Intelligent diagnosis method and system for drop-out high-voltage fuse

Abstract

The present application relates to the technical field of intelligent diagnosis, in particular to a kind of intelligent diagnosis method and system of drop-out high-voltage fuse, comprising: based on the electrical data collected, the gear of transformer is identified, and the gear of transformer is obtained;The gear of transformer is verified, and the gear of transformer after verification is obtained;Based on the gear of transformer after verification and electrical data, fuse fault diagnosis feature is extracted;According to fault diagnosis feature and pre-set fault diagnosis model, fault discrimination is carried out, and the diagnosis result of fuse is obtained.The present application realizes the accurate identification of fuse state by real-time acquisition of electrical data of high-voltage side and low-voltage side of transformer, and fault feature extraction after working condition correction combined with the gear of transformer, improves the accuracy of fault discrimination.

Description

Technical Field

[0001] This invention relates to the field of intelligent diagnostic technology, specifically to an intelligent diagnostic method and system for drop-out high-voltage fuses. Background Technology

[0002] In high-voltage power distribution systems, drop-out high-voltage fuses are widely used as important protective components for line overload, short circuit, and fault isolation protection. However, existing fuse operation monitoring methods mainly rely on manual inspections or simple open / closed status detection, which suffers from problems such as delayed response, inaccurate diagnosis, and lack of early warning capabilities.

[0003] Traditional methods struggle to achieve real-time monitoring of fuse status and analysis of fault causes. For example, fuse tripping can be caused by a variety of factors, such as short circuits, overloads, aging, or mechanical defects. However, existing systems typically only record closing or tripping events, failing to identify specific fault types or provide early warnings of potential risks.

[0004] With the development of smart grid and Internet of Things (IoT) technologies, it has become possible to use electrical data for fuse condition monitoring and fault diagnosis. However, improving the accuracy of fault identification remains a challenge. Summary of the Invention

[0005] (a) Purpose of the invention

[0006] The purpose of this invention is to provide an intelligent diagnostic method and system for drop-out high-voltage fuses. By collecting electrical data from the high-voltage and low-voltage sides of the transformer in real time, and combining the transformer tap position for operating condition correction, fault feature extraction is performed, thereby achieving accurate identification of the fuse status and improving the accuracy of fault diagnosis.

[0007] (II) Technical Solution

[0008] To address the above problems, this invention provides an intelligent diagnostic method for drop-out high-voltage fuses, wherein the drop-out high-voltage fuse is connected to a three-phase energy meter of a transformer, and the method includes:

[0009] Based on the collected electrical data, the transformer tap position is identified to obtain the transformer tap position;

[0010] The transformer tap position is verified to obtain the verified transformer tap position;

[0011] Based on the verified transformer tap position and electrical data, fuse fault diagnosis features are extracted.

[0012] Based on the fault diagnosis characteristics and the preset fault diagnosis model, fault identification is performed to obtain the diagnostic results of the fuse.

[0013] In another aspect of the present invention, preferably, the step of identifying the transformer tap position based on the collected electrical data to obtain the transformer tap position includes:

[0014] Collect three-phase voltage and three-phase current data on the high-voltage side of the transformer, as well as three-phase voltage and three-phase current data on the low-voltage side collected by the transformer's three-phase energy meter;

[0015] The three-phase voltages on the high-voltage side and the three-phase voltages on the low-voltage side are filtered to calculate the fundamental effective values ​​of the high-voltage side and the low-voltage side.

[0016] The measured turns ratio of the transformer is calculated based on the fundamental effective values ​​of the high-voltage side and the low-voltage side.

[0017] Based on the measured turns ratio of the transformer, the transformer tap position is identified to obtain the transformer tap position.

[0018] In another aspect of the present invention, preferably, the step of identifying the transformer tap position based on the measured transformer ratio to obtain the transformer tap position includes:

[0019] Based on the transformer's rated turns ratio, tap range, and number of taps, the theoretical turns ratio corresponding to each tap is generated.

[0020] The measured turns ratio is compared with the theoretical turns ratio corresponding to each gear position to determine the gear position corresponding to the measured turns ratio, which is then used as the identified transformer gear position.

[0021] In another aspect of the present invention, preferably, the step of verifying the transformer tap position to obtain the verified transformer tap position includes:

[0022] Within a preset time window, obtain the measured ratios at multiple points in time;

[0023] Statistical analysis was performed on the measured ratios at the multiple time points to obtain stable values ​​of the measured ratios;

[0024] The stability value of the measured transformer ratio is checked against the theoretical transformer ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position.

[0025] In another aspect of the present invention, preferably, the step of verifying the consistency between the stable value of the measured transformer ratio and the theoretical transformer ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position includes:

[0026] When the deviation between the stable value of the measured transformer ratio and the corresponding theoretical transformer ratio is less than or equal to a preset deviation threshold, the identified transformer tap position is determined to be the verified transformer tap position.

[0027] When the deviation between the stable value of the measured gear ratio and the corresponding theoretical gear ratio is greater than a preset deviation threshold, gear identification is performed again until the deviation between the stable value of the measured gear ratio and the corresponding theoretical gear ratio is less than or equal to the preset deviation threshold.

[0028] In another aspect of the present invention, preferably, the step of extracting fuse fault diagnosis features based on the verified transformer tap position and electrical data includes:

[0029] Based on the verified transformer tap position, the electrical data is corrected for operating conditions to obtain the tap position corrected electrical data.

[0030] Based on the electrical data before correction and the electrical data after gear correction, the fault diagnosis features of the fuse are extracted.

[0031] In another aspect of the present invention, preferably, the step of performing operating condition correction on the electrical data based on the verified transformer tap position to obtain tap position corrected electrical data includes:

[0032] The collected low-voltage side three-phase voltage and low-voltage side three-phase current data are converted to the high-voltage side equivalent voltage and equivalent current according to the measured turns ratio.

[0033] The equivalent voltage and equivalent current are used as the electrical data after gear correction.

[0034] In another aspect of the present invention, preferably, the preset fault diagnosis model is based on an LSTM / GRU network, with fault diagnosis features as input and the health score, fault risk index, and determined fuse status as output.

[0035] In another aspect of the present invention, preferably, the step of determining the fault based on the fault diagnosis features and a preset fault diagnosis model to obtain the diagnostic result of the fuse includes:

[0036] The fault diagnosis features are input into the preset fault diagnosis model;

[0037] Based on the preset fault diagnosis model, the fault diagnosis features are predicted in time series and anomalies are detected, and the health score and fault risk index of the fuse are calculated.

[0038] Based on the health score and fault risk indicators, the current status of the fuse is determined. The status of the fuse includes normal closing, fault tripping, accidental tripping, or non-tripping fault.

[0039] The system outputs the health score, fault risk indicators, and the current status of the determined fuse to obtain the diagnostic results of the fuse.

[0040] In another aspect, preferably, a smart diagnostic system for drop-out high-voltage fuses is provided, wherein the drop-out high-voltage fuse is connected to a three-phase energy meter of a transformer, and the system comprises:

[0041] Identification module: Based on the collected electrical data, it identifies the transformer tap position and obtains the transformer tap position;

[0042] Verification module: Verifies the transformer tap position to obtain the verified transformer tap position;

[0043] Extraction module: Based on the verified transformer tap position and electrical data, extract the fault diagnosis features of the fuse;

[0044] Discrimination module: Based on the fault diagnosis features and the preset fault diagnosis model, it performs fault discrimination and obtains the diagnosis result of the fuse.

[0045] (III) Beneficial Effects

[0046] The above-described technical solution of the present invention has the following beneficial technical effects:

[0047] This invention achieves accurate identification of fuse status and extraction of fault characteristics by real-time acquisition of electrical data from the high-voltage and low-voltage sides of the transformer and combining this data with transformer tap position adjustments. Through analysis of current, temperature, and trend characteristics using a pre-set fault diagnosis model, the health status, fault risk indicators, and specific fault types of the fuse can be determined in real time, effectively distinguishing between normal closing, fault drop, accidental drop, and non-drop faults. Compared with traditional manual inspection or single-state monitoring methods, this method can identify potentially high-risk fuses in advance, providing an early warning function and significantly improving operational safety and reliability. Attached Figure Description

[0048] Figure 1 This is an overall flowchart of one embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and the accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0050] Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0051] In the description of this invention, it should be noted that the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0052] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] The invention will now be described in more detail with reference to the accompanying drawings. In the various drawings, the same elements are indicated by similar reference numerals. For clarity, the various parts in the drawings are not drawn to scale.

[0054] Example 1

[0055] A method for intelligent diagnosis of drop-out high-voltage fuses, wherein the drop-out high-voltage fuses are connected to a three-phase energy meter of a transformer, wherein the three-phase energy meter can collect multi-dimensional electrical quantity data during the operation of the transformer in real time, including three-phase voltage, three-phase current, active power and reactive power. Figure 1 An overall flowchart of one embodiment of the present invention is shown, as follows: Figure 1 As shown, the method includes:

[0056] Based on the collected electrical data, the transformer tap position is identified to obtain the transformer tap position. The specific content of the electrical data is not limited here; in this embodiment, the electrical data includes the three-phase voltage and current data on the high-voltage side of the transformer, as well as the three-phase voltage and current data on the low-voltage side collected by the transformer's three-phase energy meter. By simultaneously acquiring the electrical quantity information on both the high- and low-voltage sides of the transformer, the voltage transformation characteristics and load response features of the transformer under different tap positions can be comprehensively reflected. In this embodiment, the identification of the transformer tap position based on the collected electrical data includes:

[0057] The system collects three-phase voltage and current data on the high-voltage side of the transformer, as well as three-phase voltage and current data on the low-voltage side collected by the transformer's three-phase energy meter. The high-voltage side electrical data reflects the input voltage level and load current of the transformer's primary side, while the low-voltage side electrical data reflects the output voltage and load response of the transformer's secondary side. By synchronously collecting electrical quantities on both the high and low voltage sides, the accuracy and consistency of subsequent transformer ratio calculations and tap position identification are ensured.

[0058] The three-phase voltages on the high-voltage side and the low-voltage side are filtered to calculate the fundamental RMS values ​​for both sides. To address potential harmonic components, noise interference, and transient fluctuations in the acquired voltage signals, a digital filtering algorithm is used to extract the fundamental component. Based on this, the fundamental RMS values ​​of the voltages are calculated, yielding the fundamental RMS values ​​for both the high-voltage and low-voltage sides. Calculating the fundamental RMS values ​​avoids the influence of harmonics and transient disturbances on the transformer ratio calculation results, improving the stability of the measured transformer ratio.

[0059] The measured turns ratio of the transformer is calculated based on the fundamental RMS values ​​of the high-voltage and low-voltage sides. The ratio of the fundamental RMS value on the high-voltage side to that on the low-voltage side is calculated to obtain the measured turns ratio parameter reflecting the actual voltage transformation relationship of the transformer. In a three-phase system, the corresponding turns ratio can be calculated for each of the three phase voltages, and the average value or a representative value after consistency verification is taken as the measured turns ratio under the current operating condition of the transformer.

[0060] Based on the measured turns ratio of the transformer, the transformer tap position is identified to obtain the transformer tap position.

[0061] Furthermore, in this embodiment, the transformer tap position is identified and obtained based on the measured transformer ratio, including:

[0062] Based on the transformer's rated turns ratio, tap range, and number of taps, the theoretical turns ratio corresponding to each tap position is generated. The transformer's nameplate parameters and design parameters are pre-acquired or stored, including the transformer's rated primary voltage, rated secondary voltage, rated turns ratio, allowable tap voltage regulation range, and the total number of tap positions. According to the tap range and number of taps, the rated turns ratio is divided into equal or unequal intervals, and the theoretical turns ratio value corresponding to each tap position under ideal operating conditions is calculated, thus forming a "taper position - theoretical turns ratio" correspondence table.

[0063] The measured turns ratio is compared with the theoretical turns ratio corresponding to each gear position to determine the gear position corresponding to the measured turns ratio, which is then identified as the transformer gear position. The difference or relative deviation between the measured turns ratio and each theoretical turns ratio is calculated. When the deviation between the measured turns ratio and the theoretical turns ratio corresponding to a certain gear position is less than a preset threshold, it is determined that the measured turns ratio matches that gear position, and that gear position is determined as the current operating gear position of the transformer. The preset threshold can be set according to the transformer manufacturing error, measurement error, and operating fluctuation range to ensure the reliability of gear position identification. If the deviations between the theoretical turns ratio and the measured turns ratio of multiple gear positions all meet the threshold condition, the gear position with the smallest deviation can be further selected as the final identification result; or the candidate gear positions can be comprehensively judged by combining the changing trend of the measured turns ratio in adjacent time periods and the historical gear position status, thereby avoiding gear position misjudgment caused by instantaneous fluctuations.

[0064] The transformer tap position is verified to obtain the verified transformer tap position, including:

[0065] Within a preset time window, measured turns ratios at multiple points in time are acquired. The preset time window can be a continuous or discrete time interval, and its length can be set according to the transformer's operating characteristics and sampling period, for example, from tens of seconds to several minutes. Within this time window, measured turns ratios calculated from the high-voltage and low-voltage side voltages are continuously acquired according to a predetermined sampling period, thus forming a sequence of measured turns ratios reflecting the transformer's operating status over a period of time. By introducing a time window, the impact of instantaneous disturbances, load fluctuations, or measurement noise on the results of a single measured turns ratio can be reduced.

[0066] Statistical analysis is performed on the measured turns ratios at multiple time points to obtain stable values ​​for the measured turns ratios. Specifically, the measured turns ratio sequence is statistically processed, such as calculating the mean, median, or representative value after outlier removal, to characterize the stable operating turns ratio of the transformer within that time window. When the measured turns ratio fluctuates little and the trend is gentle within the time window, the statistical results can be used as stable values ​​for the measured turns ratio, thereby reflecting the true voltage transformation relationship of the transformer at the current tap position.

[0067] The stable value of the measured transformer ratio is compared with the theoretical transformer ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position, including:

[0068] When the deviation between the stable value of the measured transformer ratio and the corresponding theoretical transformer ratio is less than or equal to a preset deviation threshold, the identified transformer tap position is determined to be the verified transformer tap position; this indicates that the tap position identification result conforms to the actual operating state of the transformer, thereby determining that the identified transformer tap position is the verified transformer tap position.

[0069] When the deviation between the stable value of the measured turns ratio and the corresponding theoretical turns ratio exceeds a preset deviation threshold, the current gear position identification result is considered inconsistent with the actual operating state of the transformer, possibly due to transient disturbances, identification errors, or the gear position switching process. In this case, a new gear position identification process is triggered, i.e., the stable value of the measured turns ratio is re-compared with the theoretical turns ratio corresponding to each gear position, and new gear position candidates are determined according to the principle of minimum deviation or meeting the threshold condition; the above consistency verification process is repeated until the deviation between the stable value of the measured turns ratio and the corresponding theoretical turns ratio is less than or equal to the preset deviation threshold, thereby obtaining the final verified transformer gear position.

[0070] Based on the verified transformer tap position and electrical data, fuse fault diagnosis features are extracted, including:

[0071] Based on the verified transformer tap position, the electrical data is corrected for operating conditions to obtain tap-corrected electrical data. Since different transformer tap positions correspond to different voltage ratios and operating conditions, under the same load conditions, different tap positions will cause systematic changes in the high and low voltage sides and current. In this embodiment, based on the verified transformer tap position, the electrical data is corrected for operating conditions to obtain tap-corrected electrical data, including:

[0072] The collected low-voltage side three-phase voltage and low-voltage side three-phase current data are converted to the high-voltage side equivalent voltage and equivalent current according to the measured transformation ratio; the equivalent voltage and equivalent current are used as the electrical data after gear position correction.

[0073] Based on the electrical data before correction and the electrical data after gear correction, fault diagnosis features of the fuse are extracted. By simultaneously utilizing the original electrical data before correction and the electrical data after gear correction, both the absolute state information of the fuse under actual operating conditions and its relative change characteristics under a unified operating condition benchmark can be taken into account, thereby improving the comprehensiveness and reliability of fault feature extraction. Fault diagnosis features may include current features and voltage features. Current features include the effective value of current, peak value, and rate of change of current. Voltage features may include the effective value of voltage, peak value of voltage, rate of change of voltage, and voltage imbalance, etc. Furthermore, trend features, trend features of current and voltage, and differential consistency features may also be included. The differential consistency features are constructed based on the electrical data before correction and the electrical data after gear correction, and are used to characterize the degree of deviation between the electrical data before correction and the electrical data after gear correction in terms of amplitude and change trend.

[0074] Based on the aforementioned fault diagnosis features and a pre-defined fault diagnosis model, fault identification is performed to obtain the diagnostic results for the fuse. The pre-defined fault diagnosis model is based on an LSTM / GRU network, with fault diagnosis features as input and the fuse's health score, fault risk index, and determined fuse status as output. The pre-defined fault diagnosis model can effectively model the input time-series data, capturing the dynamic changes of the fuse over time.

[0075] In this embodiment, based on the fault diagnosis features and the preset fault diagnosis model, fault identification is performed to obtain the diagnostic result of the fuse, including:

[0076] The fault diagnosis features are input into the preset fault diagnosis model; the fault diagnosis features may include current features, voltage features, trend features and differential consistency features; wherein, the current features, voltage features and trend features each include two sets, before correction and after correction, and the differential consistency features are one set.

[0077] Based on the preset fault diagnosis model, the fault diagnosis features are time-series predicted and anomaly detected, and the health score and fault risk index of the fuse are calculated. Specifically, the current features, voltage features, trend features and differential consistency features are aligned and normalized according to a unified time step to construct a multi-dimensional time-series feature vector sequence. After time alignment and normalization, the various features within the same time step are combined to construct a multi-dimensional time-series feature vector, and a multi-dimensional time-series feature vector sequence is formed according to the time order.

[0078] The multidimensional time-series feature vector sequence is input into a fault diagnosis model constructed based on a Long Short-Term Memory (LSTM) network or a gated recurrent unit (GRU) network. The fault diagnosis model is then used to perform time-series modeling on the multidimensional time-series feature vector sequence to obtain the predicted feature vector for the corresponding time step. The evolutionary relationship of the time-series feature vector sequence is modeled to learn the dynamic patterns of various features of the fuse changing over time under normal operating conditions. By continuously inputting historical time-series features, the model can capture short-term fluctuation characteristics and long-term trends, and output the predicted feature vector corresponding to the current time step, which is used to characterize the expected state of each feature at that time step under normal operating conditions.

[0079] Based on the difference between the predicted feature vector and the actual feature vector at the corresponding time step, the reconstruction error or prediction error at each time step is calculated, and the reconstruction error or prediction error is used as the anomaly degree reflecting the degree of abnormality of the fault diagnosis features. By calculating the difference between the two in each feature dimension, error information reflecting the degree of deviation between the actual operating state and the normal operating state predicted by the model is obtained.

[0080] Statistical analysis is performed on the anomaly degree within a time window to calculate the mean, variance, or cumulative change of the anomaly degree. A health score for the fuse is generated based on the statistical results, where a lower anomaly degree results in a higher health score. By calculating the mean, variance, or cumulative change of the anomaly degree within a preset time window, the overall operational stability and the degree of anomaly persistence of the fuse over a period of time can be comprehensively reflected. The health score of the fuse, generated based on the above statistical results, is used to quantitatively characterize the current operational health level of the fuse.

[0081] Furthermore, based on the trend or growth rate of the anomaly degree within a continuous time window, a fault risk index for the fuse is calculated to characterize the probability level of the fuse failing within a preset time range in the future. When the anomaly degree continues to rise or the growth rate increases significantly within multiple time windows, it indicates that the fuse's operating state is evolving towards an unstable or faulty state. The fault risk index for the fuse, calculated based on the aforementioned trend or growth rate, characterizes the likelihood of the fuse failing within a preset time range in the future, thereby achieving early warning of potential fuse failures.

[0082] Pre-correction features accurately reflect the original electrical state changes of the fuse under actual operating conditions, fully preserving the impact of load fluctuations, environmental changes, and equipment aging on electrical parameters, which is beneficial for capturing sudden anomalies and obvious fault characteristics. Post-correction features, on the other hand, characterize the fuse's operating state under a unified operating condition benchmark, weakening the impact of external disturbances and measurement deviations, making it easier for the model to learn normal operating modes and improving the ability to identify subtle anomalies and progressive faults. By simultaneously introducing pre-correction and post-correction features, the correspondence between the two types of features can be learned. When an anomaly occurs in the equipment, this correspondence deviates, thereby reducing the risk of false alarms and missed alarms. Furthermore, differential consistency features directly characterize the degree of consistency change between the pre-correction and post-correction states, effectively identifying structural anomalies caused by internal structural changes or performance degradation, and showing higher sensitivity to early faults and non-dropout faults. Collaborative modeling enables fault diagnosis results to simultaneously reflect absolute anomalies, relative deviations, and consistency violations, thereby significantly improving the accuracy and stability of fuse fault diagnosis.

[0083] Based on the health score and fault risk indicators, the current status of the fuse is determined. The fuse status includes normal closing, fault tripping, accidental tripping, or non-tripping fault. "Normal closing" indicates that the fuse is in normal working condition; "fault tripping" indicates that the fuse tripped due to overload, short circuit, or other faults; "accidental tripping" indicates that the fuse tripped unexpectedly without a fault; and "non-tripping fault" indicates that although the fuse has an anomaly, it did not trip, potentially posing a safety risk. Status determination can be achieved through rule-based judgment combining health scores and fault risk indicator thresholds, or the model can directly output the status classification results.

[0084] The system outputs the health score, fault risk indicators, and the determined current status of the fuse to obtain the fuse's diagnostic results. The calculated health score, fault risk indicators, and determined current fuse status are then output together to form a comprehensive fuse diagnostic result. This diagnostic result can be used for power system operation monitoring and maintenance decisions, providing maintenance personnel with real-time fuse health status, potential risk warnings, and fault analysis basis, thereby achieving intelligent monitoring, preventative maintenance, and safe operation management of fuses.

[0085] This invention achieves accurate identification of fuse status and extraction of fault characteristics by real-time acquisition of electrical data from the high-voltage and low-voltage sides of the transformer and combining this data with transformer tap position adjustments. Through analysis of current, temperature, and trend characteristics using a pre-set fault diagnosis model, the health status, fault risk indicators, and specific fault types of the fuse can be determined in real time, effectively distinguishing between normal closing, fault drop, accidental drop, and non-drop faults. Compared with traditional manual inspection or single-state monitoring methods, this method can identify potentially high-risk fuses in advance, providing an early warning function and significantly improving operational safety and reliability.

[0086] Example 2

[0087] A smart diagnostic system for drop-out high-voltage fuses, wherein the drop-out high-voltage fuse is connected to a three-phase energy meter of a transformer, the system comprising:

[0088] Identification module: Based on the collected electrical data, it identifies the transformer tap position and obtains the transformer tap position;

[0089] Verification module: Verifies the transformer tap position to obtain the verified transformer tap position;

[0090] Extraction module: Based on the verified transformer tap position and electrical data, extract the fault diagnosis features of the fuse;

[0091] Discrimination module: Based on the fault diagnosis features and the preset fault diagnosis model, it performs fault discrimination and obtains the diagnosis result of the fuse.

[0092] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of the invention should be included within the protection scope of the invention. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

[0093] The present invention has been described above with reference to embodiments thereof. However, these embodiments are merely illustrative and not intended to limit the scope of the invention. The scope of the invention is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of the invention, and all such substitutions and modifications should fall within the scope of the invention.

[0094] Although embodiments of the present invention have been described in detail, it should be understood that various changes, substitutions, and modifications can be made to the embodiments of the present invention without departing from the spirit and scope of the invention.

[0095] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for intelligent diagnosis of drop-out high-voltage fuses, characterized in that, The method for connecting the drop-out high-voltage fuse to the three-phase energy meter of the transformer includes: Based on the collected electrical data, the transformer tap position is identified to obtain the transformer tap position; The transformer tap position is verified to obtain the verified transformer tap position; Based on the verified transformer tap position and electrical data, fuse fault diagnosis features are extracted. Based on the fault diagnosis characteristics and the preset fault diagnosis model, fault identification is performed to obtain the diagnostic results of the fuse; The step of identifying the transformer tap position based on the collected electrical data to obtain the transformer tap position includes: Collect three-phase voltage and three-phase current data on the high-voltage side of the transformer, as well as three-phase voltage and three-phase current data on the low-voltage side collected by the transformer's three-phase energy meter; The three-phase voltages on the high-voltage side and the three-phase voltages on the low-voltage side are filtered to calculate the fundamental effective values ​​of the high-voltage side and the low-voltage side. The measured turns ratio of the transformer is calculated based on the fundamental effective values ​​of the high-voltage side and the low-voltage side. Based on the measured turns ratio of the transformer, the transformer tap position is identified to obtain the transformer tap position; The step of verifying the transformer tap position to obtain the verified transformer tap position includes: Within a preset time window, obtain the measured ratios at multiple points in time; Statistical analysis was performed on the measured ratios at the multiple time points to obtain stable values ​​of the measured ratios; The stability value of the measured turns ratio is checked against the theoretical turns ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position. Based on the verified transformer tap position and electrical data, the following features are extracted for fuse fault diagnosis: Based on the verified transformer tap position, the electrical data is corrected for operating conditions to obtain tap-corrected electrical data, including: The collected low-voltage side three-phase voltage and low-voltage side three-phase current data are converted to the high-voltage side equivalent voltage and equivalent current according to the measured turns ratio. The equivalent voltage and equivalent current are used as the electrical data after gear position correction; Based on the electrical data before correction and the electrical data after gear correction, the fault diagnosis features of the fuse are extracted.

2. The intelligent diagnostic method for drop-out high-voltage fuses according to claim 1, characterized in that, The step of identifying the transformer tap position based on the measured transformer ratio to obtain the transformer tap position includes: Based on the transformer's rated turns ratio, tap range, and number of taps, the theoretical turns ratio corresponding to each tap is generated. The measured turns ratio is compared with the theoretical turns ratio corresponding to each gear position to determine the gear position corresponding to the measured turns ratio, which is then used as the identified transformer gear position.

3. The intelligent diagnostic method for drop-out high-voltage fuses according to claim 2, characterized in that, The step of verifying the consistency between the stable value of the measured transformer ratio and the theoretical transformer ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position includes: When the deviation between the stable value of the measured transformer ratio and the corresponding theoretical transformer ratio is less than or equal to a preset deviation threshold, the identified transformer tap position is determined to be the verified transformer tap position. When the deviation between the stable value of the measured gear ratio and the corresponding theoretical gear ratio is greater than a preset deviation threshold, gear identification is performed again until the deviation between the stable value of the measured gear ratio and the corresponding theoretical gear ratio is less than or equal to the preset deviation threshold.

4. The intelligent diagnostic method for drop-out high-voltage fuses according to claim 3, characterized in that, The preset fault diagnosis model is based on an LSTM / GRU network. The input is fault diagnosis features, and the output is the fuse's health score, fault risk index, and the determined fuse status.

5. The intelligent diagnostic method for drop-out high-voltage fuses according to claim 4, characterized in that, The step of determining the fault based on the fault diagnosis features and the preset fault diagnosis model to obtain the fault diagnosis result of the fuse includes: The fault diagnosis features are input into the preset fault diagnosis model; Based on the preset fault diagnosis model, the fault diagnosis features are predicted in time sequence and anomalies are detected, and the health score and fault risk index of the fuse are calculated. Based on the health score and fault risk indicators, the current status of the fuse is determined. The status of the fuse includes normal closing, fault tripping, accidental tripping, or non-tripping fault. The system outputs the health score, fault risk indicators, and the current status of the determined fuse to obtain the diagnostic results of the fuse.

6. A smart diagnostic system for drop-out high-voltage fuses, characterized in that, The drop-out high-voltage fuse is connected to the three-phase energy meter of the transformer, and the system includes: Identification module: Based on the collected electrical data, it identifies the transformer tap position and obtains the transformer tap position; Verification module: Verifies the transformer tap position to obtain the verified transformer tap position; Extraction module: Based on the verified transformer tap position and electrical data, extract the fault diagnosis features of the fuse; The discrimination module: Based on the fault diagnosis features and the preset fault diagnosis model, it performs fault discrimination and obtains the diagnosis result of the fuse; The step of identifying the transformer tap position based on the collected electrical data to obtain the transformer tap position includes: Collect three-phase voltage and three-phase current data on the high-voltage side of the transformer, as well as three-phase voltage and three-phase current data on the low-voltage side collected by the transformer's three-phase energy meter; The three-phase voltages on the high-voltage side and the three-phase voltages on the low-voltage side are filtered to calculate the fundamental effective values ​​of the high-voltage side and the low-voltage side. The measured turns ratio of the transformer is calculated based on the fundamental effective values ​​of the high-voltage side and the low-voltage side. Based on the measured turns ratio of the transformer, the transformer tap position is identified to obtain the transformer tap position; The step of verifying the transformer tap position to obtain the verified transformer tap position includes: Within a preset time window, obtain the measured ratios at multiple points in time; Statistical analysis was performed on the measured ratios at the multiple time points to obtain stable values ​​of the measured ratios; The stability value of the measured turns ratio is checked against the theoretical turns ratio corresponding to the identified transformer tap position to obtain the verified transformer tap position. Based on the verified transformer tap position and electrical data, the following features are extracted for fuse fault diagnosis: Based on the verified transformer tap position, the electrical data is corrected for operating conditions to obtain tap-corrected electrical data, including: The collected low-voltage side three-phase voltage and low-voltage side three-phase current data are converted to the high-voltage side equivalent voltage and equivalent current according to the measured turns ratio. The equivalent voltage and equivalent current are used as the electrical data after gear position correction; Based on the electrical data before correction and the electrical data after gear correction, the fault diagnosis features of the fuse are extracted.

Images