A transformer oil conservator oil level full-range verification management method and system

Abstract

The application provides a transformer oil conservator oil level full-range verification management method and system, and belongs to the technical field of verification management, and specifically comprises the following steps: determining the matching degree of the range verification result of the oil conservator and the prediction result of the monitoring data based on the oil temperature by using the range verification management method; when it is determined that the optimization processing of the monitoring data of the oil temperature monitoring device is needed based on the matching degree, determining the deviation checking processing process in different subintervals based on the matching degree; and determining the abnormal processing strategy of the oil temperature monitoring device by combining the correlation degree of different subintervals with different oil temperature intervals in the verification analysis process, so that the accuracy of the transformer oil conservator oil level range verification management is improved.

Description

Technical Field

[0001] This invention belongs to the field of verification management technology, and in particular relates to a method and system for full-range verification management of oil level in transformer oil conservator. Background Technology

[0002] The transformer oil conservator (also known as an oil tank) is a crucial component of a transformer. It compensates for the volume expansion and contraction of transformer oil due to temperature changes and provides gas storage space for the gas relay. The oil level in the conservator is a critical monitoring parameter for the transformer's operating status—a low oil level may indicate oil leakage or seal failure, while a high level may indicate a ruptured oil tank bladder or a malfunctioning level gauge. Accurately determining the true oil level in the conservator is essential for the safe operation of the transformer. After long-term operation, level gauges (such as magnetic needle level gauges and float level gauges) may experience malfunctions such as jamming, pointer deviation, and wear of the transmission mechanism, leading to discrepancies between the displayed and actual oil levels. However, maintenance personnel may find it difficult to determine the accuracy of the level gauge.

[0003] To address the aforementioned issues, existing technical solutions utilize magnetic needle-type oil level gauges and remote transmitters for periodic automated verification of the oil level gauges. However, this approach suffers from the following technical problems: As the transformer oil temperature fluctuates, the transformer oil level will also fluctuate. Therefore, if the reliability of the transformer oil temperature monitoring data is ignored, it will be difficult to effectively identify abnormal fluctuations in the transformer oil level. Thus, determining the verification and analysis method for the oil temperature monitoring device based on the risk of transformer oil level fluctuations, and determining the full-range verification and management method for the transformer oil conservator oil level and the abnormal handling method for the oil temperature monitoring device based on the verification and analysis results of the oil temperature monitoring device, thereby ensuring the reliability of the oil level verification and analysis, has become an urgent technical problem to be solved.

[0004] Specifically, this application provides a method and system for full-range verification and management of oil level in transformer oil conservator. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides a method for full-range verification and management of transformer oil level in a transformer oil conservator, which includes: S1 uses the monitoring data of the transformer oil conservator to determine the oil level fluctuation range of the transformer oil conservator, and uses the oil level fluctuation range and the running time in different sub-ranges within the oil level fluctuation range to determine the verification analysis and management strategy of the oil temperature monitoring device. S2 uses the verification analysis management strategy to determine the consistency between the oil temperature monitoring device and the oil temperature prediction model, and based on the consistency in different oil temperature ranges, determines the suspected deviation oil temperature range, and combines the running time and consistency in the suspected deviation oil temperature range to determine the range verification management method of the transformer oil conservator. S3 uses the range verification management method to determine the degree of matching between the range verification result of the oil tank and the prediction result based on the oil temperature monitoring data. When it is determined based on the degree of matching that the monitoring data of the oil temperature monitoring device needs to be optimized, proceed to the next step. S4 determines the deviation verification process within different sub-intervals based on the matching degree, and determines the abnormal handling strategy of the oil temperature monitoring device by combining the correlation degree between different sub-intervals and different oil temperature intervals during the verification analysis process.

[0006] The beneficial effects of this invention are as follows: Based on the oil level fluctuation range and the running time in different sub-ranges within the oil level fluctuation range, the reliability of oil level verification analysis and processing in different sub-ranges is determined. Based on the reliability of oil level verification analysis and processing in different sub-ranges, the verification analysis and management strategy of oil temperature detection device is determined, thus laying the foundation for determining the range verification management method of oil level using the reliability of monitoring and processing of oil temperature monitoring device.

[0007] Based on the deviation verification and processing process in different sub-intervals and the correlation between different sub-intervals and oil temperature ranges, the reliability of oil temperature monitoring in different oil temperature ranges and its impact on the overall reliability of oil level range monitoring are determined. Based on the impact of the reliability of oil temperature monitoring in different oil temperature ranges on the overall reliability of oil level range monitoring, anomaly handling strategies for the oil temperature monitoring device are determined, thereby ensuring the reliability and accuracy of oil temperature monitoring data and range verification and analysis results.

[0008] Furthermore, the monitoring data of the transformer oil conservator includes the monitoring results of the oil level of the transformer oil conservator at different time periods.

[0009] Furthermore, the oil level fluctuation range of the transformer oil conservator is determined based on the range corresponding to the monitoring results of the oil level in the transformer oil conservator.

[0010] Furthermore, the method for determining the verification analysis and management strategy of the oil temperature monitoring device is as follows: Determine the deviation rate between the endpoints of the oil level fluctuation range based on the oil level fluctuation range; The percentage of runtime of each sub-interval is determined by the runtime in different sub-intervals within the oil level fluctuation range, and this percentage is used as the runtime matching coefficient of the sub-interval. By utilizing the deviation rate between the endpoints of the oil level fluctuation range and the operating matching coefficients of different sub-ranges, the verification analysis and management strategy of the oil temperature monitoring device is determined.

[0011] Furthermore, the method for determining the abnormal handling strategy of the oil temperature monitoring device is as follows: Based on the deviation verification process within the sub-interval, the verification impact coefficient of the sub-interval is determined; Based on the degree of correlation between the sub-interval and different oil temperature ranges during the verification analysis, the correlation weight value between the sub-interval and different oil temperature ranges is determined. Based on the verification influence coefficients of different sub-intervals and the correlation weight values ​​between the sub-intervals and different oil temperature intervals, the abnormal handling strategy of the oil temperature monitoring device is determined.

[0012] Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for full-range verification and management of oil level in a transformer oil conservator when running the computer program.

[0013] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0014] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0015] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0016] Figure 1 This is a flowchart of a method for full-range verification and management of transformer oil level in a transformer oil conservator. Figure 2 This is a flowchart illustrating the method for determining the verification analysis and management strategy for oil temperature monitoring devices; Figure 3 This is a flowchart illustrating the method for determining the range verification management method for transformer oil conservators. Detailed Implementation

[0017] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0018] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements / components / etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements / components / etc. may exist in addition to the listed elements / components / etc.

[0019] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, a method for full-range verification and management of transformer oil level in a transformer oil conservator is provided, specifically including: S1 uses the monitoring data of the transformer oil conservator to determine the oil level fluctuation range of the transformer oil conservator, and uses the oil level fluctuation range and the running time in different sub-ranges within the oil level fluctuation range to determine the verification analysis and management strategy of the oil temperature monitoring device. The core objective of this embodiment is to determine the verification and management strategy for the oil temperature monitoring device by analyzing the oil level fluctuation range and sub-range operating time of the transformer oil conservator. It then determines the suspected deviation oil temperature range and range verification management method based on the consistency between the oil temperature monitoring device and the oil temperature prediction model. Next, it judges whether monitoring data optimization is needed based on the matching degree between the range verification results and the prediction results. Finally, it determines the anomaly handling strategy based on the deviation verification process and the correlation between the sub-range and the oil temperature range. The core logic uses oil conservator monitoring data as input. First, it determines the verification and management strategy based on the endpoint deviation rate of the oil level fluctuation range and the sub-range operating matching coefficient. Then, it determines the suspected deviation oil temperature range and range verification management method based on the consistency of the oil temperature range and the deviation verification process. Next, it judges whether optimization is needed based on the matching degree. Finally, it determines the anomaly handling strategy based on the verification influence coefficient and correlation weight value. The overall logic follows the process of "verification and analysis management strategy determination → range verification management method determination → optimization judgment → anomaly handling strategy determination".

[0020] Furthermore, the monitoring data of the transformer oil conservator includes the monitoring results of the oil level of the transformer oil conservator at different time periods.

[0021] The monitoring data refers to the oil level records collected by the oil level sensor in the transformer oil tank at different time periods, and the oil temperature records collected by the oil temperature monitoring device.

[0022] Assuming that a transformer oil conservator continuously collects data through an oil level sensor and an oil temperature monitoring device during its historical operation, the oil level and oil temperature records for each time period constitute the monitoring data of the oil conservator.

[0023] The content of the monitoring data was clarified, and the actual oil level and oil temperature data were used as the objective basis for verifying and analyzing the management strategy, thereby avoiding the one-sidedness of relying solely on theoretical models to determine the strategy.

[0024] Specifically, the oil level fluctuation range of the transformer oil conservator is determined based on the range corresponding to the monitoring results of the oil level in the transformer oil conservator.

[0025] The oil level fluctuation range refers to the range determined by the maximum and minimum values ​​of the oil level monitoring results in the transformer oil tank during historical operation, reflecting the fluctuation amplitude of the oil level.

[0026] Assuming that the lowest oil level in a certain oil storage tank during its historical operation was 25% and the highest was 75%, then the oil level fluctuation range is 25% to 75%.

[0027] The method for determining the oil level fluctuation range was clarified, and the actual fluctuation range of historical oil level monitoring results was used as the basis for determining the subsequent sub-range division and verification analysis management strategy.

[0028] Specifically, such as Figure 2 As shown, the method for determining the verification analysis and management strategy of the oil temperature monitoring device is as follows: In this embodiment, the reliability of oil level verification analysis is determined based on the oil level fluctuation range and the running time in different sub-ranges within the oil level fluctuation range. Based on the reliability of oil level verification analysis in different sub-ranges, the verification analysis management strategy of the oil temperature detection device is determined, thus laying the foundation for determining the range verification management method of oil level using the reliability of the monitoring and processing of the oil temperature monitoring device.

[0029] The verification analysis management strategy refers to the decision-making scheme for determining under what conditions the oil temperature monitoring device should be verified and analyzed, including the baseline period verification strategy and the baseline period plus cumulative duration trigger verification strategy; the verification analysis process refers to the process of evaluating the monitoring reliability of the oil temperature monitoring device by comparing the monitoring data of the oil temperature monitoring device with the prediction results of the oil temperature prediction model.

[0030] If the oil level fluctuates significantly and some sub-intervals have short operating times, it indicates that the verification reliability in these sub-intervals is insufficient. Therefore, a more frequent verification strategy should be adopted to detect monitoring deviations in a timely manner.

[0031] By determining the verification analysis and management strategy based on the oil level fluctuation range and the running time of the sub-range, the oil level fluctuation information and the running distribution information of the sub-range are transformed into the basis for selecting the verification strategy, thus laying the foundation for determining the range verification management method based on the reliability of oil temperature monitoring.

[0032] S11, using the oil level fluctuation range, determine the deviation rate between the endpoints of the oil level fluctuation range; The deviation rate between the endpoints refers to the ratio of the difference between the upper and lower limits of the oil level fluctuation range to the lower limit, reflecting the severity of the oil level fluctuation.

[0033] Assuming the oil level fluctuation range is 25% to 75%, then the deviation rate = (75-25) / 25 = 2.0.

[0034] The degree of oil level fluctuation is quantified by the endpoint deviation rate, providing a quantitative basis for determining the verification strategy based on the difference in the degree of fluctuation.

[0035] Specifically, if the deviation rate between the endpoints of the oil level fluctuation range is less than the preset deviation rate threshold, it indicates that the oil level in the transformer oil tank is relatively stable. Therefore, the verification analysis and management strategy for the oil temperature monitoring device is determined to be to perform verification analysis processing of the oil temperature monitoring device according to the benchmark cycle.

[0036] The preset deviation rate threshold refers to the critical value for judging whether the oil level fluctuation is stable; the benchmark period refers to the time interval for verification analysis.

[0037] If the deviation rate is less than the preset deviation rate threshold, it indicates that the oil level is relatively stable. In this case, the benchmark cycle verification strategy can be adopted, and verification can be carried out at a fixed cycle.

[0038] When the oil level is stable, a baseline periodic verification strategy is adopted. When the oil level fluctuates little, regular verification can ensure the reliability of monitoring, thereby avoiding unnecessary frequent verification and saving resources.

[0039] It is also understood that if the deviation rate between the endpoints of the oil level fluctuation range is not less than the preset deviation rate threshold, the process proceeds to step S12. S12 determines the percentage of runtime of the sub-interval based on the runtime in different sub-intervals within the oil level fluctuation range, and uses it as the runtime matching coefficient of the sub-interval. The sub-interval refers to multiple sub-intervals divided according to the oil level fluctuation range; the operation matching coefficient refers to the proportion of the running time in a certain sub-interval to the total running time of the oil level fluctuation range, and the value ranges from 0 to 1. The larger the value, the more fully the sub-interval is operated.

[0040] Assuming the oil level fluctuation range of 25% to 75% is divided into 5 sub-ranges, each sub-range being 10% wide, and the running time of a certain sub-range accounting for 30% of the total running time, then the running matching coefficient of that sub-range is 0.30.

[0041] The running matching coefficient is calculated by the proportion of running time, which quantifies the sufficiency of running in each interval from the perspective of sub-intervals, and provides a quantitative basis for the refined determination of subsequent verification strategies.

[0042] S13 uses the deviation rate between the endpoints of the oil level fluctuation range and the operating matching coefficient of different sub-ranges to determine the verification analysis and management strategy of the oil temperature monitoring device.

[0043] Assuming the deviation rate and the matching coefficients of each sub-interval have been determined, the verification analysis management strategy is determined based on the judgment results of Case 1 and Case 2 respectively.

[0044] The verification strategy is determined by combining the deviation rate and the operational matching coefficient, and the adaptive hierarchical adjustment of the verification strategy determination strategy is realized.

[0045] It is understandable that the verification analysis and management strategy for the oil temperature monitoring device is determined by utilizing the deviation rate between the endpoints of the oil level fluctuation range and the operating matching coefficients of different sub-ranges, specifically including: It should be noted that the determination of the verification analysis management strategy includes the following two situations.

[0046] Case 1: If the matching coefficient in different sub-intervals is greater than the preset matching coefficient threshold, then the verification analysis management strategy of the oil temperature monitoring device is determined to be to perform verification analysis processing of the oil temperature monitoring device according to the benchmark cycle in all oil temperature intervals. The preset matching coefficient threshold refers to the critical value for determining whether the matching coefficient of the sub-interval is sufficient.

[0047] If the running matching coefficient of all sub-intervals is greater than the preset matching coefficient threshold, it indicates that the running of each sub-interval is sufficient and the verification reliability is good. In this case, the benchmark period verification strategy can be adopted.

[0048] When all sub-intervals have been running sufficiently, a baseline periodic verification strategy is adopted. When the verification reliability of each sub-interval is good, the monitoring reliability can be guaranteed by periodic verification, thereby avoiding unnecessary frequent verification.

[0049] Scenario 2: If the matching coefficients in different sub-intervals are not all greater than the preset matching coefficient threshold, the sub-intervals with matching coefficients not greater than the preset matching coefficient threshold are taken as matching deviation intervals. It is determined whether the number of matching deviation intervals is greater than the preset value for the number of deviation intervals. If so, the verification analysis management strategy of the oil temperature monitoring device is determined to be that, in addition to performing verification analysis processing of the oil temperature monitoring device according to the benchmark cycle in all oil temperature intervals, verification analysis processing of the oil temperature monitoring device is performed as long as the cumulative running time in any sub-interval reaches the first running time in the most recent preset time period. If not, the verification analysis management strategy of the oil temperature monitoring device is determined to be that, in addition to performing verification analysis processing of the oil temperature monitoring device according to the benchmark cycle in all oil temperature intervals, verification analysis processing of the oil temperature monitoring device is performed as long as the cumulative running time in any matching deviation interval reaches the first running time in the most recent preset time period.

[0050] The matching deviation interval refers to the sub-interval where the running matching coefficient does not meet the standard, that is, the running of the interval is insufficient and the verification reliability is insufficient; the preset value of the number of deviation intervals refers to the critical value for judging whether the number of matching deviation intervals is too large; the first running time refers to the cumulative running time threshold for triggering additional verification.

[0051] If the number of matching deviation intervals is large, indicating that multiple sub-intervals are not running sufficiently, then in addition to the baseline period verification, verification will be performed when the cumulative running time of any sub-interval reaches the first running time; if the number of matching deviation intervals is small, then in addition to the baseline period verification, verification will be performed only when the cumulative running time of any matching deviation interval reaches the first running time.

[0052] When the sub-interval is not running sufficiently, a verification strategy is adopted that uses a base period plus cumulative duration. When some sub-intervals are not running sufficiently, additional verification is triggered by the cumulative running time to detect monitoring deviations in a timely manner. At the same time, the trigger range is dynamically adjusted when the number of matching deviation intervals is different, thereby achieving an adaptive balance between verification reliability and processing efficiency.

[0053] This embodiment, through steps S11 to S13, realizes a complete process from calculating the deviation rate of oil level fluctuation range to analyzing the matching coefficient of sub-range operation and determining the verification analysis management strategy. Its core value is reflected in three aspects: First, by judging the deviation rate, a benchmark cycle verification strategy is adopted when the oil level is stable to save resources; second, by judging the operation matching coefficient, a benchmark cycle verification strategy is adopted when all sub-ranges have been fully operated to ensure efficiency; third, by judging the hierarchical nature of the number of matching deviation ranges, the verification trigger range is dynamically adjusted when the sub-range operation is insufficient, thereby achieving an adaptive balance between verification reliability and processing efficiency.

[0054] During a 30-day operation, the oil level monitoring results for a transformer's oil conservator showed a minimum of 30% and a maximum of 70%, with a fluctuation range of 30% to 70%. This fluctuation range was divided into four sub-ranges: Z1 (30% to 40%), Z2 (40% to 50%), Z3 (50% to 60%), and Z4 (60% to 70%). The operating time of each sub-range over the 30 days was: Z1 72 hours, Z2 216 hours, Z3 288 hours, and Z4 144 hours, for a total operating time of 720 hours.

[0055] Step S11: Deviation rate = (70-30) / 30 = 1.33.

[0056] Determine if the deviation rate of 1.33 is less than the preset deviation rate threshold of 1.0: 1.33 is not less than 1.0, no, proceed to S12.

[0057] Step S12: Calculate the running matching coefficient for each sub-interval.

[0058] Z1: 72 / 720 = 0.10; Z2: 216 / 720 = 0.30; Z3: 288 / 720 = 0.40; Z4: 144 / 720 = 0.20; Step S13: Determine whether the matching coefficient of each sub-interval is greater than the preset matching coefficient threshold of 0.15.

[0059] Z1: 0.10 is not greater than 0.15, and is not uniformly greater than 0.15, which belongs to case 2.

[0060] The matching deviation range is Z1 (0.10≤0.15), with a total of 1 range.

[0061] Determine if the number of matching deviation intervals (1) is greater than the preset value of the number of deviation intervals (2): 1 is not greater than 2, no.

[0062] The verification analysis management strategy is as follows: In addition to performing verification analysis processing according to the baseline cycle of 24 hours in all oil temperature ranges, verification analysis processing of the oil temperature monitoring device will be performed whenever the cumulative running time in any matching deviation range (i.e., Z1) reaches the first running time of 20 hours in the most recent 7 days.

[0063] S2 uses the verification analysis management strategy to determine the consistency between the oil temperature monitoring device and the oil temperature prediction model, and based on the consistency in different oil temperature ranges, determines the suspected deviation oil temperature range, and combines the running time and consistency in the suspected deviation oil temperature range to determine the range verification management method of the transformer oil conservator. Specifically, such as Figure 3As shown, the method for determining the range verification management method of the transformer oil conservator is as follows: In this embodiment, the deviation risk of the monitoring and analysis deviation of the transformer oil conservator in different oil temperature ranges is determined based on the degree of consistency in different oil temperature ranges. Based on the deviation risk of the monitoring and analysis deviation of the transformer oil conservator in different oil temperature ranges, the range verification management method of the transformer oil conservator is determined. This reduces the difficulty of verification and analysis while ensuring the reliable verification and processing of the transformer oil conservator in combination with oil temperature monitoring data throughout the entire range.

[0064] The range verification management method refers to the decision-making scheme that determines under what conditions oil level prediction verification is performed based on oil temperature monitoring data; the consistency degree refers to the degree of agreement between the monitoring data of the oil temperature monitoring device and the prediction results of the oil temperature prediction model; the deviation verification process refers to the verification process when the deviation rate between the range verification result and the prediction result based on oil temperature exceeds a preset deviation rate threshold.

[0065] If the consistency between the oil temperature monitoring device and the prediction model is low within a certain oil temperature range, it indicates that the monitoring deviation risk in that range is relatively high. Therefore, a more stringent range verification management method should be adopted to detect monitoring deviations in a timely manner.

[0066] S21 determines the deviation verification process in the transformer oil conservator in different oil temperature ranges based on the degree of consistency in different oil temperature ranges. The deviation verification process refers to the verification process in which the deviation rate between the range verification result and the prediction result based on oil temperature exceeds a preset deviation rate threshold during the verification analysis process in a certain oil temperature range.

[0067] If, during the verification process within a certain oil temperature range, the deviation rate exceeds the preset deviation rate threshold multiple times, then these verification processes are identified as deviation verification processes.

[0068] By screening the deviation rate during the deviation verification process, processes with large deviations are identified, providing a basis for determining the oil temperature range of suspected deviations in the future.

[0069] S22, the oil temperature range where the quantity of deviation verification and processing does not meet the requirements is regarded as the suspected deviation oil temperature range; The suspected deviation oil temperature range refers to the oil temperature range where the number of deviation verification processes is relatively large and exceeds the preset deviation process number threshold, meaning that the risk of deviation monitoring in this range is relatively high.

[0070] If the number of deviation verification processes in a certain oil temperature range reaches or exceeds the preset deviation process number threshold, then the oil temperature range is identified as a suspected deviation oil temperature range.

[0071] By screening suspected deviation oil temperature ranges through the deviation verification process, the ranges with higher monitoring deviation risk are identified from all oil temperature ranges, providing a basis for determining the subsequent range verification management method.

[0072] S23 determines the range verification management method for the transformer oil conservator based on the number of suspected deviation oil temperature ranges, the deviation verification process in different suspected deviation oil temperature ranges, and the running time.

[0073] Assuming the number of suspected deviation oil temperature ranges and the deviation verification process and runtime for each range have been determined, the range verification management method is determined based on whether the quantity meets the requirements and the judgment results of Case 1 and Case 2.

[0074] The range verification management method is determined by combining the number of suspected deviation intervals and deviation verification data, and the range verification management method is adaptively classified.

[0075] It should be noted that the suspected deviation oil temperature range is a sub-range where the number of deviation verification processes exceeds a preset threshold for the number of deviation processes.

[0076] It is understandable that if the number of suspected deviation oil temperature ranges does not meet the requirements, i.e., it is greater than a certain range number threshold, then the reliability of range verification processing based on oil temperature is low. Therefore, as long as the total duration of the suspected deviation oil temperature ranges is less than the preset duration threshold within the most recent preset time period, i.e., based on the monitoring data of oil temperature at different times within the most recent preset time period, the oil level prediction processing is performed to obtain the prediction result, and the prediction result is used to perform the verification analysis processing of the transformer oil conservator.

[0077] The term "not meeting the requirements" refers to the situation where the number of suspected deviation oil temperature intervals exceeds the preset interval number threshold, i.e., there are many deviation intervals and the reliability of verification based on oil temperature is low; the preset interval number threshold is the critical value for judging whether the number of suspected deviation intervals is too large; the preset duration threshold is the lower limit of the total duration for triggering verification based on oil temperature prediction.

[0078] If the number of suspected deviation intervals is large, it indicates that the reliability of oil temperature-based verification is low. Therefore, as long as the total duration of suspected deviation intervals in the most recent preset time period is greater than the preset duration threshold, verification processing based on oil temperature prediction will be performed.

[0079] When there are many suspected deviation ranges, an oil temperature-based verification strategy is adopted. When the reliability of oil temperature-based verification is low, verification is performed using oil temperature prediction results to ensure the timeliness of verification.

[0080] Additionally, it should be noted that if the number of suspected deviation oil temperature ranges meets the requirements, it includes the following: It should be noted that the range verification management method for determining when the number of suspected deviation ranges meets the requirements includes the following two situations.

[0081] Scenario 1: If the number of suspected deviation oil temperature intervals meets the requirements, the verification analysis impact value of the suspected deviation oil temperature interval is determined by the proportion of deviation verification processing processes and the proportion of running time within the suspected deviation oil temperature interval. When there is no suspected deviation oil temperature interval with a verification analysis impact value greater than the preset impact threshold, as long as the total duration of the suspected deviation oil temperature interval is less than the second preset duration threshold and there is no verification processing process within the most recent preset time period, the oil level prediction processing is performed based on the oil temperature monitoring data at different times within the most recent preset time period to obtain the prediction result. The prediction result is then used for the verification analysis processing of the transformer oil conservator.

[0082] The verification analysis impact value refers to a quantitative indicator calculated based on the proportion of deviation verification processing processes and the proportion of runtime; the preset impact threshold refers to a critical value for judging whether the verification analysis impact value is high; the second preset duration threshold refers to a stricter upper limit of total duration that is less than the preset duration threshold. Duration threshold relationship: Second preset duration threshold < Preset duration threshold.

[0083] The calculation method for the impact value of the verification analysis is as follows: Let P1 be the proportion of the number of deviation verification processes within a certain suspected deviation oil temperature range, and P2 be the proportion of the running time. Then: Impact value of verification analysis = P1 + P2.

[0084] Assuming there are no suspected deviation oil temperature ranges where the verification analysis impact value is greater than the preset impact threshold, indicating that the degree of deviation impact of each suspected deviation range is controllable, verification based on oil temperature prediction is performed when the total duration of the suspected deviation range is less than the second preset duration threshold and there has been no recent verification process.

[0085] When there is no high-impact value range, a more lenient prediction verification trigger condition is adopted. When the deviation impact of each suspected deviation range is controllable, the timeliness of verification is ensured by a more lenient trigger condition, thereby achieving a balance between verification reliability and processing efficiency.

[0086] Scenario 2: When there is a suspected deviation oil temperature range whose verification analysis impact value is greater than the preset impact threshold, the verification impact coefficient is determined based on the proportion of suspected deviation oil temperature ranges in the sub-range and the verification analysis impact value between different suspected deviations. It is then determined whether the verification impact coefficient is greater than the preset impact coefficient threshold. If so, as long as the total duration of the suspected deviation oil temperature range is less than the preset duration threshold in the most recent preset time period, the oil level prediction is performed based on the oil temperature monitoring data at different times in the most recent preset time period to obtain the prediction result. The prediction result is then used for the verification analysis of the transformer oil conservator. If not, as long as the total duration of the suspected deviation oil temperature range is less than the preset duration threshold (greater than the second preset duration threshold) in the most recent preset time period and there is no verification process in the most recent preset time period, the oil level prediction is performed based on the oil temperature monitoring data at different times to obtain the prediction result. The prediction result is then used for the verification analysis of the transformer oil conservator.

[0087] The verification impact coefficient refers to a quantitative indicator calculated by combining the proportion of suspected deviation oil temperature ranges and the impact value of verification analysis; the preset impact coefficient threshold refers to the critical value for judging whether the verification impact coefficient is high.

[0088] The verification impact coefficient is calculated as follows: Let P be the proportion of suspected deviation oil temperature intervals in the sub-intervals, and let Iavg be the mean value of the verification analysis impact value of different suspected deviation oil temperature intervals. Then: Verification impact coefficient = P interval × Iavg.

[0089] If the calculated verification impact coefficient is greater than the preset impact coefficient threshold, it indicates that the suspected deviation interval has a large impact. In this case, a more lenient triggering condition is adopted (verification is performed if the total duration is less than the preset duration threshold). If it is not greater than the threshold, a more stringent triggering condition is adopted (verification is performed only if the total duration is less than the preset duration threshold and there has been no verification process recently).

[0090] When there is a high-impact value range, the triggering conditions are further distinguished by verifying the impact coefficient. When the degree of deviation impact is different, the triggering strictness of the prediction verification is dynamically adjusted, thereby achieving a fine balance between verification reliability and processing efficiency.

[0091] This embodiment, through steps S21 to S23, realizes a complete process from determining the deviation verification process to identifying the suspected deviation oil temperature range and then determining the range verification management method. Its core value is reflected in three aspects: First, by judging the number of suspected deviation ranges, an oil temperature prediction-based verification strategy is adopted to ensure the timeliness of verification when there are many deviation ranges; second, by judging the impact value of verification, a more lenient triggering condition is adopted to ensure efficiency when there are no high impact value ranges; third, by judging the graded impact coefficient of verification, the triggering strictness is dynamically adjusted when there are high impact value ranges, thereby achieving an adaptive balance between verification reliability and processing efficiency.

[0092] Based on the calculation results of S1, the verification analysis management strategy has been determined. The oil temperature range is divided into three intervals: T1 (0 to 40℃), T2 (40 to 60℃), and T3 (60 to 90℃). In the verification analysis process, the number of deviation verification processes for each oil temperature interval is as follows: 3 for T1, 6 for T2, and 2 for T3, with a total of 15, 20, and 10 verification processes respectively.

[0093] Step S21: Determine the deviation verification process for each oil temperature range (verification process when the deviation rate exceeds the preset deviation rate threshold).

[0094] T1: Three deviation verification and processing procedures; T2: Six deviation verification and processing procedures; T3: Two deviation verification and processing procedures; Step S22: Determine whether each oil temperature range is a suspected deviation oil temperature range (the number of deviation verification and processing steps is ≥5).

[0095] T1: 3 < 5, not belonging. T2: 6 ≥ 5, belonging. T3: 2 < 5, not belonging.

[0096] The suspected deviation oil temperature range is T2, with a total of 1 range.

[0097] Step S23: Determine whether the number of suspected deviation oil temperature ranges 1 does not meet the requirements (i.e., exceeds the preset range number threshold 2): 1 is not greater than 2, which meets the requirements (not many).

[0098] It belongs to case 1 or case 2.

[0099] Calculate the impact value of verification analysis for suspected deviation oil temperature range T2: the proportion of deviation verification process number = 6 / 20 = 0.30, and the proportion of runtime is assumed to be 0.40.

[0100] The impact value of the verification analysis is 0.30 + 0.40 = 0.7.

[0101] Determine if there is a suspected deviation oil temperature range where the verification analysis impact value is greater than the preset impact threshold of 0.8: T2's 0.7 is not greater than 0.8, so it does not exist and belongs to case 1.

[0102] According to scenario 1, the range verification management method is as follows: as long as the total duration of the suspected deviation oil temperature range (T2) is less than the second preset duration threshold (2 minutes) within the most recent preset time period and there is no verification process within the most recent preset time period (e.g., 3 hours), the oil level prediction process is performed based on the monitoring data of oil temperature at different times within the most recent preset time period (20 minutes) to obtain the prediction result, and the prediction result is used for the verification analysis of the transformer oil conservator.

[0103] S3 uses the range verification management method to determine the degree of matching between the range verification result of the oil tank and the prediction result based on the oil temperature monitoring data. When it is determined based on the degree of matching that the monitoring data of the oil temperature monitoring device needs to be optimized, proceed to the next step. It is understandable that it is necessary to optimize the monitoring data from the oil temperature monitoring device, specifically including: In this embodiment, the deviation risk of the range verification analysis and processing results is determined based on the degree of matching between the range verification results of the oil tank and the prediction results based on the oil temperature monitoring data in different sub-intervals. Based on the deviation risk of the range verification analysis and processing results, it is determined whether the monitoring data of the oil temperature monitoring device needs to be optimized, thereby further improving the reliability of the range verification analysis and processing results.

[0104] The optimization process refers to the process of correcting or recalibrating the monitoring data of the oil temperature monitoring device; the matching degree refers to the degree of agreement between the range verification results and the oil temperature prediction results.

[0105] If the verification results deviate significantly from the predicted results in multiple sub-intervals, it indicates a high risk of range verification deviation. In such cases, it is necessary to optimize the monitoring data of the oil temperature monitoring device to improve the reliability of the verification.

[0106] S31 Based on the matching degree, determine the verification process where the deviation rate between the range verification result of the oil tank and the prediction result based on the oil temperature monitoring data is greater than a preset deviation rate threshold in different sub-intervals, and use it as the deviation verification process. The deviation verification process (used in S3) refers to the verification process when the deviation rate between the range verification result and the oil temperature-based prediction result exceeds a preset deviation rate threshold.

[0107] If, during multiple verification processes within a certain sub-interval, some deviation rates exceed a preset deviation rate threshold, these processes are identified as deviation verification processes.

[0108] S32 determines whether optimization processing of the monitoring data from the oil temperature monitoring device is required by checking the deviation in different sub-intervals.

[0109] Assuming that the deviation verification process for each sub-interval has been determined, the determination of whether optimization is needed is based on the judgment result of the verification influence coefficient.

[0110] The significance of determining whether optimization is needed through the deviation verification process lies in achieving adaptive judgment in the optimization decision.

[0111] It is understandable that the deviation verification process within different sub-intervals is used to determine whether optimization processing of the oil temperature monitoring device's monitoring data is necessary, specifically including: The verification impact coefficient of the sub-interval is determined by the proportion of the deviation verification process in different sub-intervals among all verification processes. The verification impact coefficient refers to the proportion of the number of deviation verification processes within a certain sub-interval to the total number of verification processes in that sub-interval.

[0112] Suppose that a certain sub-interval has 20 verification processes, of which 6 are deviation verification processes, then the verification impact coefficient = 6 / 20 = 0.30.

[0113] The impact coefficient of the deviation verification process is calculated by measuring the proportion of deviation verification. The impact of deviation verification is quantified from the sub-interval dimension, providing a quantitative basis for judging whether subsequent optimization is needed.

[0114] When the number of sub-intervals with an impact coefficient greater than the preset impact coefficient threshold exceeds the preset number of intervals, it is necessary to accurately determine the monitoring reliability of different oil temperature intervals to ensure the reliability of oil level verification. Therefore, it is determined that the monitoring data of the oil temperature monitoring device needs to be optimized.

[0115] The preset influence coefficient threshold (used in S3) refers to the critical value for judging whether the influence coefficient of the sub-interval is high; the preset value of the number of intervals refers to the critical value for judging whether the number of sub-intervals with high influence coefficients is large.

[0116] If the number of sub-intervals whose verification influence coefficient exceeds the preset influence coefficient threshold reaches a preset value for the number of intervals, it indicates that the verification of deviations in multiple sub-intervals has a significant impact and the risk of deviation in range verification is high. In this case, it is necessary to optimize the monitoring data of the oil temperature monitoring device.

[0117] When there are many sub-intervals with excessive influence coefficients, optimization processing is triggered. When the verification deviation risk of multiple sub-intervals is high, optimization processing is used to improve the accuracy of monitoring data, thereby ensuring the reliability of range verification.

[0118] S4 determines the deviation verification process within different sub-intervals based on the matching degree, and determines the abnormal handling strategy of the oil temperature monitoring device by combining the correlation degree between different sub-intervals and different oil temperature intervals during the verification analysis process.

[0119] Specifically, the method for determining the abnormal handling strategy of the oil temperature monitoring device is as follows: In this embodiment, based on the deviation verification process in different sub-intervals and the correlation between different sub-intervals and oil temperature ranges, the reliability of oil temperature monitoring in different oil temperature ranges and its impact on the overall reliability of oil level range monitoring are determined. Based on the impact of the reliability of oil temperature monitoring in different oil temperature ranges on the overall reliability of oil level range monitoring, an anomaly handling strategy for the oil temperature monitoring device is determined, thereby ensuring the reliability and accuracy of oil temperature monitoring data and range verification analysis results.

[0120] The anomaly handling strategy refers to the decision-making scheme for determining which oil temperature ranges to conduct real-time anomaly identification and analysis, including real-time identification of the entire range and real-time identification of a portion of the range; the correlation degree refers to the degree of correlation between the sub-range and different oil temperature ranges during the verification and analysis process.

[0121] If multiple sub-intervals are highly correlated with a certain oil temperature interval and there are many deviation checks, then the oil temperature interval has a significant impact on the overall monitoring reliability and real-time anomaly identification is required.

[0122] By determining the anomaly handling strategy through the deviation verification process and correlation degree, the sub-interval deviation information and oil temperature interval correlation information are transformed into the basis for anomaly handling decisions, thereby ensuring the reliability of oil temperature monitoring data and range verification results.

[0123] S41 determines the verification influence coefficient of the sub-interval based on the deviation verification process within the sub-interval; The verification impact coefficient (used in S41) is the same index as the verification impact coefficient in S3, which refers to the proportion of the number of deviation verification processes in a certain sub-interval to the total number of verification processes in that sub-interval.

[0124] If the verification impact coefficient for a certain sub-interval has been calculated, it can be directly used for subsequent mean determination.

[0125] The deviation verification process determines the verification impact coefficient, providing a quantitative basis for subsequently determining anomaly handling strategies based on the differences in mean values.

[0126] Specifically, if the average value of the verification influence coefficients of different sub-intervals is greater than the preset influence coefficient threshold, then the abnormal handling strategy of the oil temperature monitoring device is determined to be to perform real-time abnormal identification and analysis in all oil temperature intervals, determine the consistency between the oil temperature prediction model and the monitoring data of the oil temperature monitoring device, and thus achieve reliable identification and updating of suspected deviation oil temperature intervals in different oil temperature intervals.

[0127] The preset influence coefficient threshold (used in S41) is the same threshold as the preset influence coefficient threshold in S3.

[0128] If the average value of the influence coefficient of each sub-interval is greater than the preset influence coefficient threshold, it indicates that the overall deviation check has a large impact. Therefore, real-time anomaly identification and analysis are performed in all oil temperature intervals.

[0129] When the average value of the impact coefficient is high, a real-time identification strategy for the entire range is adopted. When the overall deviation has a large impact, the suspected deviation oil temperature range is updated in a timely manner through real-time identification of the entire range, thereby ensuring the reliability of monitoring.

[0130] Additionally, it is understandable that if the average value of the cross-checking influence coefficients of different sub-intervals is not greater than the preset influence coefficient threshold, proceed to step S42. If the average value of the verification impact coefficient is not greater than the preset impact coefficient threshold, it indicates that the overall deviation verification impact is controllable, and then proceed to S42 for more detailed analysis.

[0131] S42 determines the correlation weight value between the sub-interval and different oil temperature ranges based on the degree of correlation between the sub-interval and different oil temperature ranges during the verification analysis process; The correlation weight value refers to the quantified value of the degree of correlation between a certain sub-interval and a certain oil temperature interval during the verification and analysis process. The calculation method of the correlation weight value for preliminary verification needs to be clearly defined.

[0132] The correlation weight value is calculated as follows: Let the total number of verification analysis processes in a certain sub-interval be Nproc, and the number of processes whose running time ratio in a certain oil temperature interval reaches or exceeds the preset value of the running time ratio be Ncorrelation. Then: the correlation weight value between the sub-interval and the oil temperature interval = Ncorrelation / Nproc.

[0133] If a certain sub-interval has 20 verification analysis processes, and the running time of a certain oil temperature interval reaches or exceeds the preset value in 12 of them, then the correlation weight value = 12 / 20 = 0.60.

[0134] The correlation weight value is calculated by the percentage of times the oil temperature range running time meets the standard during the verification and analysis process. This quantifies the correlation between the sub-range and the oil temperature range from the perspective of the verification and analysis process, providing a quantitative basis for the refined determination of subsequent anomaly handling strategies.

[0135] S43 determines the abnormal handling strategy of the oil temperature monitoring device based on the verification influence coefficient of different sub-intervals and the correlation weight value between the sub-intervals and different oil temperature intervals.

[0136] Assuming that the verification impact coefficient and correlation weight value of each sub-interval have been determined, the anomaly handling strategy is determined based on the judgment results of S431 and S432.

[0137] By comprehensively verifying the impact coefficient and the associated weight value, the anomaly handling strategy is determined, and the anomaly handling strategy determination strategy is adaptively and hierarchically adjusted.

[0138] Specifically, the association weight value between the sub-interval and different oil temperature intervals is determined based on the proportion of verification analysis processes in which the running time of the oil temperature interval is above a preset time proportion during the verification analysis process of the sub-interval.

[0139] The calculation method for the association weight value was clarified, ensuring that the determination of the association weight value has a clear mathematical definition, thereby guaranteeing the repeatability of the subsequent anomaly handling strategy determination.

[0140] Furthermore, based on the verification influence coefficients of different sub-intervals and the correlation weight values ​​between the sub-intervals and different oil temperature ranges, the anomaly handling strategy of the oil temperature monitoring device is determined, specifically including: S431 uses the correlation weight values ​​of different sub-intervals and different oil temperature intervals to determine the sum of the correlation weight values ​​of the oil temperature interval and different sub-intervals. The sum of the correlation weight values ​​of the oil temperature interval and different sub-intervals is used as the correlation value of the oil temperature interval. It is determined whether the average value of the correlation values ​​of different oil temperature intervals is greater than the preset correlation threshold. If so, the abnormal handling strategy of the oil temperature monitoring device is determined to perform real-time abnormal identification and analysis processing in all oil temperature intervals. The consistency between the oil temperature prediction model and the monitoring data of the oil temperature monitoring device is determined, thereby realizing the reliable identification and update processing of suspected deviation oil temperature intervals in different oil temperature intervals. If not, proceed to step S432. The correlation value refers to the sum of the correlation weight values ​​between a certain oil temperature range and all sub-ranges; the preset correlation threshold refers to the critical value for judging whether the mean of the correlation value is high.

[0141] If the average correlation value of each oil temperature range is greater than the preset correlation threshold, it indicates that the overall correlation is high. In this case, real-time anomaly identification and analysis will be performed in all oil temperature ranges.

[0142] When the mean correlation value is high, a real-time identification strategy across the entire range is adopted. When the overall correlation is high, the suspected deviation range is updated in a timely manner through real-time identification across the entire range, thereby ensuring the reliability of monitoring.

[0143] S432 defines the sub-intervals with a verification influence coefficient greater than a preset influence coefficient threshold as verification influence intervals. It then determines whether the number of oil temperature intervals whose sum of correlation weight values ​​with all verification influence intervals is greater than a preset oil temperature interval number threshold. If so, the anomaly handling strategy of the oil temperature monitoring device is determined to be to perform real-time anomaly identification and analysis processing in all oil temperature intervals whose sum of correlation weight values ​​with all verification influence intervals is greater than a preset threshold, thereby determining the consistency between the oil temperature prediction model and the monitoring data of the oil temperature monitoring device, and thus achieving reliable identification and updating processing of suspected deviation oil temperature intervals in different oil temperature intervals. If not, the anomaly handling strategy of the oil temperature monitoring device is determined to be to perform real-time anomaly identification and analysis processing in all oil temperature intervals whose sum of correlation weight values ​​with all verification influence intervals is greater than a preset threshold, as well as in oil temperature intervals where there are verification influence intervals with correlation weight values ​​greater than a preset weight threshold, thereby determining the consistency between the oil temperature prediction model and the monitoring data of the oil temperature monitoring device, and thus achieving reliable identification and updating processing of suspected deviation oil temperature intervals in different oil temperature intervals.

[0144] The verification influence interval refers to the sub-interval where the verification influence coefficient exceeds the preset influence coefficient threshold; the preset threshold refers to the critical value for judging whether the sum of the correlation weight values ​​between the oil temperature interval and the verification influence interval is high; the preset oil temperature interval quantity threshold refers to the critical value for judging whether the number of highly correlated oil temperature intervals is large; the preset weight threshold refers to the critical value for judging whether a single correlation weight value is high.

[0145] If there are a large number of oil temperature ranges with correlation weight values ​​above the preset threshold and the corresponding influence ranges, then real-time anomaly identification will be performed in these oil temperature ranges; if there are not many, then real-time anomaly identification will be performed in these oil temperature ranges and in oil temperature ranges with a single high correlation weight correlation influence range.

[0146] By further differentiating the anomaly handling strategies by the number of highly correlated oil temperature ranges, the real-time identification range is dynamically adjusted according to the number of highly correlated ranges when the average correlation value is not high, thereby achieving a fine balance between anomaly identification coverage and processing efficiency.

[0147] This embodiment, through steps S41 to S43, realizes a complete process from verifying the calculation of the influence coefficient to analyzing the correlation weight value and determining the anomaly handling strategy. Its core value is reflected in three aspects: First, by verifying the mean of the influence coefficient, a real-time identification strategy across the entire range is adopted to ensure reliability when the overall deviation has a large impact. Second, by judging the mean of the correlation value, the real-time identification range is dynamically adjusted when the overall correlation degree is different. Third, by judging the number of highly correlated oil temperature ranges in a graded manner, the real-time identification range is finely selected when the correlation degree is not high, thereby achieving an adaptive balance between anomaly identification coverage and processing efficiency.

[0148] Based on the calculation results of S2, the range verification management method has been determined. The number of deviation verification processes and the total number of verification processes for the four sub-intervals Z1 to Z4 in the verification analysis are as follows: Z1: 3 deviations out of 10, check the impact coefficient = 3 / 10 = 0.30; Z2: 5 deviations out of 15, check the impact coefficient = 5 / 15 ≈ 0.33; Z3: 2 deviations out of 12, check the impact coefficient = 2 / 12 ≈ 0.17; Z4: 4 deviations out of 8, check the impact coefficient = 4 / 8 = 0.50; Step S3: Determine whether the influence coefficient of each sub-interval is greater than the preset influence coefficient threshold of 0.25.

[0149] Z1: 0.30 > 0.25, greater than. Z2: 0.33 > 0.25, greater than. Z3: 0.17 not greater than 0.25, not greater than. Z4: 0.50 > 0.25, greater than.

[0150] There are 3 subintervals greater than 0.25 (Z1, Z2, Z4).

[0151] Determine if 3 is greater than or equal to the preset value of 2 in the interval: if 3 is greater than 2, then it is determined that the monitoring data of the oil temperature monitoring device needs to be optimized and proceed to S4.

[0152] Step S4: Step S41: Check the influence coefficients of each sub-interval as described above.

[0153] The mean of the impact coefficient is calculated as (0.30 + 0.33 + 0.17 + 0.50) / 4 = 1.30 / 4 = 0.325.

[0154] Determine if the mean of 0.325 is greater than the preset influence coefficient threshold of 0.25: If 0.325 is greater than 0.25, then yes.

[0155] Based on the path that satisfies the conditions in S41, the anomaly handling strategy of the oil temperature monitoring device is determined to be to perform real-time anomaly identification and analysis in all oil temperature ranges, determine the consistency between the oil temperature prediction model and the monitoring data of the oil temperature monitoring device, and thus achieve reliable identification and updating of suspected deviation oil temperature ranges in different oil temperature ranges.

[0156] It should be noted that the oil temperature prediction model is constructed as follows: Using ambient temperature, load current, cooling system operating status, and historical oil temperature monitoring data from the transformer oil conservator's historical operating period as input features, and the oil temperature monitoring value after a preset future time period as the output label, a Long Short-Term Memory (LSTM) network is employed to construct the oil temperature prediction model. This model captures the dynamic characteristics of oil temperature changes through the temporal memory capability of LSTM, using ambient temperature and load current as external driving variables and historical oil temperature as an autoregressive variable for multivariate time-series prediction. During training, continuous time periods from historical operating data are used as training samples, and the model parameters are optimized through backpropagation. This enables the model to output a predicted oil temperature value after a preset future time period, given the current ambient temperature, load current, cooling system status, and recent oil temperature monitoring data.

[0157] The oil level prediction model based on oil temperature is constructed as follows: Oil temperature monitoring data or prediction results from the oil temperature prediction model at different times in the transformer oil conservator are used as input features, and the actual oil level verification results at the corresponding time are used as output labels. A multilayer perceptron (MLP) neural network is employed to construct the oil level prediction model. This model establishes the mapping relationship between oil temperature and oil level through the nonlinear mapping capability of the MLP. Considering the nonlinear range of thermal expansion and contraction characteristics of oil level in the transformer oil conservator with oil temperature changes, the hidden layer of the MLP uses the ReLU activation function to capture the nonlinear mapping characteristics between oil temperature and oil level. During training, historical oil temperature monitoring data and corresponding oil level verification results are used as training samples. The model parameters are optimized through the backpropagation algorithm, enabling the model to output the predicted oil level value at the corresponding time given the oil temperature monitoring data or oil temperature prediction results. The model is invoked when the range verification management method is triggered. It predicts the oil level based on oil temperature monitoring data at different times and compares the prediction results with the range verification results of the oil tank. When the deviation rate between the two exceeds the preset deviation rate threshold, it is determined to be a deviation verification process, thereby realizing the identification of deviation risks in the range verification analysis and processing results of the oil tank.

[0158] This embodiment implements a full-range verification and management method for transformer oil tank levels through the complete process from S1 to S4. S1, through hierarchical analysis of the oil level fluctuation range deviation rate and sub-range operation matching coefficient, determines the verification analysis and management strategy, achieving an adaptive balance between verification reliability and processing efficiency. S2, through hierarchical analysis of the oil temperature range consistency and deviation verification process, determines the range verification management method, achieving a refined balance between verification reliability and processing efficiency. S3, through quantitative analysis of the verification influence coefficient, determines whether monitoring data optimization is needed, feeding back verification deviation risk information into the optimization decision. S4, through hierarchical analysis of the verification influence coefficient and associated weight values, determines the anomaly handling strategy, achieving an adaptive balance between anomaly identification coverage and processing efficiency. Overall, this method constructs a complete technical closed loop from data acquisition to anomaly handling strategy determination through the coordinated linkage of oil level fluctuation analysis, verification strategy determination, graded determination of range verification management methods, optimization processing judgment, and dynamic management of anomaly handling strategies. Its core value is reflected in four aspects: First, it ensures the adaptability of the verification strategy through dual judgment of deviation rate and operational matching coefficient; second, it achieves adaptive selection of range verification management methods through graded analysis of suspected deviation oil temperature ranges and verification influence coefficients; third, it achieves accurate judgment of optimization processing through feedback analysis of verification influence coefficients; and fourth, it achieves dynamic determination of anomaly handling strategies through graded judgment of correlation weight values ​​and the number of highly correlated ranges. This ensures the reliability of oil temperature monitoring data and range verification results while reasonably controlling the resource consumption for verification and anomaly identification.

[0159] Example 2 Secondly, the present invention provides a computer system comprising: a memory and a processor connected in communication, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the above-described method for full-range verification and management of oil level in a transformer oil conservator when running the computer program.

[0160] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0161] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0162] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for full-range verification and management of oil level in a transformer oil conservator, characterized in that, Specifically, it includes: Based on the monitoring data of the transformer oil conservator, the oil level fluctuation range of the transformer oil conservator is determined. Based on the oil level fluctuation range and the running time in different sub-ranges within the oil level fluctuation range, the verification analysis and management strategy of the oil temperature monitoring device is determined. The verification analysis management strategy is used to determine the consistency between the oil temperature monitoring device and the oil temperature prediction model. Based on the consistency in different oil temperature ranges, suspected deviation oil temperature ranges are determined. The range verification management method for the transformer oil conservator is determined by combining the running time and consistency in the suspected deviation oil temperature ranges. The range verification management method is used to determine the degree of matching between the range verification result of the oil tank and the prediction result based on the oil temperature monitoring data. When it is determined that the monitoring data of the oil temperature monitoring device needs to be optimized based on the degree of matching, the process proceeds to the next step. Based on the degree of matching, the deviation verification process within different sub-intervals is determined, and the abnormal handling strategy of the oil temperature monitoring device is determined by combining the degree of correlation between different sub-intervals and different oil temperature ranges during the verification analysis process.

2. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 1, characterized in that, The monitoring data of the transformer oil conservator includes the monitoring results of the oil level in the transformer oil conservator at different time periods.

3. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 1, characterized in that, The oil level fluctuation range of the transformer oil conservator is determined based on the range corresponding to the monitoring results of the oil level in the transformer oil conservator.

4. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 1, characterized in that, The method for determining the verification and analysis management strategy for the oil temperature monitoring device is as follows: Determine the deviation rate between the endpoints of the oil level fluctuation range based on the oil level fluctuation range; The percentage of runtime of each sub-interval is determined by the runtime in different sub-intervals within the oil level fluctuation range, and this percentage is used as the runtime matching coefficient of the sub-interval. By utilizing the deviation rate between the endpoints of the oil level fluctuation range and the operating matching coefficients of different sub-ranges, the verification analysis and management strategy of the oil temperature monitoring device is determined.

5. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 4, characterized in that, If the deviation rate between the endpoints of the oil level fluctuation range is less than the preset deviation rate threshold, the verification analysis management strategy for the oil temperature monitoring device is determined to be to perform verification analysis processing of the oil temperature monitoring device according to the benchmark cycle.

6. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 4, characterized in that, By utilizing the deviation rate between the endpoints of the oil level fluctuation range and the operating matching coefficients of different sub-ranges, a verification analysis and management strategy for the oil temperature monitoring device is determined, specifically including: If the matching coefficient in different sub-intervals is greater than the preset matching coefficient threshold, then the verification analysis management strategy of the oil temperature monitoring device is determined to be to perform verification analysis processing of the oil temperature monitoring device according to the benchmark cycle in all oil temperature intervals.

7. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 1, characterized in that, The method for determining the range verification management method for the transformer oil conservator is as follows: The deviation verification process of the transformer oil conservator in different oil temperature ranges is determined by the degree of consistency in different oil temperature ranges. Oil temperature ranges where the quantity of deviation verification and processing does not meet the requirements are considered as suspected deviation oil temperature ranges. Based on the number of suspected deviation oil temperature ranges, the deviation verification and processing procedures in different suspected deviation oil temperature ranges, and the running time, the range verification management method for the transformer oil conservator is determined.

8. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 7, characterized in that, The suspected deviation oil temperature range is a sub-range where the number of deviation verification processes exceeds a preset threshold for the number of deviation processes.

9. The method for full-range verification and management of transformer oil level in a transformer oil conservator as described in claim 1, characterized in that, The method for determining the abnormal handling strategy of the oil temperature monitoring device is as follows: Based on the deviation verification process within the sub-interval, the verification impact coefficient of the sub-interval is determined; Based on the degree of correlation between the sub-interval and different oil temperature ranges during the verification analysis, the correlation weight value between the sub-interval and different oil temperature ranges is determined. Based on the verification influence coefficients of different sub-intervals and the correlation weight values ​​between the sub-intervals and different oil temperature intervals, the abnormal handling strategy of the oil temperature monitoring device is determined.

10. A computer system, comprising: A memory and processor connected by communication, and a computer program stored in the memory and capable of running on the processor, characterized in that, when the processor runs the computer program, it executes a method for full-range verification and management of oil level in a transformer oil conservator as described in any one of claims 1-9.

Images