actual line topology and equipment association logic.

[0018] Since the propagation of voltage anomalies must rely on the actual electrical connections of the power grid, simulations that are detached from the real topology may result in unreasonable situations where the fault is transmitted to unconnected devices. Only when based on the real topology can the subsequent simulated propagation path correspond to which monitoring points in the actual fault showed anomalies in sequence, thus providing a reliable inference framework for tracing the source.

[0019] When constructing a dynamic model of fault propagation based on this, the time-series electrical parameters from the multi-dimensional data pool must be used as the core input. The time-series electrical parameters are equivalent to the initial disturbance signal of the fault. By inputting this signal, the model can simulate the electrical propagation law of voltage anomalies in the topology network.

[0020] This combination of electrical propagation and equipment status assessment allows the model to track the entire chain of faults, from electrical anomalies to equipment damage and then to fault escalation.

[0021] S300, Analyze the time series data of the electrical parameters, identify the start time of voltage anomalies at each monitoring point, and screen out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate start point set; The initial stage of grid voltage anomalies often manifests as a transient process, where the voltage changes abruptly within a very short time. Steady-state data, however, can only reflect the stable state after a fault and cannot capture this instantaneous change. Transient analysis can accurately identify the voltage abrupt change point at each voltage monitoring point. For example, if the voltage at a monitoring point drops sharply from the rated 10kV to 6kV within 0.5 milliseconds, this abrupt change point is the starting moment of the voltage anomaly at that monitoring point. Recording this moment is the core basis for subsequent judgment of "which monitoring point experienced an anomaly first," because the monitoring point or upstream equipment corresponding to the true fault starting point will inevitably be the first to experience an anomaly. Transient analysis aims to capture this time signal and avoid missing initial fault clues by relying on steady-state data.

[0022] Subsequently, the onset times of anomalies at all monitoring points are globally compared, and monitoring points within a 5-millisecond time window are selected. Since electromagnetic waves propagate at near the speed of light in power lines, a 5-millisecond time window can cover a line propagation distance of tens of kilometers. This ensures that all monitoring points directly related to the initial fault are included, while effectively excluding secondary anomalies caused by subsequent protection device actions or emergency load switching. After the initial fault occurs, the protection device may activate and disconnect the faulty line after 10 milliseconds. Voltage anomalies at the end-of-line monitoring points at this time constitute secondary anomalies. Without time window filtering, monitoring points unrelated to the initial fault would be included in the analysis, leading to an uncontrolled expansion of the candidate equipment range. Limiting the window to 5 milliseconds essentially filters monitoring points directly related to the physical propagation of the initial fault through time correlation, ensuring a direct causal relationship between the subsequently identified equipment and the initial fault.

[0023] Based on the actual line topology and equipment association logic, the direct upstream power equipment of the monitoring points after screening is extracted. The propagation of voltage anomalies in the power grid follows the physical law of upstream equipment affecting downstream equipment. That is, the anomalies of downstream monitoring points are often caused by the failure of upstream equipment, rather than the other way around.

[0024] S400, combine the physical status data of the power equipment to calculate the clue matching degree for each power equipment in the candidate starting point set, sort all candidate starting points in the entire candidate starting point set according to the matching degree calculation result, and determine the preliminary fault starting point based on the priority sorting result. When calculating the physical state matching degree for each power device in the candidate starting point set, the physical state data of the device within a preset time window before the fault occurs is selected. The device state data after the fault occurs may be the result of voltage anomalies, rather than the cause of the fault. Only the data before the fault can reflect whether the device already has potential faults. For example, if the winding temperature, vibration value, and partial discharge signal in the 10 minutes before the fault show a deterioration trend similar to historical faults, it indicates that the device itself is already in an unhealthy state and is more likely to be the source of voltage anomalies. Conversely, if the state data before the fault remains stable and only changes after the voltage anomaly occurs, the device is more likely to passively receive abnormal signals rather than be the initial fault point.

[0025] The preset fault database provides a standard reference for fault precursors. This database is built based on a large number of historical fault cases of similar equipment and includes typical state characteristics corresponding to different fault types.

[0026] Candidate starting points are prioritized based on their physical state matching degree from high to low, with the device whose state and fault characteristics match the highest being placed first. Although all devices in the candidate starting point set meet the conditions of being the earliest and upstream, their own health states differ, and the ranking can intuitively distinguish between high-suspect and low-suspect devices.

[0027] Finally, the candidate starting point ranked first is selected as the initial fault starting point. If the initial starting point is not locked by matching degree ranking, subsequent verification will require simulation of all candidate devices one by one, which will greatly increase the workload. Focusing on the device with the highest matching degree can reduce the resource consumption of the verification process and ensure that the verification direction is focused on the most likely source, laying an efficient and accurate foundation for finally determining the fault starting point.

[0028] The S500, based on a set of candidate fault origins, combined with a multi-dimensional data pool and a fault propagation dynamic model, determines the final fault origin through forward matching verification and reverse elimination verification, thus completing the voltage anomaly fault tracing assessment.

[0029] The core logic of positive consistency verification is to use the initial starting point as the source to verify whether it can reproduce the full picture of the actual fault. Although the initial fault starting point is obtained through physical state matching degree sorting, it is still necessary to confirm whether it has the ability to trigger all subsequent cascading anomalies. Specifically, the time-series electrical parameter data and physical state data of the power equipment corresponding to the initial starting point are input into the fault propagation dynamic model. The model will simulate the fault propagation path and cascading point status from this starting point based on the actual power grid topology and equipment association logic. Then, these simulation results are compared with the actual situation: the actual situation needs to be obtained from two aspects: first, the status of the cascading points monitored on site; and second, the actual fault propagation path deduced from the actual line topology and equipment association logic.

[0030] Reverse elimination verification aims to eliminate the possibility of other candidate starting points and ensure the uniqueness of the final starting point. In the candidate starting point set, although other devices besides the initial starting point have a low matching degree, there is still a possibility that the matching degree calculation may be inaccurate due to data deviation, requiring further elimination through model simulation. Specifically, these non-initial starting point candidate devices are input one by one into the fault propagation dynamic model. Each simulation generates a corresponding fault propagation path, which is then compared with the reverse-engineered actual fault propagation path for consistency verification. The consistency verification here focuses on whether there is a fundamental contradiction: if the simulated path of a candidate starting point completely deviates from the actual path, or if the affected cascading points do not overlap with the actual ones, it indicates that the candidate starting point cannot cause an actual fault and can be directly eliminated; if there are slight differences between the simulated path and the actual one, further judgment is needed based on its physical state matching degree, but the core logic is that if there is an inexplicable fundamental contradiction, the candidate is eliminated.

[0031] Power grid faults directly impact power supply safety. If multiple candidate starting points exist that have not been ruled out, or if the initial starting point fails positive verification, it may lead to incorrect maintenance direction, causing secondary power outages or escalating the fault. Only when the initial starting point can reproduce the actual fault and there are no other candidate starting points that can be reproduced can the absolute accuracy of the final starting point be ensured.

[0032] like Figure 2 As shown, the acquisition of time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid specifically includes: S110, Collect voltage and current waveform sequences at each node of the power grid to form the time-series data of the electrical parameters; S120: Collect readings of temperature, vibration, and partial discharge signals of the power equipment to form physical state data of the power equipment; S130, through the data fusion processing layer, the electrical parameter time series data and the power equipment physical status data are aligned, cleaned and associated and stored under a unified time axis to form the multi-dimensional data pool.

[0033] like Figure 3 As shown, the construction of the fault propagation dynamic model to dynamically simulate the chain reaction caused by voltage anomaly faults specifically includes: S210, parses the SCD model file and wiring diagram of the power grid to obtain the actual line topology and equipment association logic containing the connection relationships of power equipment; The actual line topology refers to the physical electrical connection diagram between all primary equipment in the power grid, such as circuit breakers, disconnectors, transformers, busbars, and lines. Equipment association logic defines the functional dependencies between equipment; this step defines the possible propagation paths and ranges of faults (voltage anomalies). Without this topology logic, it is impossible to determine which direction a fault should propagate, nor can isolated monitoring points be linked into an organic whole.

[0034] S220, using the actual line topology and equipment association logic, establish a fault propagation dynamic model, and use the electrical parameter time series data in the multi-dimensional data pool as the basic input to simulate the propagation process of voltage abnormality on the actual line topology; S230, Real-time access to the physical status data of the equipment in the multi-dimensional data pool is used to determine the failure probability and operating limit of the power equipment under the current electrical stress, and to dynamically determine the fault propagation path.

[0035] Specifically: 1. Use the abnormal data of the preliminary fault starting point or candidate starting point as the initial disturbance.

[0036] 2. Based on the current power grid status, calculate the power flow distribution in the network after the fault occurs, and obtain the current, voltage and apparent power flowing through each device.

[0037] 3. Equipment Status Assessment: This involves retrieving the physical status data of the power equipment (transformer winding temperature). ), calculate the real-time failure probability of the device .

[0038] ; in: This represents the basic failure rate of the equipment under rated operating conditions. This represents the actual current flowing through the device under the current fault scenario. This is the rated current of the equipment; The maximum permissible operating temperature designed for this device. For shape parameters, constants, used to describe the degree of influence of current and temperature on aging failure; This is the simulated step size.

[0039] 4. Path determination: If a certain device If the operating limit is exceeded, the device is identified as the next fault point, and the fault propagates along the topology logic to the device, updating the power grid topology.

[0040] 5. Iteration: Repeat steps 2-4 until the fault is completely isolated or the system returns to stability, thereby dynamically generating a complete fault propagation path.

[0041] like Figure 4 As shown, forming the candidate starting point set specifically includes: S310, Perform transient analysis on the electrical parameter time series data, identify the voltage mutation point of each voltage monitoring point, and record the time corresponding to the voltage mutation point as the start time of the voltage anomaly of that voltage monitoring point; S320: Perform a global comparison of the start times of voltage anomalies at all voltage monitoring points, and filter out all monitoring points that detect the start time of voltage anomalies within a 5-millisecond time window after the zero point of the time. The electromagnetic transient process of power grid faults is extremely rapid. The true fault initiation point is always the earliest point at which voltage anomalies begin. Limiting the time window to within 5 milliseconds after the first anomaly point (time zero) is based on engineering experience regarding the propagation speed of electromagnetic waves in the power grid and the operating time of protection devices. This ensures that all monitoring points directly related to the initial fault are captured, while excluding voltage anomalies that occur after delayed tripping of line protection. This significantly narrows down the range of candidate initiation points, improving the efficiency and accuracy of source tracing.

[0042] S330, based on the actual line topology and equipment association logic, extract the direct upstream power equipment corresponding to the monitoring points selected within the 5-millisecond time window to form a candidate starting point set.

[0043] like Figure 5 As shown, determining the initial fault starting point specifically includes: S410, For each power device in the candidate starting point set, calculate the degree of match between the physical state data of the power device within a preset time window before the fault occurs and the standard fault feature data of the corresponding type in the preset fault database, and obtain the physical state matching degree. ; in: The closer the value is to 1, the higher the degree of matching in the calculated physical state. For candidate power equipment, within a preset time window before the fault occurs, the first The actual measured values ​​of each state parameter; For equipment of the same type as the candidate power equipment, retrieved from the preset fault database, before a known fault occurs, the first... Typical fault precursor characteristic values ​​of each state parameter; For the first The weighting coefficients of each state parameter.

[0044] S420, sort all candidate starting points in the candidate starting point set according to the physical state matching degree from high to low; S430, select the candidate starting point that ranks first in the priority sorting results as the initial fault starting point.

[0045] like Figure 6 As shown, the process of determining the final fault origin through forward matching verification and reverse elimination verification specifically includes: S510, the electrical parameter time series data and power equipment physical state data corresponding to the initial fault starting point are input into the fault propagation dynamic model in the forward direction. The simulated output of the cascading point status and fault propagation path is compared with the actual cascading point status and the actual fault propagation path deduced from the actual line topology and equipment association logic to verify the degree of consistency. S520, In reverse, the other candidate starting points in the candidate starting point set, except for the initial fault starting point, are sequentially input into the fault propagation dynamic model for fault simulation. The consistency of the fault propagation path obtained from each fault simulation with the reverse-inferred actual fault propagation path is checked, and candidate starting points that have a fundamental contradiction with the actual path are eliminated. S530, the initial fault starting point is determined as the final fault starting point if and only if the simulation results of the initial fault starting point pass the consistency verification and all other candidate starting points are excluded.

[0046] The purpose of positive verification is to test whether the initially suspected object (the initial point of failure) is truly capable of, and logically, creating the entire observed failure scene. Through verification, supporting evidence is sought for the initial conclusions.

[0047] The purpose of reverse verification is to eliminate all other reasonable possibilities. By determining whether other equipment malfunctions are possible, and by systematically examining and eliminating other candidate starting points, it ensures that no other causes exist, thereby enhancing the uniqueness and reliability of the final conclusion.

[0048] Figure 7 This is a structural block diagram of the voltage anomaly fault tracing and assessment system provided in an embodiment of the present invention, as shown below. Figure 7 As shown, the system includes: The data acquisition and integration module 100 is used to acquire time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrate them into a multi-dimensional data pool. The fault propagation dynamic simulation module 200 is used to construct a fault propagation dynamic model based on a multi-dimensional data pool and combined with the actual line topology of the power grid and the association logic of power equipment. It dynamically simulates the chain reaction caused by voltage abnormality faults. The fault propagation dynamic model is used to characterize the impact of each data in the multi-dimensional data on fault propagation. The candidate starting point identification module 300 is used to analyze the electrical parameter time series data, identify the start time of voltage anomalies at each monitoring point, and filter out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate starting point set. The preliminary fault origin determination module 400 is used to calculate the clue matching degree of each power device in the candidate origin set by combining the physical state data of the power equipment, sort all candidate origins in the entire candidate origin set according to the matching degree calculation result, and determine the preliminary fault origin by combining the priority sorting result. The starting point verification module 500 is used to determine the final fault starting point based on the set of candidate fault starting points, combined with a multi-dimensional data pool and a fault propagation dynamic model, through forward matching verification and reverse elimination verification, and to complete the voltage anomaly fault tracing assessment.

[0049] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0050] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0051] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for tracing and assessing the source of voltage anomalies, characterized in that, The method includes: Acquire time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrate them into a multi-dimensional data pool; Based on a multidimensional data pool, and combined with the actual line topology of the power grid and the correlation logic of power equipment, a dynamic model of fault propagation is constructed to dynamically simulate the chain reaction caused by voltage anomaly faults. The dynamic model of fault propagation is used to characterize the impact of each data in the multidimensional data on fault propagation. Analyze the time series data of the electrical parameters, identify the start time of voltage anomalies at each monitoring point, and screen out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate start point set; Based on the physical status data of the power equipment, the clue matching degree of each power equipment in the candidate starting point set is calculated. Based on the matching degree calculation results, all candidate starting points in the entire candidate starting point set are prioritized and the preliminary fault starting point is determined based on the priority ranking results. Based on the set of candidate fault starting points, combined with a multi-dimensional data pool and a fault propagation dynamic model, the final fault starting point is determined through forward matching verification and reverse elimination verification, thus completing the voltage anomaly fault tracing assessment.

2. The method according to claim 1, characterized in that, The acquisition of time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid specifically includes: Collect voltage and current waveform sequences at various nodes of the power grid to form the time-series data of the electrical parameters; The readings of temperature, vibration, and partial discharge signals of the power equipment are collected to form the physical state data of the power equipment; Through the data fusion processing layer, the electrical parameter time-series data and the power equipment physical status data are aligned, cleaned, and associated and stored under a unified time axis to form the multi-dimensional data pool.

3. The method according to claim 2, characterized in that, The construction of a dynamic fault propagation model to dynamically simulate the chain reaction triggered by an abnormal voltage fault specifically includes: Parse the SCD model file and wiring diagram of the power grid to obtain the actual line topology and equipment association logic containing the connection relationships of power equipment; Based on the actual line topology and equipment association logic, a fault propagation dynamic model is established, and the electrical parameter time series data in the multi-dimensional data pool is used as the basic input to simulate the propagation process of voltage abnormality on the actual line topology. The system calls upon the physical status data of the equipment in the multidimensional data pool in real time to determine the failure probability and operating limits of the power equipment under the current electrical stress and dynamically determine the fault propagation path.

4. The method according to claim 3, characterized in that, The formation of the candidate starting point set specifically includes: Transient analysis is performed on the time series data of the electrical parameters to identify the voltage mutation points of each voltage monitoring point, and the time corresponding to the voltage mutation point is recorded as the start time of the voltage anomaly at that voltage monitoring point. A global comparison of the start times of voltage anomalies at all voltage monitoring points is performed, and all monitoring points that detect the start time of voltage anomalies within a 5-millisecond time window after the zero point of the time are selected. Based on the actual line topology and equipment association logic, the direct upstream power equipment corresponding to the selected monitoring points within the 5-millisecond time window is extracted to form a candidate starting point set.

5. The method according to claim 4, characterized in that, The determination of the initial fault initiation point specifically includes: For each power device in the candidate starting point set, the degree of match between the physical state data of the power device within a preset time window before the fault occurs and the standard fault feature data of the corresponding type in the preset fault database is calculated to obtain the physical state matching degree. Based on the physical state matching degree, all candidate starting points in the candidate starting point set are sorted by priority from high to low; The candidate starting point ranked first in the priority sorting results is selected as the initial fault starting point.

6. The method according to claim 5, characterized in that, The process of determining the final fault origin through forward matching verification and reverse elimination verification specifically includes: The electrical parameter time series data and power equipment physical state data corresponding to the initial fault initiation point are input into the fault propagation dynamic model in the forward direction. The simulated output cascading point status and fault propagation path are compared with the actual cascading point status and the actual fault propagation path deduced from the actual line topology and equipment association logic to verify the degree of consistency. In reverse, the candidate starting points other than the initial fault starting point in the candidate starting point set are sequentially input into the fault propagation dynamic model for fault simulation. The consistency of the fault propagation path obtained from each fault simulation with the actual fault propagation path is checked, and candidate starting points that have fundamental contradictions with the simulated path and the actual path are eliminated. The initial fault origin is determined as the final fault origin if and only if the simulation results of the initial fault origin pass the consistency verification and all other candidate origins are excluded.

7. A voltage anomaly fault tracing and assessment system, characterized in that, The system includes: The data acquisition and integration module is used to acquire time-series data of electrical parameters and physical status data of power equipment related to voltage anomalies in the power grid, and integrate them into a multi-dimensional data pool. The fault propagation dynamic simulation module is used to construct a fault propagation dynamic model based on a multi-dimensional data pool and combined with the actual line topology of the power grid and the association logic of power equipment. It dynamically simulates the chain reaction caused by voltage abnormality faults. The fault propagation dynamic model is used to characterize the impact of each data in the multi-dimensional data on fault propagation. The candidate starting point identification module is used to analyze the electrical parameter time series data, identify the start time of voltage anomalies at each monitoring point, and filter out power equipment whose voltage anomalies start earlier than all interlocking points to form a candidate starting point set. The preliminary fault origin determination module is used to calculate the clue matching degree of each power device in the candidate origin set by combining the physical state data of the power equipment, sort all candidate origins in the entire candidate origin set according to the matching degree calculation result, and determine the preliminary fault origin by combining the priority sorting result. The starting point verification module is used to determine the final fault starting point based on the set of candidate fault starting points, combined with a multi-dimensional data pool and a fault propagation dynamic model, through forward matching verification and reverse elimination verification, and to complete the voltage anomaly fault tracing assessment.

Images