Winter power transmission line adaptive reclosing method and system based on evidence theory fusion

Abstract

The invention discloses a winter power transmission line adaptive reclosing method and system based on evidence theory fusion, and electronic equipment, and the method comprises the steps: carrying out the calculation based on a Bayesian formula model, and obtaining the posterior probability of different subdivided fault causes under corresponding weather characteristics when a winter power transmission line has a fault; the method comprises the following steps of: selecting a plurality of fault reasons, introducing the fault reasons as weight distribution into a fault reason identification classification result output by a pre-constructed waveform image identification model, subdividing a large class of faults to obtain different fault reason identification probabilities after subdivision, and fusing a posterior probability and an identification probability based on a D-S evidence theory fusion model to obtain a fault reason identification result. And obtaining different fault cause probabilities after fusion, and dynamically adjusting the operation mode, the reclosing time and the number of times of reclosing on the basis of the different fault cause probabilities. Compared with the prior art, the method has the advantages that the fault reason identification precision is greatly improved, the reclosing system can dynamically adjust according to weather changes and fault analysis, and effective decision support is provided for power transmission line adaptive reclosing in winter weather.

Description

Technical Field

[0001] The present invention belongs to the technical field of power transmission line protection and automation, and in particular relates to a winter power transmission line adaptive reclosing method, system and electronic equipment based on evidence theory fusion. Background Art

[0002] At present, the fault recovery of transmission lines mainly relies on the automatic reclosing mechanism. When a fault occurs, the circuit breaker is automatically reset to determine whether the fault is a transient fault and perform recovery. However, the traditional reclosing method usually adopts a fixed recovery time and number of times, and does not dynamically optimize according to the specific weather and fault characteristics. Especially in winter, under cold climate conditions, the impact of severe weather such as snow, ice and strong wind on transmission lines is particularly significant. These weather factors make the causes of faults more complex and changeable. For example, de-icing jumps, pollution flashover and other faults are more common in winter, and most of them are transient faults. The traditional reclosing system does not make flexible adjustments based on the characteristics of these faults, resulting in the system often being unable to efficiently cope with their variability and instantaneity when facing winter faults. Especially when the reclosing operation is invalid for multiple times or the cause of the fault is misjudged, the traditional strategy may lead to instability of the power system, and may even cause equipment damage or large-scale power outages.

[0003] Although the existing reclosing systems have introduced weather data and fault analysis to a certain extent, most systems still rely on static diagnostic methods and fail to make dynamic adjustments based on real-time weather changes, line status, and fault cause analysis. This static analysis method makes it impossible for the system to flexibly adapt to the fault characteristics under different weather conditions, especially in extreme winter climates, and cannot effectively deal with complex faults such as wind deviation, wildfires, and foreign objects. Therefore, there is an urgent need for an adaptive reclosing method for transmission lines that adapts to winter weather conditions. This strategy should be able to dynamically adjust the delay, number, and operation mode of the reclosing according to real-time weather changes and fault analysis, so as to quickly recover from transient faults, and take locking reclosing measures for permanent faults to ensure the stability and safety of the power system. Summary of the invention

[0004] Based on the above technical problems, the purpose of the present invention is to provide a winter transmission line adaptive reclosing method, system and electronic equipment based on evidence theory fusion, which introduces the posterior probability obtained by calculating the Bayesian formula model as a weight into the fault cause identification and classification result output by the waveform image recognition model, subdivides the major fault categories, and fuses the subdivided different fault cause identification probabilities with the posterior probability obtained by calculating the Bayesian formula model based on the DS evidence theory fusion model, and uses the fused fault cause probability as the decision basis for reclosing, dynamically adjusts the delay, number and operation mode of the reclosing, so that transient faults can be quickly restored, and takes locking reclosing measures for permanent faults, thereby ensuring the stability and safety of the entire power system.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] The first aspect of the present invention discloses an adaptive reclosing method for winter power transmission lines based on evidence theory fusion, comprising the following steps: S1, obtaining corresponding weather characteristics when a winter power transmission line fails; S2, based on the corresponding weather characteristics obtained, and based on a pre-built Bayesian formula model, calculating and obtaining the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics; S3, inputting the electrical quantity waveform of the winter power transmission line of the fault to be identified into a pre-built waveform image recognition model to obtain the fault cause identification classification result; S4, judging whether the fault cause identification classification result in step S3 has a large category of faults, and if so, introducing the fault cause identification classification result in step S2. The obtained posterior probabilities of occurrence of different subdivided fault causes are used as allocation weights to further subdivide the fault cause identification and classification results in step S3 to obtain the subdivided identification probabilities of different fault causes; S5, based on the DS evidence theory fusion model, the posterior probabilities of occurrence of different subdivided fault causes obtained based on the Bayesian formula model in step S2 and the different fault cause identification probabilities based on the waveform graphic recognition model and subdivided based on the posterior probability in step S4 are fused and processed to output the fused probabilities of different fault causes; step S6, based on the different fault cause probabilities after fusion of weather characteristics outputted in step S5, dynamically adjust the operation mode, reclosing time and number of reclosing switches.

[0007] In the above-mentioned disclosed steps, the present invention introduces the posterior probability obtained by calculation based on the Bayesian formula model as a weight into the fault cause identification classification result output by the waveform image recognition model, subdivides the large category of faults, and fuses the subdivided different fault cause identification probabilities with the posterior probability obtained by calculation based on the Bayesian formula model based on the DS evidence theory fusion model, and uses the fused fault cause probability as the decision basis for reclosing, which effectively solves the problem that the existing waveform image recognition model is limited in identifying fault causes with extremely similar features and is difficult to quickly and accurately guide reclosing decisions.

[0008] In addition, weather characteristic information is introduced as non-electrical quantity features into the waveform image recognition model, and then effectively integrated with the DS evidence theory, which improves the accuracy of fault cause identification. At the same time, it enables the reclosing system to adapt to the fault characteristics under different weather conditions, and dynamically adjust according to real-time weather changes and fault analysis, providing effective decision support for adaptive reclosing of transmission lines in winter weather.

[0009] In addition, the present invention can quickly recover from transient faults by dynamically adjusting the delay, number of times and operation mode of the reclosing switch, and takes locking reclosing measures for permanent faults, thereby ensuring the stability and safety of the entire power system.

[0010] In the present invention, in step S1, the corresponding weather characteristics of the transmission line failure in winter are mainly derived from the weather records in the fault line patrol report, which are usually based on the real-time data of the meteorological monitoring station in the fault line section or the weather forecast information issued by the local meteorological bureau. The meteorological monitoring station provides specific real-time weather data, such as temperature, humidity, snowfall, etc., and the forecast information of the meteorological bureau is used to supplement the weather conditions in the area.

[0011] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, before step S1, it also includes step S0, pre-building a Bayesian formula model that can calculate the posterior probability of different subdivided fault causes based on known specific weather characteristics; it includes the following steps:

[0012] Construct a historical fault sample library for winter power transmission lines. The data in the historical fault sample library includes the time, location, type, detailed fault cause and historical weather characteristics of each historical fault period.

[0013] The fault cause of the fault event of the transmission line in winter is taken as variable A, and the weather characteristics corresponding to the fault event is taken as variable B. The frequency of occurrence of each fault cause in all fault events is counted to obtain the prior probability P(A i );

[0014] For each fault cause, the frequency of weather characteristics that occur when the fault cause occurs is counted to obtain the conditional probability P(B|A i );

[0015] Based on the prior probability P(A i ) and the conditional probability P(B|A i ) constructs the calculation of the posterior probability P(A i |B) Bayesian formula model:

[0016]

[0017] Among them, ∑ j P(B|A j )P(A j )=P(B), which represents the total probability of variable B occurring.

[0018] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, in the historical sample database, each subdivided fault cause is assigned a unique label, including wildfire, wind deviation, icing-ice shedding jump, lightning strike, foreign matter, pollution flashover-bird droppings, icing-insulator shed bridging, pollution flashover-other pollution flashover; and the weather characteristics are standardized to include sunny days, cloudy days, thunderstorms, strong winds, severe convective winds, typhoons, heavy rain, light rain, heavy snow, light snow, heavy fog and dust.

[0019] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, in step S4, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 are introduced as allocation weights, and the steps of further subdividing the fault cause identification and classification results in step S3 at least include:

[0020] Normalize the posterior probabilities of windage fault and icing-deicing jump fault in the fault causes subdivided in step S2, and use them as weight allocation factors to subdivide the windage and icing-deicing jump fault probabilities output by the waveform image recognition model;

[0021] The posterior probabilities of the flashover-guano fault, icing-insulator shed bridging fault and flashover-other flashover faults in the fault causes subdivided in step S2 are normalized and used as weight allocation factors to subdivide the probability of the major flashover faults output by the waveform image recognition model.

[0022] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, in step S5, fusion processing is performed based on the DS evidence theory fusion model, specifically including:

[0023] The identification probability of various fault causes of the segmented waveform image recognition model is used as the first independent evidence;

[0024] The posterior probability of each type of fault cause under different weather conditions based on the output of the Bayesian formula is used as the second independent evidence;

[0025] The basic distribution probability value of the two pieces of evidence is calculated based on the Dempster combination rule, and the prediction probability of each type of fault cause after fusion is finally output.

[0026] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, in step S6, the step of dynamically adjusting the operation mode of the reclosing includes:

[0027] Perform reclosing operations based on the predicted probabilities of different fault causes;

[0028] If reclosing fails, forced power supply or locked reclosing operation will be performed according to the corresponding criteria of different fault causes.

[0029] In the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention, in step S6, the step of performing forced power transmission or locking reclosing operation according to the criteria corresponding to different fault causes includes:

[0030] If the fault cause is identified as a de-icing jump fault, the reclosing is performed with a delay of Δt. Δt is determined according to the bounce recovery time of the de-icing jump, and the statistical value is 3-6s.

[0031] If the fault cause is identified as a dirty flashover fault, reclosing is performed normally;

[0032] If the fault cause is identified as a wildfire fault, check whether the wildfire is extinguished. If it is extinguished, perform forced power transmission. If it is not extinguished, lock the reclosing switch.

[0033] If the fault cause is identified as a foreign object fault, check whether the foreign object has been removed. If it has been removed, perform a forced power supply operation. If it has not been removed, lock the reclosing switch.

[0034] If the cause of the fault is identified as a wind deviation fault, the adaptive reclosing system needs to make another reclosing attempt based on the observed wind speed data from the nearby automatic weather station. Only when the wind speed drops below the design wind speed value of the corresponding line can the forced power supply operation be performed, otherwise the reclosing system will be locked.

[0035] The second aspect of the present invention discloses a winter power transmission line adaptive reclosing system based on evidence theory fusion, comprising:

[0036] A weather feature acquisition unit, configured to acquire corresponding weather features when a power transmission line fails in winter;

[0037] The fault cause posterior probability acquisition unit is constructed to calculate and obtain the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics according to the acquired corresponding weather characteristics and based on the pre-built Bayesian formula model;

[0038] The fault cause graphic identification unit is constructed to input the electrical quantity waveform of the winter power transmission line with a fault to be identified into a pre-built waveform image recognition model to obtain a fault cause identification classification result;

[0039] The fault cause subdivision processing unit is configured to determine whether the fault cause identification and classification result has a large category of faults. If it is determined to be yes, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 are introduced as allocation weights, and the fault cause identification and classification result in step S3 is further subdivided to obtain subdivided different fault cause identification probabilities;

[0040] A fusion processing unit is constructed to fuse the posterior probabilities of different subdivided fault causes obtained based on the Bayesian formula model and the different fault cause identification probabilities obtained based on the waveform pattern recognition model and subdivided based on the posterior probability based on the DS evidence theory fusion model, and output the fused probabilities of different fault causes;

[0041] The reclosing unit is constructed to dynamically adjust the operation mode, reclosing time and number of reclosing according to the probabilities of different fault causes after integrating the weather characteristics output by the fusion processing module.

[0042] In the winter transmission line adaptive reclosing system based on evidence theory fusion disclosed in the second aspect of the present invention, it also includes a Bayesian formula model construction unit, which is constructed to pre-build a Bayesian formula model that can calculate the posterior probability of occurrence of different subdivided fault causes based on known specific weather characteristics.

[0043] The third aspect of the present invention discloses an electronic device, including a memory, a processor and a bus, wherein the memory stores a computer program executable by the processor. When the electronic device is running, the memory and the processor communicate through the bus, and the processor executes the computer program to execute the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention.

[0044] The beneficial effects of the present invention are as follows:

[0045] The present invention introduces the posterior probability calculated based on the Bayesian formula model as a weight into the fault cause identification classification result output by the waveform image recognition model, subdivides the major fault categories, and fuses the subdivided different fault cause identification probabilities with the posterior probability calculated based on the Bayesian formula model based on the DS evidence theory fusion model, and uses the fused fault cause probability as the decision basis for the reclosing operation, which greatly improves the accuracy and robustness of fault cause identification compared with the prior art; and introduces weather feature information as a non-electrical quantity feature into the waveform image recognition model, and then effectively fuses it in combination with the DS evidence theory, so that the reclosing system can adapt to the fault characteristics under different weather conditions, and dynamically adjust according to real-time weather changes and fault analysis, providing effective decision support for the adaptive reclosing of winter transmission lines; in addition, the present invention can quickly recover from transient faults by dynamically adjusting the delay, number and operation mode of the reclosing, and take locking reclosing measures for permanent faults, thereby ensuring the stability and safety of the entire power system.

[0046] The following discloses in detail the winter transmission line adaptive reclosing method, system and electronic equipment based on evidence theory fusion of the present invention in conjunction with the embodiments shown in the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a step diagram of the winter transmission line adaptive reclosing method based on evidence theory fusion of the present invention;

[0048] Figure 2 It is a fusion flow chart based on DS evidence theory when the present invention is implemented;

[0049] Figure 3 It is an adaptive reclosing execution decision diagram when the present invention is specifically implemented;

[0050] Figure 4 The present invention is a flow chart of the winter transmission line adaptive reclosing system based on evidence theory fusion. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict.

[0052] like Figure 1As shown, the present invention discloses a winter transmission line adaptive reclosing method based on evidence theory fusion, comprising the following steps: S1, obtaining the corresponding weather characteristics when a winter transmission line fails; S2, according to the corresponding weather characteristics obtained, based on a pre-built Bayesian formula model, calculating and obtaining the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics; S3, inputting the electrical quantity waveform of the winter transmission line of the fault to be identified into a pre-built waveform image recognition model to obtain the fault cause identification classification result; S4, judging whether the fault cause identification classification result in step S3 has a large category of faults, and if so, introducing the fault cause identification classification result in step S2. The obtained posterior probabilities of the occurrence of different subdivided fault causes are used as allocation weights to further subdivide the fault cause identification and classification results in step S3 to obtain the subdivided identification probabilities of different fault causes; S5, based on the DS evidence theory fusion model, the posterior probabilities of the occurrence of different subdivided fault causes obtained based on the Bayesian formula model in step S2 and the different fault cause identification probabilities based on the waveform graphic recognition model and subdivided based on the posterior probability in step S4 are fused and processed to output the fused probabilities of different fault causes; S6, based on the different fault cause probabilities after integrating the weather characteristics output in step S5, the operation mode, reclosing time and number of reclosing switches are dynamically adjusted.

[0053] In the present invention, the posterior probability calculated based on the Bayesian formula model is introduced as a weight into the fault cause identification classification result output by the waveform image recognition model, the major faults are subdivided, and the subdivided different fault cause identification probabilities are fused with the posterior probability calculated based on the Bayesian formula model based on the DS evidence theory fusion model, and the fused fault cause probability is used as the decision basis for the reclosing operation, which greatly improves the accuracy and robustness of the fault cause identification compared with the prior art; and the weather characteristic information is introduced into the waveform image recognition model as a non-electrical quantity feature, and then effectively fused in combination with the DS evidence theory, so that the reclosing system can adapt to the fault characteristics under different weather conditions, and dynamically adjust according to real-time weather changes and fault analysis, providing effective decision support for the adaptive reclosing of winter transmission lines; and the present invention can quickly recover from transient faults by dynamically adjusting the delay, number and operation mode of the reclosing, and take locking reclosing measures for permanent faults, thereby ensuring the stability and safety of the entire power system.

[0054] In a preferred embodiment, before step S1, step S0 is further included to pre-build a Bayesian formula model that can calculate the posterior probability of occurrence of different subdivided fault causes based on known specific weather characteristics, specifically in the following manner:

[0055] 1) Constructing a winter transmission line fault sample library

[0056] First, collect and organize data on historical winter transmission line fault events. The sample library should ensure that the data includes the time, location, type, cause of the fault, and detailed weather characteristics of each fault event period. The data sources mainly include event records in fault patrol reports, real-time weather data provided by meteorological monitoring stations, and weather forecast information released by local meteorological bureaus. Through these sources, ensure that fault events and corresponding weather characteristics can be accurately matched.

[0057] After the data is collected, it is cleaned and preprocessed. Outliers and missing values ​​in the sample library are removed or filled to ensure the integrity and accuracy of the data. In addition, to ensure data consistency, the data needs to be standardized so that different weather characteristics and fault types can be uniformly measured and compared in subsequent analysis.

[0058] The fault causes are classified in detail, including: wildfire, wind deviation, ice-de-ice jumping, lightning strike, foreign matter, pollution flashover-bird droppings, ice-insulator shed bridging, pollution flashover-other pollution flashover, etc. Each fault cause is given a unique label to ensure the clarity and accuracy of the classification, which is convenient for subsequent analysis.

[0059] At the same time, the collected weather characteristics also need to be standardized, and different meteorological monitoring data are uniformly classified to facilitate matching and analysis with the cause of the fault. Specifically, the standardized weather characteristics are divided into N categories, including: sunny, cloudy, thunderstorm, strong wind, strong convective wind, typhoon, heavy rain, light rain, heavy snow, light snow, heavy fog and dust, etc. After each type of weather characteristics is standardized, it can accurately reflect the specific meteorological conditions when the fault occurs, and provide reliable input data for subsequent Bayesian statistical analysis.

[0060] Based on the prior probability P(A i ) and the conditional probability P(B|A i ) constructs the calculation of the posterior probability P(A i |B) Bayesian formula model:

[0061]

[0062] Among them, ∑ j P(B|A j )P(A j )=P(B), which represents the total probability of variable B occurring.

[0063] 2) Obtain the prior distribution and sample conditional probability of the sample library data

[0064] Set variables A and B, A represents the fault cause of the fault event of the transmission line in winter, and B represents the weather characteristics corresponding to the fault event;

[0065] The prior distribution reflects the basic frequency of different fault causes in all fault events. By counting the historical fault sample library, the frequency of each fault cause in all fault events is counted to obtain the prior probability P(A i );

[0066] The conditional probability reflects the correlation between the fault cause and the weather conditions. For each fault cause, the frequency of weather characteristics that occur when the fault cause occurs is counted to obtain the conditional probability P(B|A i ).

[0067] 3) Calculate the posterior probability based on the Bayesian formula

[0068] Furthermore, the posterior distribution of different fault causes under different weather characteristics is determined, and the posterior probability P(A i |B), the calculation formula is:

[0069]

[0070] Among them, ∑ j P(B|A j )P(A j )=P(B), which represents the total probability of variable B occurring.

[0071] This formula allows the calculation of the i-th fault cause A when a specific weather feature B is known. i probability of occurrence.

[0072] In the embodiment of the present invention, in step S4, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 are introduced as allocation weights, and the steps of further subdividing the fault cause identification and classification results in step S3 at least include:

[0073] Normalize the posterior probabilities of windage fault and icing-deicing jump fault in the fault causes subdivided in step S2, and use them as weight allocation factors to subdivide the windage and icing-deicing jump fault probabilities output by the waveform image recognition model;

[0074] The posterior probabilities of the flashover-guano fault, icing-insulator shed bridging fault and flashover-other flashover faults in the fault causes subdivided in step S2 are normalized and used as weight allocation factors to subdivide the probability of the major flashover faults output by the waveform image recognition model.

[0075] like Figure 2 As shown, in the embodiment of the present invention, in step S5, fusion processing is performed based on the DS evidence theory fusion model, and the specific method is as follows:

[0076] Firstly, the electrical quantity waveform of the fault case to be identified is input into the pre-trained waveform image recognition model to obtain the preliminary fault cause classification result;

[0077] Furthermore, the fault cause identification results of the waveform image recognition model are further subdivided. The posterior probability calculated in step S2 is introduced, and the posterior probability of wind deviation and icing-deicing jump faults is normalized as a weight factor to further refine the wind deviation and icing-deicing jump fault category probabilities output by the waveform recognition model;

[0078] The posterior probabilities of pollution flashover-guano, icing-insulator shed bridging and pollution flashover-other pollution flashover faults are normalized and used as weight factors to further refine the probability of major categories of pollution flashover faults output by the waveform recognition model.

[0079] Construct a DS evidence theory fusion model to fuse the identification probabilities of various fault causes of the segmented waveform image recognition model and the posterior probabilities of various fault causes based on Bayesian formula statistics. The specific construction methods include:

[0080] 1) Establish an evidence theory identification framework Θ. In this embodiment, the identification framework is a set of winter transmission line fault causes, and the winter transmission line fault causes can be divided into M categories in total:

[0081] Θ={F1,F2,…,F m}

[0082] Where m is the category number of the fault cause, F m Indicates the cause of the mth type of transmission line fault.

[0083] 2) Determine the basic probability distribution function m i (F j ), calculated by the following formula:

[0084]

[0085] In the formula, m i (F j ) is the basic probability distribution value of the i-th evidence of the j-th fault cause, Y ij is the probability of the i-th evidence for the j-th fault cause.

[0086] 3) Determine the conflict coefficient k of independent evidence m1 and m2, calculated by the following formula:

[0087]

[0088] In the formula, m1(A i ) is the predicted probability of the i-th fault cause output by the waveform image recognition model, m2(Bj ) is the posterior probability of the jth fault cause output by the Bayesian statistical model.

[0089] 4) Based on the Dempster combination rule, the basic probability distribution values ​​of the two models are integrated. The combination rule formula is:

[0090]

[0091] In the formula, represents the basic probability distribution value after fusion, and A is the proposition A i and Proposition B j In this embodiment, A is the same fault cause output by the two models.

[0092] The present invention can effectively fuse the recognition results of the waveform image recognition model and the posterior probability results of the Bayesian statistical model through the DS evidence theory, thereby improving the accuracy and robustness of fault cause identification.

[0093] like Figure 3 As shown, this embodiment shows an adaptive reclosing execution decision diagram. The decision diagram dynamically adjusts the reclosing operation strategy based on the integrated weather information and the corrected predicted probabilities of various fault causes. The adaptive reclosing method is mainly optimized for transient faults, while for permanent faults, the reclosing will be locked, and forced power will be sent after the fault cause is confirmed. The specific methods include:

[0094] 1) Determine the cause of the transmission circuit failure

[0095] According to the modified DS evidence fusion results, the predicted probabilities of various fault causes are sorted from high to low, and the fault cause with the highest probability is selected as the final identification result.

[0096] 2) Adaptive reclosing method

[0097] According to different fault causes, corresponding adaptive reclosing methods are adopted, including:

[0098] De-icing jump fault: Since this type of fault has a short bounce recovery time, it is recommended to perform the reclosing operation after a delay of Δt=3-6s.

[0099] Pollution flashover fault: This type of fault is usually a transient fault, so reclosing can be performed normally.

[0100] Lightning fault: Lightning fault is usually also a transient fault, and reclosing can be performed normally.

[0101] Wildfire fault: First perform the reclosing operation. If the reclosing fails, check whether the wildfire in the faulty line section has been extinguished:

[0102] If the wildfire has been extinguished, the forced power supply operation will be carried out;

[0103] If the wildfire has not been extinguished, the reclosing switch will be locked and power supply will be restored only after the wildfire is extinguished.

[0104] Foreign object fault: First perform the reclosing operation. If the reclosing fails, check whether the foreign object in the faulty line section has been removed:

[0105] If the foreign object has been removed, perform a forced power supply operation;

[0106] If the foreign object is not removed, the reclosing switch will be locked and power will be restored after the foreign object is removed.

[0107] Wind deviation fault: First perform the reclosing operation. If the reclosing fails, obtain real-time wind speed data from the meteorological station near the faulty line section:

[0108] If the wind speed is less than the design wind speed, the forced power transmission operation will be performed;

[0109] If the wind speed is greater than the design wind speed, the reclosing switch will be locked and power supply will be restored after the wind speed decreases.

[0110] Through this adaptive reclosing method, the reclosing operation mode can be dynamically adjusted to ensure that transient faults are restored in time, while safer handling measures are taken for permanent faults, thereby optimizing the operational reliability of the power system.

[0111] This embodiment analyzes some transmission line failure cases in a western region in the past year to make reclosing decisions. The specific implementation cases are as follows:

[0112] Failure Case 1

[0113] At 01:37 on January 1, 2024, the 220 kV MH East Line failed in section 041 to 042. On the day of the failure, there was continuous fog and the temperature was below zero.

[0114] The DS evidence theory fusion model was used to identify the cause of the fault, and it was confirmed that the cause of the fault was ice-de-icing jump. According to the fault identification results, the reclosing method was optimized and adjusted. Considering that it takes a certain amount of time to recover after de-icing bounce, a 5-second delay reclosing method was adopted, that is, reclosing was attempted 5 seconds after the fault was isolated. The purpose of this strategy is to reserve sufficient time for the conductor to recover from bounce and ice to fall off, thereby reducing the frequency of tripping and reclosing and improving power supply reliability.

[0115] The operational results after implementation showed that this 5-second delay reclosing method effectively avoided the problem of multiple tripping and achieved a reasonable reclosing delay. This measure successfully restored the line to normal operation and ensured power supply reliability.

[0116] Failure Case 2

[0117] At 17:49 on February 16, 2024, a fault occurred on the B-phase conductor of the 110 kV TD line pole 33. The weather when the fault occurred was windy, and the maximum instantaneous wind speed during the fault period reached 25.6 m / s.

[0118] The DS evidence theory fusion model was used to identify the cause of the fault and confirmed that the cause of the fault was wind deviation. Considering that the high wind speed may continue or fluctuate, and that multiple trips may occur during the wind weakening process, the reclosing method was optimized and adjusted. Given that the fault occurred when the wind speed exceeded the design wind speed of 23.5m / s, it was finally decided to use the locked reclosing method.

[0119] The operational results after implementation show that the locking reclosing method effectively avoids the problem of multiple tripping caused by strong winds.

[0120] Failure Case 3

[0121] At 14:19 on February 21, 2024, a fault occurred on the 188th tower of the 220 kV YB line. According to the subsequent on-site inspection report, when the fault occurred, the snow on the surface of the A-phase insulator (snow continued to fall from the 18th to the early morning of the 21st) melted and mixed with the dust on the surface to form icicles. As the temperature gradually rose and the weather cleared during the day on the 21st, icicles accumulated on the surface of the insulator, resulting in the formation of a conductive channel at the insulator shed, which eventually triggered a flashover and caused the line to trip.

[0122] The DS evidence theory fusion model was used to identify the cause of the fault, and it was confirmed that the cause of the fault was ice-insulator shed bridge. According to the fault identification results, the reclosing method was optimized and adjusted. Since the arc during the dirty flashover process generated high temperature, it quickly heated and melted the ice channel, thereby restoring the electrical insulation performance of the insulator. This type of fault is an instantaneous event. Considering that the insulator restored normal electrical insulation after the ice melted and the arc subsided, it was decided to operate according to the normal reclosing method, that is, to perform the reclosing operation immediately.

[0123] The operational results after implementation showed that the rapid execution of the reclosing method successfully restored power supply and ensured power supply continuity.

[0124] like Figure 4 As shown, the present invention also discloses a winter power transmission line adaptive reclosing system based on evidence theory fusion, comprising: a weather feature acquisition unit, configured to acquire corresponding weather features when a winter power transmission line fails;

[0125] The fault cause posterior probability acquisition unit is constructed to calculate and obtain the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics according to the acquired corresponding weather characteristics and based on the pre-built Bayesian formula model;

[0126] The fault cause graphic identification unit is constructed to input the electrical quantity waveform of the winter power transmission line with a fault to be identified into a pre-built waveform image recognition model to obtain a fault cause identification classification result;

[0127] The fault cause subdivision processing unit is configured to determine whether the fault cause identification and classification result has a large category of faults. If it is determined to be yes, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 are introduced as allocation weights, and the fault cause identification and classification result in step S3 is further subdivided to obtain subdivided different fault cause identification probabilities;

[0128] A fusion processing unit is constructed to fuse the posterior probabilities of different subdivided fault causes obtained based on the Bayesian formula model and the different fault cause identification probabilities obtained based on the waveform pattern recognition model and subdivided based on the posterior probability based on the DS evidence theory fusion model, and output the fused probabilities of different fault causes;

[0129] The reclosing unit is constructed to dynamically adjust the operation mode, reclosing time and number of reclosing according to the probabilities of different fault causes after integrating the weather characteristics output by the fusion processing module.

[0130] In the embodiment of the present invention, a Bayesian formula model construction unit is also included, which is constructed to pre-construct a Bayesian formula model that can calculate the posterior probabilities of occurrence of different segmented fault causes based on known specific weather characteristics.

[0131] The present invention also discloses an electronic device, including a memory, a processor and a bus, wherein the memory stores a computer program executable by the processor. When the electronic device is running, the memory and the processor communicate through the bus, and the processor executes the computer program to execute the winter transmission line adaptive reclosing method based on evidence theory fusion disclosed in the first aspect of the present invention.

[0132] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0133] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0134] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0135] The above description is only an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0136] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0137] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A winter transmission line adaptive reclosing method based on evidence theory fusion, characterized in that: The steps include: S1, obtain the corresponding weather characteristics when the transmission line fails in winter; S2, according to the corresponding weather characteristics obtained, based on the pre-built Bayesian formula model, calculate and obtain the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics; S3, inputting the electrical quantity waveform of the winter transmission line with a fault to be identified into a pre-built waveform image recognition model to obtain a fault cause identification and classification result; S4, judging whether the fault cause identification and classification result in step S3 has a large category of faults. If it is judged to be yes, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S2 are introduced as allocation weights, and the fault cause identification and classification result in step S3 is further subdivided to obtain the subdivided identification probabilities of different fault causes; S5, based on the DS evidence theory fusion model, the posterior probabilities of the occurrence of different subdivided fault causes obtained based on the Bayesian formula model in step S2 and the different fault cause identification probabilities obtained based on the waveform pattern recognition model and subdivided based on the posterior probabilities in step S4 are fused and processed, and the fused probabilities of different fault causes are output; S6, dynamically adjusting the operation mode, reclosing time and number of times of the reclosing switch based on the probabilities of different fault causes after integrating the weather characteristics output in step S5.

2. The winter power transmission line adaptive reclosing method based on evidence theory fusion according to claim 1 is characterized in that: Before step S1, step S0 is also included, which is to pre-build a Bayesian formula model that can calculate the posterior probability of occurrence of different subdivided fault causes based on known specific weather characteristics; It includes the following steps: Constructing a historical fault sample library of winter power transmission lines, wherein the data in the historical fault sample library includes the time, location, type, detailed fault cause and historical weather characteristics of each historical fault occurrence period; The fault cause of the fault event of the transmission line in winter is taken as variable A, and the weather characteristics corresponding to the fault event is taken as variable B. The frequency of occurrence of each fault cause in all fault events is counted to obtain the prior probability P(A i ); For each fault cause, the frequency of weather characteristics that occur when the fault cause occurs is counted to obtain the conditional probability P(B|A i ); Based on the prior probability P(A i ) and the conditional probability P(B|A i ) Construct and calculate the posterior probability P(A i |B) Bayesian formula model: Among them, ∑ j P(B|A j )P(A j )=P(B), which represents the total probability of variable B occurring.

3. The winter power transmission line adaptive reclosing method based on evidence theory fusion according to claim 2 is characterized in that: In the historical sample database, each detailed fault cause is assigned a unique label, including wildfire, wind deviation, icing-deicing jumping, lightning strike, foreign object, pollution flashover-bird droppings, icing-insulator shed bridging, pollution flashover-other pollution flashover; and the weather characteristics are standardized to include sunny, cloudy, thunderstorm, strong wind, severe convective wind, typhoon, heavy rain, light rain, heavy snow, light snow, heavy fog and dust.

4. The winter power transmission line adaptive reclosing method based on evidence theory fusion according to claim 3 is characterized in that: In step S4, the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 are introduced as allocation weights to further subdivide the fault cause identification and classification results in step S3, which at least includes: Normalize the posterior probabilities of windage fault and icing-deicing jump fault in the fault causes subdivided in step S2, and use them as weight allocation factors to subdivide the windage and icing-deicing jump fault probabilities output by the waveform image recognition model; The posterior probabilities of the flashover-guano fault, icing-insulator shed bridging fault and flashover-other flashover faults in the fault causes subdivided in step S2 are normalized and used as weight allocation factors to subdivide the probability of the major flashover faults output by the waveform image recognition model.

5. The winter transmission line adaptive reclosing method based on evidence theory fusion according to claim 4 is characterized in that: In step S5, fusion processing is performed based on the DS evidence theory fusion model, specifically including: The identification probability of various fault causes of the segmented waveform image recognition model is used as the first independent evidence; The posterior probability of each type of fault cause under different weather conditions based on the output of the Bayesian formula is used as the second independent evidence; The basic distribution probability value of the two pieces of evidence is calculated based on the Dempster combination rule, and the prediction probability of each type of fault cause after fusion is finally output.

6. The winter transmission line adaptive reclosing method based on evidence theory fusion according to claim 5 is characterized in that: In step S6, the step of dynamically adjusting the operation mode of the reclosing circuit breaker includes: Perform reclosing operations based on the predicted probabilities of different fault causes; If reclosing fails, forced power supply or locked reclosing operation will be performed according to the corresponding criteria of different fault causes.

7. The winter transmission line adaptive reclosing method based on evidence theory fusion according to claim 6 is characterized in that: In step S6, the steps of performing forced power transmission or locking reclosing operation according to the judgment criteria corresponding to different fault causes include: If the fault cause is identified as a de-icing jump fault, the reclosing is performed with a delay of Δt. Δt is determined according to the bounce recovery time of the de-icing jump, and the statistical value is 3-6s. If the fault cause is identified as a dirty flashover fault, reclosing is performed normally; If the fault cause is identified as a wildfire fault, check whether the wildfire is extinguished. If it is extinguished, perform forced power transmission. If it is not extinguished, lock the reclosing switch. If the fault cause is identified as a foreign object fault, check whether the foreign object has been removed. If it has been removed, perform a forced power supply operation. If it has not been removed, lock the reclosing switch. If the cause of the fault is identified as a wind deviation fault, the adaptive reclosing system needs to make another reclosing attempt based on the observed wind speed data from the nearby automatic weather station. Only when the wind speed drops below the design wind speed value of the corresponding line can the forced power supply operation be performed, otherwise the reclosing system will be locked.

8. A winter transmission line adaptive reclosing system based on evidence theory fusion, characterized in that: include: A weather feature acquisition unit, configured to acquire corresponding weather features when a power transmission line fails in winter; The fault cause posterior probability acquisition unit is constructed to calculate and obtain the posterior probability of occurrence of different subdivided fault causes based on the corresponding weather characteristics according to the acquired corresponding weather characteristics and based on the pre-built Bayesian formula model; The fault cause graphic identification unit is constructed to input the electrical quantity waveform of the winter power transmission line with a fault to be identified into a pre-built waveform image recognition model to obtain a fault cause identification classification result; a fault cause subdivision processing unit, configured to determine whether the fault cause identification and classification result has a large category of faults, and if so, introduce the posterior probabilities of occurrence of different subdivided fault causes obtained in step S3 as allocation weights, and further subdivide the fault cause identification and classification result in step S3 to obtain subdivided different fault cause identification probabilities; A fusion processing unit is constructed to fuse the posterior probabilities of occurrence of different subdivided fault causes obtained based on the Bayesian formula model and the different fault cause identification probabilities obtained based on the waveform pattern recognition model and subdivided based on the posterior probabilities based on the DS evidence theory fusion model, and output the fused probabilities of different fault causes; The reclosing unit is constructed to dynamically adjust the operation mode, reclosing time and number of reclosing according to the probabilities of different fault causes after integrating the weather characteristics output by the fusion processing module.

9. The winter power transmission line adaptive reclosing system based on evidence theory fusion according to claim 8 is characterized in that: It also includes a Bayesian formula model building unit, which is constructed to pre-build a Bayesian formula model that can calculate the posterior probability of occurrence of different segmented fault causes based on known specific weather characteristics.

10. An electronic device, characterized in that: It includes a memory, a processor and a bus, the memory stores a computer program executable by the processor, when the electronic device is running, the memory and the processor communicate through the bus, and the processor executes the computer program to execute the winter transmission line adaptive reclosing method based on evidence theory fusion as described in any one of claims 1-7.

Images