Ultra-high frequency partial discharge stability test method and device for hydro-generator

By sampling the ultra-high frequency local discharge signals of the hydro turbine generator extreme point and counting amplitude, combined with cluster analysis of local discharge events, the problem of high fluctuation in the test results is solved, and the accuracy of the insulation state of the stator winding is achieved. Identification and stability improvement.

CN118465475BActive Publication Date: 2024-11-15GUODIAN SCI & TECH RES INST +5
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410912445.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-09
Publication Date
2024-11-15
Estimated Expiration
2044-07-09

AI Technical Summary

Technical Problem

The existing ultra-high frequency partial discharge testing technology has problems such as high data fluctuation and poor test results in hydrowheel generators, which leads to the insulating state mapping relationship that cannot be accurately identified and established.

Method used

By obtaining the periodic pulse sequence of the local discharge signal, performing extreme point sampling and differential operations, generating extreme value sequences, dividing short periods and long periods, calculating amplitude statistical parameters and local burst stability indicators, and combining the clustering of local discharge events Analysis and determine the insulation state of the stator winding.

Benefits of technology

It effectively reduces the volatility of the test results, improves the stability and accuracy of the test results, accurately identifys the insulation state of the stator winding, and overcomes the influence of external interference and randomness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118465475B_ABST
    Figure CN118465475B_ABST
Patent Text Reader

Abstract

The present invention provides a method and device for testing the stability of ultra-high frequency partial discharge of a hydro-generator, and belongs to the technical field of ultra-high frequency partial discharge testing. The method comprises: obtaining a periodic pulse sequence obtained after the partial discharge signal is equally dispersed; sampling the extreme points of the periodic pulse sequence within the sampling period to generate an extreme value sequence, and associating the amplitude of each extreme point in the extreme value sequence with its corresponding occurrence time value; dividing each short period within the sampling period; determining the amplitude statistical parameters of the short period according to the extreme value sequence within the short period, and determining the partial discharge stability index of the short period according to the amplitude statistical parameters; when the partial discharge stability index of the short period meets the preset requirements, the short period is determined as the target short period; the amplitude statistical parameters of each target short period within the sampling period are analyzed to determine the insulation state of the stator winding. The present invention improves the stability of the partial discharge test results.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of ultra-high frequency partial discharge measurement, and specifically relates to an ultra-high frequency partial discharge stability test method for a hydro-turbine generator, an ultra-high frequency partial discharge stability test device for a hydro-turbine generator, a machine-readable storage medium and an electronic device. Background Art

[0002] The partial discharge test of a hydro-turbine generator is used to diagnose early insulation faults. The process mainly includes: installing partial discharge online test equipment at the partial discharge test point of the stator winding, and monitoring the changes in the test data to infer the insulation state of the stator winding and identify the type of partial discharge. The key to the above process is to establish a mapping relationship between the test results and the insulation state.

[0003] At present, the mainstream partial discharge online test equipment in the market mainly adopts low-frequency partial discharge test technology, and its frequency range is 10kHz~500kHz. When applied to hydro-turbine generators, the monitored test data fluctuates greatly due to the interference signal during the operation stage of the generator, resulting in difficulties in identifying partial discharge patterns. In order to overcome the above-mentioned defects of low-frequency partial discharge test technology, many scholars have carried out research on ultra-high frequency partial discharge test technology for hydro-turbine generators, and existing research has shown that compared with low-frequency partial discharge test technology, ultra-high frequency partial discharge test technology can avoid some external interference, and the stability of test results is better than that of low-frequency partial discharge test technology. However, affected by many external interferences and the randomness of partial discharge itself, the test results of ultra-high frequency partial discharge test technology are still very volatile. For example, affected by the large capacitance characteristics of the stator winding of the hydro-turbine generator to the ground, as the capacitance increases, the signal-to-noise ratio of the partial discharge signal will be significantly reduced, which is very unfavorable for partial discharge testing. When the test data has large fluctuations, it is impossible to accurately determine the test results, and thus it is impossible to accurately establish the mapping relationship between the test results and the insulation state.

[0004] In summary, in order to improve the data volatility and enhance the stability of test results in ultra-high frequency partial discharge test technology, it is urgent to propose a partial discharge stability test method for hydro-generators. Summary of the invention

[0005] In view of this, the purpose of the embodiments of the present invention is to provide a hydro-turbine generator ultra-high frequency partial discharge stability test method, a hydro-turbine generator ultra-high frequency partial discharge stability test device, a machine-readable storage medium and an electronic device, so as to overcome the technical problems of large data volatility and poor test result stability existing in the existing ultra-high frequency partial discharge test technology applied to hydro-turbine generators.

[0006] In order to achieve the above-mentioned object, a first aspect of an embodiment of the present invention provides a method for testing the stability of ultra-high frequency partial discharge of a hydro-turbine generator, the method comprising:

[0007] Obtaining a periodic pulse sequence formed by discretizing the partial discharge signal;

[0008] Sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, and associating the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value;

[0009] Divide each short period within the sampling period;

[0010] Determine the amplitude statistical parameters of the short period according to the extreme value sequence in the short period, and determine the stability index of the partial discharge amount of the short period according to the amplitude statistical parameters;

[0011] When the partial discharge stability index of a short cycle meets the preset requirements, the short cycle is determined as the target short cycle;

[0012] The amplitude statistical parameters of each target short period within the sampling period are analyzed to determine the insulation state of the stator winding.

[0013] Optionally, the step of sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, and associating the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value, includes:

[0014] During the sampling period, the amplitude of each pulse in the periodic pulse sequence is differentially calculated;

[0015] Determine the extreme point based on the difference operation result;

[0016] An extreme value sequence is composed of all extreme value points. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value.

[0017] Optionally, determining the extreme point according to the difference operation result includes:

[0018] If the amplitude of a pulse position in the periodic pulse sequence is greater than 0, and the differential operation result for the pulse position satisfies the first determination condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point; the first determination condition is that the amplitude of the pulse position is greater than the amplitude of the pulse position adjacent to it;

[0019] If the amplitude of a pulse position in a periodic pulse sequence is less than 0, and the differential operation result of the pulse amplitude meets the second judgment condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point; the second judgment condition is that the amplitude of the pulse position is smaller than the amplitude of the pulse position adjacent to it.

[0020] Optionally, the amplitude statistical parameter includes at least one of a maximum value, a 95% value and an average value.

[0021] Optionally, the partial discharge stability index includes at least one of a partial discharge fluctuation value index and a 95% value statistical deviation index;

[0022] The partial discharge fluctuation value index represents the variation range of the maximum value compared with the average value in a short period;

[0023] The 95% value statistical deviation index represents the variation range of the maximum value of the 95% value in each periodic time period in the short cycle compared with the minimum value of the 95% value in each periodic time period.

[0024] Optionally, the method further includes:

[0025] Divide each long period within the sampling period;

[0026] Associating each extreme point in the extreme value sequence with its corresponding phase angle;

[0027] According to the amplitude and phase angle of the extreme point, the extreme value sequence in the long period is clustered to determine the local discharge events occurring in the long period;

[0028] The partial discharge events of each long period are analyzed, and the analysis results of the partial discharge events of each long period are combined with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding.

[0029] Optionally, analyzing each long-period partial discharge event and combining the analysis results of each long-period partial discharge event with the amplitude statistical parameter analysis results of each target short period in the sampling period to determine the insulation state of the stator winding includes:

[0030] Analyze the changing trend of the amplitude statistical parameters of each target short period within the sampling period;

[0031] Analyze the changing trend of the number of partial discharge events in each long period within the sampling period;

[0032] The variation trend of the amplitude statistical parameters of each target short period within the sampling period and the variation trend of the number of partial discharge events of each long period are combined and analyzed to determine the insulation state of the stator winding.

[0033] Optionally, the combining and analyzing the variation trend of the amplitude statistical parameter of each target short period and the variation trend of the number of partial discharge events of each long period within the sampling period to determine the insulation state of the stator winding includes:

[0034] If the change trend of the amplitude statistical parameters of each target short period within the sampling period is an increasing trend, and the change trend of the number of partial discharge events of each long period within the sampling period is an increasing trend, it is used to determine that an insulation fault has occurred in the stator winding.

[0035] Optionally, clustering the extreme value sequence within a long period according to the amplitude and phase angle of the extreme value point to determine the local discharge event occurring within the long period includes:

[0036] Calculate the Euclidean distance between any two extreme value points in the extreme value sequence in the two-dimensional vector space composed of amplitude and phase angle. If the Euclidean distance is less than the first threshold, it is determined that the current two extreme value points are the same event, otherwise they are not the same event.

[0037] The total number of extreme value points that have been determined to be the same event is counted, and if the total number of extreme value points exceeds a second threshold, it is determined to be a partial discharge event.

[0038] Optionally, there are multiple process object links to which the periodic pulse sequence within the sampling period belongs, the extreme value sequence includes multiple extreme value subsequences, and each extreme value subsequence is divided into each process object link one by one, and the method further includes:

[0039] If the change of process object link A causes a delay in the response of process object link B, the time distance between a certain extreme value point in the extreme value subsequence corresponding to process object link A and each extreme value point in the extreme value subsequence corresponding to process object link B is calculated respectively, and the minimum value is taken to obtain the minimum time distance. Similarly, the minimum time distances corresponding to each extreme value point in the extreme value subsequence corresponding to process object link A are obtained, and all minimum time distances form a set;

[0040] Count the number of occurrences of each minimum time interval in the set to obtain the mode of the set, and determine the mode as the time interval of the delay; when associating each extreme point in the extreme value subsequence corresponding to the process object link B with its corresponding occurrence time value, the occurrence time value is the difference between the time value of the pulse position corresponding to the extreme point and the time interval.

[0041] A second aspect of an embodiment of the present invention provides a hydro-generator ultra-high frequency partial discharge stability test device, the device comprising:

[0042] An acquisition module, used to acquire a periodic pulse sequence formed after the partial discharge signal is discretized;

[0043] An extreme value sampling module is used to sample extreme value points within a sampling period of a periodic pulse sequence to generate an extreme value sequence, and associate the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value;

[0044] A short cycle division module is used to divide each short cycle within the sampling period;

[0045] A partial discharge stability index determination module is used to determine the amplitude statistical parameters of the short period according to each extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters;

[0046] A target short cycle determination module is used to determine the short cycle as a target short cycle when the partial discharge stability index of the short cycle meets the preset requirements;

[0047] The insulation state mapping module is used to analyze the amplitude statistical parameters of each target short period within the sampling period to determine the insulation state of the stator winding.

[0048] A third aspect of an embodiment of the present invention provides a machine-readable storage medium having instructions stored thereon, which, when executed by a processor, configures the processor to execute the ultra-high frequency partial discharge stability test method for a hydro-turbine generator as described in the first aspect of the embodiment of the present invention.

[0049] A fourth aspect of an embodiment of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the ultra-high frequency partial discharge stability test method for a hydro-turbine generator as described in the first aspect of an embodiment of the present invention is implemented.

[0050] In the above technical scheme, the partial discharge signal is discretized to obtain a periodic pulse sequence. It is known that the discretization is usually equidistant discretization, and then the periodic pulse sequence is sampled at extreme points. The extreme point sampling adopts methods such as differential operation to ensure the accurate acquisition of the maximum value of the partial discharge pulse wave. Then, in order to achieve accurate mapping of the test results and the insulation state of the stator winding, it is necessary to obtain the accurate change trend of the amplitude statistical parameters within the sampling period. Based on this, the present invention also constructs a partial discharge stability index, and excludes the corresponding short-period extreme value sequences whose partial discharge value fluctuation state does not meet the requirements, that is: it is not used for the change trend analysis of the amplitude statistical parameters, thereby eliminating the influence of abnormal fluctuations, thereby overcoming the volatility problem of the test results and improving the stability of the test results.

[0051] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0053] Figure 1 A schematic flow chart of a method for testing the stability of ultra-high frequency partial discharge of a hydro-generator provided by an embodiment of the present invention;

[0054] Figure 2 Another schematic flow chart of a method for testing the stability of ultra-high frequency partial discharge of a hydro-generator provided by an embodiment of the present invention;

[0055] Figure 3 A schematic diagram of equidistant dispersion processing of partial discharge signals in an application example;

[0056] Figure 4 It is a schematic diagram of the extreme value points in the extreme value sequence in the application example;

[0057] Figure 5 A schematic diagram of the variation trend of amplitude statistical parameters in the application example;

[0058] Figure 6 A schematic diagram of the event clustering results in the application example;

[0059] Figure 7 A schematic diagram of the clustering results of partial discharge events in the application example;

[0060] Figure 8 A schematic diagram of the trend of the number of partial discharge events in the application example;

[0061] Fig. 9 A schematic diagram of the composition of an ultra-high frequency partial discharge stability testing device for a hydro-turbine generator provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. It should be understood that the specific implementation methods described herein are only used to illustrate and explain the embodiments of the present invention, and are not used to limit the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0063] The following embodiments should be read together Figures 1 to 8 .

[0064] Embodiment 1

[0065] See also Figure 1 This embodiment provides a method for testing the stability of ultra-high frequency partial discharge of a hydro-turbine generator, which specifically includes the following implementation steps:

[0066] S100. Obtaining a periodic pulse sequence formed by equidistant dispersion of a partial discharge signal;

[0067] S200. Sample the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value, that is, the extreme value sequence includes the amplitude of each extreme value point and the occurrence time value corresponding to the extreme value point. The amplitude of the extreme value point is the amplitude of its corresponding pulse position, and the occurrence time value of the extreme value point is the time value of its corresponding pulse position;

[0068] S300. Divide each short period within the sampling period;

[0069] S400. Determine the amplitude statistical parameters of the short period according to the extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters, the partial discharge stability index characterizing the stability of the partial discharge value;

[0070] S500. When the partial discharge stability index of the short cycle meets the preset requirements, the short cycle is determined as the target short cycle;

[0071] S600. Analyze the amplitude statistical parameters of each target short period within the sampling period to determine the insulation state of the stator winding of the hydro-generator.

[0072] It is known that in order to identify the insulation state of the stator winding, it is usually obtained by analyzing the change trend of the partial discharge value. Therefore, a specific implementation process of S600 is as follows:

[0073] Analyze the changing trend of the amplitude statistical parameters of each target short period within the sampling period to determine the insulation state of the stator winding.

[0074] For example: within a sampling period, if the amplitude statistical parameter change trend is a rapid increase and exceeds the set threshold, it is determined that the stator winding has insulation faults such as corona or slot discharge. Another example: after counting the amplitude statistical parameter change trend for several consecutive sampling periods, if the amplitude statistical parameter change trend is a slow increase and exceeds the set threshold, it is determined that the stator winding insulation is aged.

[0075] Exemplarily, the amplitude statistical parameter may include at least one of the maximum value, the 95% value and the average value. The partial discharge stability index includes at least one of the partial discharge fluctuation value index and the 95% value statistical deviation index. Among them, the 95% value refers to the maximum value of the remaining amplitude after the amplitudes of each extreme value point in the extreme value sequence in the short period are arranged in order from large to small, and the 5% maximum value is removed. The partial discharge volatility index characterizes the change range of the maximum value of the amplitude of the extreme value sequence in the short period compared with the average value. The 95% value statistical deviation index characterizes the change range of the maximum value of the 95% value in each periodic time period in the short period compared with the minimum value of the 95% value in each periodic time period. It should be understood that the periodic time period refers to a plurality of repeated time periods with the same time interval divided into a short period in order to achieve the quantification of the 95% value statistical deviation index.

[0076] As a preferred method, the partial discharge fluctuation value index is quantified by formula 1:

[0077] (Formula 1);

[0078] In the formula, represents the partial discharge fluctuation value, is the maximum value of the extreme value sequence amplitude in a short period, It is the average value of the extreme value sequence amplitude within a short period.

[0079] As another preferred method, the 95% value statistical deviation index is quantified by formula 2:

[0080] (Formula 2);

[0081] In the formula, is the 95% statistical deviation, It is the maximum value of 95% value in each period of short cycle. It is the minimum value of 95% value in each period of short cycle.

[0082] As another preferred method, the amplitude statistical parameters include the maximum value, the 95% value and the average value. The partial discharge stability index includes the partial discharge volatility index and the 95% value statistical deviation index. By comprehensively considering the three statistical parameters of the maximum value, the 95% value and the average value, the evaluation of the change trend of the amplitude statistical parameters is more comprehensive and accurate. For example, when the maximum value change trend, the 95% value change trend and the average value change trend are all growth trends, the amplitude statistical parameters within the sampling period are determined to be a growth trend, which improves the credibility of the amplitude parameter change trend evaluation and eliminates abnormal fluctuations in the partial discharge value to the greatest extent.

[0083] Exemplarily, the extreme point sampling method used is a differential operation method. Accordingly, a specific implementation process of S200 is as follows:

[0084] S210. Perform differential calculation on the amplitude of each pulse in the periodic pulse sequence within the sampling period;

[0085] S220. Determine the extreme point according to the difference operation result;

[0086] S230. An extreme value sequence is formed by all extreme value points. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value.

[0087] The differential operation refers to calculating the amplitude difference between two adjacent pulse positions in the periodic pulse sequence. It can be seen that when the amplitude at the pulse position is greater than 0, the amplitude of the pulse position is greater than the amplitudes of the two adjacent pulse positions, which is an extreme point; when the amplitude at the pulse position is less than 0, the amplitude of the pulse position is less than the amplitudes of the two adjacent pulse positions, which is an extreme point. Accordingly, a specific implementation process of S220 is:

[0088] If the amplitude of a pulse position in the periodic pulse sequence is greater than 0, and the differential operation result for the pulse position satisfies the first determination condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point, and the first determination condition is that the amplitude of the pulse position is greater than the amplitude of the pulse position adjacent to it;

[0089] If the amplitude of a pulse position in a periodic pulse sequence is less than 0, and the differential operation result of the pulse amplitude meets the second judgment condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point. The second judgment condition is that the amplitude of the pulse position is smaller than the amplitude of the adjacent pulse position.

[0090] Embodiment 2

[0091] See also Figure 2 This embodiment provides a method for testing the stability of ultra-high frequency partial discharge of a hydro-turbine generator, which is an improvement of the first embodiment and specifically includes the following implementation steps:

[0092] SS100. Obtain a periodic pulse sequence formed after the partial discharge signal is discretized. This step is the same as step S100;

[0093] SS200. Sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, in which the amplitude of each extreme value point is associated with its corresponding occurrence time value and phase angle; wherein, sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, in which the amplitude of each extreme value point is associated with its corresponding occurrence time value, the process is the same as S200;

[0094] It can be seen that the range of the phase angle is [0,2π], and the phase angle can be calculated by formula 3:

[0095] (Formula 3);

[0096] In the formula, θ n represents the phase angle, t n %20ms means t n Divide by 20 milliseconds and take the remainder. 20ms is the period of one power frequency cycle.

[0097] SS300. Divide the short periods and long periods within the sampling period;

[0098] SS400. Determine the amplitude statistical parameters of the short period according to the extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters, the partial discharge stability index characterizes the stability of the partial discharge value; and cluster the extreme value sequence in the long period according to the amplitude and phase angle of the extreme point to determine the partial discharge event occurring in the long period; wherein, determine the amplitude statistical parameters of the short period according to the extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters, the process is the same as S400;

[0099] SS500. When the partial discharge stability index of a short cycle meets the preset requirements, the short cycle is determined as the target short cycle. This step is the same as S500;

[0100] SS600 analyzes each long-period partial discharge event and combines it with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding.

[0101] It is known that, in order to identify the insulation state of the stator winding, it is usually obtained by analyzing the change trend of the partial discharge value. In this embodiment, the statistical results of partial discharge events are added for joint analysis. Therefore, a specific implementation process of SS600 is as follows:

[0102] SS610. Analyze the changing trend of amplitude statistical parameters of each target short period within the sampling period;

[0103] SS620. Analyze the changing trend of the number of partial discharge events in each long period within the sampling period;

[0104] SS630. Combine and analyze the changing trends of the amplitude statistical parameters of each target short period within the sampling period and the changing trends of the number of partial discharge events of each long period in order to determine the insulation status of the stator winding.

[0105] For example, within a sampling period, if the variation trend of the amplitude statistical parameter and the variation trend of the number of partial discharge events are rapidly increasing and exceed the set threshold, it is determined that the stator winding has insulation faults such as corona or slot discharge. Another example: after counting the variation trend of the amplitude statistical parameter and the variation trend of the number of partial discharge events for several consecutive sampling periods, if the variation trend of the amplitude statistical parameter and the variation trend of the number of partial discharge events are slowly increasing and exceed the set threshold, it is determined that the stator winding insulation has aged.

[0106] When external interference is excluded, that is, under ideal conditions, as the insulation fault continues, the amplitude statistical parameter increases with time, and accordingly, the cumulative number of partial discharge events also increases with time. Therefore, in the above embodiment, based on the principle that the partial discharge value of the same insulation defect has a certain repeatability and shows aggregation characteristics with the amplitude and phase angle, the partial discharge events are identified, and the changing trend of the number of partial discharge events identified in a long period is combined with the changing trend of the amplitude statistical parameter in the target short period to map the insulation state of the stator winding. Compared with the first embodiment in which the insulation state of the stator winding is identified only by the changing trend of the amplitude statistical parameter, the identification result is more accurate.

[0107] Exemplarily, based on the principle that the partial discharge values ​​of the same insulation defect present aggregation characteristics, a clustering algorithm is used in SS400 to identify partial discharge events, such as a Euclidean distance clustering algorithm. Accordingly, according to the amplitude and phase angle of the extreme point, the extreme value sequence in a long period is clustered to determine the partial discharge events occurring in the long period, which specifically includes the following implementation steps:

[0108] SS410. Calculate the Euclidean distance between any two extreme value points in the extreme value sequence in the two-dimensional vector space composed of amplitude and phase angle. If the Euclidean distance is less than the first threshold, it is determined that the current two extreme value points are the same event, otherwise they are not the same event;

[0109] SS420. Count the total number of extreme value points that have been determined to be the same event. If the total number of extreme value points exceeds a second threshold, it is determined to be a partial discharge event.

[0110] It should be understood that the Euclidean distance clustering algorithm in the above content is only a preferred implementation of this embodiment. It is known that other clustering algorithms can be used based on the principle that the partial discharge values ​​of the same insulation defect present aggregation characteristics. This embodiment does not elaborate on this part.

[0111] The following content is combined Figures 3 to 8 , a brief description of an application example of this embodiment is given. In this application example, the sampling period T is 7 days, and the short period T 1 Take 5min, long period T 2 Take 60 minutes. A specific process of the application example is as follows:

[0112] A1. Combination Figure 3 As shown in the figure, the local discharge signal is dispersed in an equidistant manner: Assume that the sampling frequency is F = 1 GHz, and the collected local discharge signal After equidistant dispersion, we get a periodic pulse sequence , , Indicates partial discharge signal, represents the equidistance scatter function, represents the total number of sample points in the periodic pulse train, , the time interval between each sample point: , Indicates the sampling frequency.

[0113] A2. The partial discharge value indicates the healthy state of the stator winding insulation. When the partial discharge value is large, it means that the insulation state is poor and the risk of inducing insulation failure is higher. Therefore, the discrete post-cycle pulse sequence The largest partial discharge value is selected to form an extreme value sequence P ( n ), which is used to describe the insulation state of the stator winding in this sampling period. Figure 4 As shown, the extreme value sequence is obtained based on the difference operation P ( n The specific process of ) is as follows:

[0114] Periodic pulse train X ( n ) If the amplitude of each pulse position meets the preset conditions, the extreme value sequence is written P ( n ) and write the occurrence time value of the pulse position corresponding to the extreme point t n And by calculating and writing the corresponding phase angle , It can be expressed as: , phase angle The calculation is based on the above formula 3. Figure 4 middle, p 1 , p 2 , p n Etc. indicates writing extreme value sequence P ( n ) in the sample points.

[0115] A3. Combination Figure 5 As shown in the figure, the trend diagram of partial discharge value changes in each short period within the statistical sampling period:

[0116] The partial discharge value is characterized by three amplitude statistical parameters in a short period. The three amplitude statistical parameters are the maximum amplitude of each extreme point in the short period. P 1 , 95% value P 2 and average P 3 ;

[0117] Calculate the stability index of partial discharge in each short period. The stability of partial discharge is characterized by two indexes, namely partial discharge fluctuation value index and 95% value statistical deviation index.

[0118] The short period whose partial discharge stability index meets the preset requirements is determined as the target short period, and the change trend of partial discharge in the target short period is statistically analyzed; it is assumed that the extreme value sequence in a single short period is expressed as , Depend on K values, for the convenience of 95% example, assuming K The value is 100. ,Will After sorting from largest to smallest, we get , then the maximum value of the extreme value sequence amplitude in this short period , 95% value ,average value ;

[0119] Draw a trend chart of the three amplitude statistical parameters in a short period.

[0120] A4. Combination Figures 6 to 8 As shown in the figure, the trend diagram of partial discharge events in each long period within the statistical sampling period is as follows:

[0121] The amplitude and phase angle of each extreme point in the extreme value sequence in the long period are normalized, that is, the amplitude is divided by the preset reference value X base , The phase angle is divided by the preset reference value θ base;

[0122] The Euclidean distance in a two-dimensional vector space can be expressed as: ,in, Indicates the extreme point in the extreme value sequence in the long period and The Euclidean distance between , Indicates extreme point The normalized amplitude, Indicates extreme point Normalized amplitude , Indicates extreme point Normalized phase angle , Indicates extreme point Normalized phase angle;

[0123] Set the first threshold ξ ,when dis ( n )< ξ , then the two extreme points in the long period and Divided into the same event, assuming a long period T 2 After clustering, it is divided into H events, see Figure 6 As shown;

[0124] A second threshold is set, and the total number of extreme value points that have been determined to be the same event is counted. If the total number of extreme value points exceeds the second threshold, it is determined to be a partial discharge event. Figure 7 As shown, the number of partial discharge events in this long period is H 1 ;

[0125] Plot the changing trend of the number of partial discharge events over a long period.

[0126] A5. Stator winding insulation fault determination: The insulation status of the stator winding is determined by combining the change trend of the amplitude statistical parameters of each target short period within the sampling period and the change trend of the number of partial discharge events of each long period.

[0127] For example, if the change trend of the amplitude statistical parameters of each target short period within the sampling period is an increasing trend, and the change trend of the number of partial discharge events of each long period within the sampling period is an increasing trend, it is determined that an insulation fault has occurred in the stator winding.

[0128] Embodiment 3

[0129] The present embodiment provides a method for testing the stability of ultra-high frequency partial discharge of a hydro-turbine generator. The difference from the first and second embodiments is that: in actual program design, the process object links to which the periodic pulse sequence in the sampling period belongs are multiple, and the generated extreme value sequence includes multiple extreme value subsequences, and each extreme value subsequence is divided into each process object link one by one. It can be seen that the change of a certain process link during the extreme value point sampling process may cause a delay in the response change of the subsequent process link. In order to overcome the defect of inaccurate occurrence time value corresponding to the extreme value point caused by this problem, the method for testing the stability of ultra-high frequency partial discharge of a hydro-turbine generator provided by the present embodiment also includes the following implementation steps, which are included in step S200 or step SS200:

[0130] S001. If the change of process object link A causes a delay in the response change of process object link B, the time distance between a certain extreme value point in the extreme value subsequence corresponding to process object link A and each extreme value point in the extreme value subsequence corresponding to process object link B is calculated respectively, and the minimum value is taken to obtain the minimum time distance. Similarly, the minimum time distances corresponding to each extreme value point in the extreme value subsequence corresponding to process object link A are obtained, and all the minimum time distances form a set;

[0131] S002. Count the number of occurrences of each minimum time interval in the set to obtain the mode of the set, and determine the mode as the time interval for responding to the change delay in S001. When associating each extreme point in the extreme value subsequence corresponding to the process object link B with its corresponding occurrence time value, the occurrence time value is the difference between the time value of the pulse position corresponding to the extreme point and the above time interval.

[0132] For example, for any two process object links on the periodic pulse sequence and process object link , process object link A change in the process object link The delay of a response change is recorded as By sampling the extreme points of the periodic pulse sequence, the process object link The corresponding extreme value subsequence is expressed as , process object link The corresponding extreme value subsequence is expressed as , Represents the process object link The number of extreme points in the corresponding extreme value subsequence, Represents the process object link The number of extreme points in the corresponding extreme value subsequence, and They represent the time values ​​of the pulse positions corresponding to the extreme value points in the extreme value subsequence. The time distance between extreme value points is defined as follows: In Location The moment at the extreme point is , In Location The time at , then the time distance between the extreme points at different positions of two extreme subsequences is . Calculate the time interval The process is divided into two parts. The first part is to calculate the time distance between the extreme points at different positions of the two process object links, and the second part is to take the mode of the minimum time distance.

[0133] The first part is specifically: calculation In Location At the extreme point The time distance of each extreme point is obtained exist Position extreme point for link The result set of the time interval of each extreme point is recorded as , take the minimum value from the result set and record it as ; and so on, Calculate the extreme points at other positions respectively, record the minimum value each time, and get Each extreme point is The minimum time interval set of all extreme points is recorded as .

[0134] The second part is specifically: for the minimum value set Sort to get the sorted ordered set , then search the sorted set , count the number of occurrences of each minimum value, and get the mode of the set, recorded as ,So and The delay time between .

[0135] In the above embodiment, when a change occurs in a certain process object link causing a delay in responding to the change in another process object link, if the extreme point in the other process object link is still associated with the time value of the corresponding pulse position, an error occurs. This error makes the subsequent trend analysis of the change of the amplitude statistical parameters and the trend analysis of the change of the number of partial discharge events inaccurate. Through the delay time compensation method defined in this embodiment, the time value of the pulse position corresponding to the extreme point is corrected, thereby approaching the actual time value of each extreme point in the partial discharge signal, thereby improving the accuracy of the partial discharge test results.

[0136] Embodiment 4

[0137] See also Fig. 9 The present embodiment provides a hydro-generator ultra-high frequency partial discharge stability test device, which includes an acquisition module, an extreme value sampling module, a short cycle division module, a partial discharge stability index determination module, a target short cycle determination module and an insulation state mapping module connected in sequence.

[0138] The acquisition module is used to obtain the periodic pulse sequence formed after the partial discharge signal is discretized.

[0139] The extreme value sampling module is used to sample the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value.

[0140] The short cycle division module is used to divide each short cycle within the sampling period.

[0141] The partial discharge stability index determination module is used to determine the amplitude statistical parameters of the short period according to each extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters.

[0142] The target short cycle determination module is used to determine the short cycle as the target short cycle when the partial discharge stability index of the short cycle meets the preset requirements.

[0143] The insulation state mapping module is used to analyze the amplitude statistical parameters of each target short period within the sampling period in order to determine the insulation state of the stator winding.

[0144] Optionally, the extreme value points within the sampling period of the periodic pulse sequence are sampled to generate an extreme value sequence. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value. The specific process is as follows:

[0145] During the sampling period, the amplitude of each pulse in the periodic pulse sequence is differentially calculated;

[0146] Determine the extreme point based on the difference operation result;

[0147] An extreme value sequence is composed of all extreme value points. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value.

[0148] Optionally, the extreme point is determined according to the result of the differential operation. The specific process is as follows:

[0149] If the amplitude of a pulse position in the periodic pulse sequence is greater than 0, and the differential operation result for the pulse position satisfies the first determination condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point, and the first determination condition is that the amplitude of the pulse position is greater than the amplitude of the pulse position adjacent to it;

[0150] If the amplitude of a pulse position in a periodic pulse sequence is less than 0, and the differential operation result of the pulse amplitude meets the second judgment condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point. The second judgment condition is that the amplitude of the pulse position is smaller than the amplitude of the adjacent pulse position.

[0151] Optionally, the amplitude statistical parameters include at least one of a maximum value, a 95% value and an average value.

[0152] Optionally, the partial discharge stability index includes at least one of a partial discharge fluctuation value index and a 95% value statistical deviation index, the partial discharge fluctuation value index characterizes the variation range of the maximum value compared with the average value in a short period, and the 95% value statistical deviation index characterizes the variation range of the maximum value of 95% values ​​in each periodic time period in the short period compared with the minimum value of 95% values ​​in the aforementioned periodic time periods.

[0153] Optionally, the device further includes a long period division module, a phase angle correlation module, a partial discharge event clustering module and a joint analysis module, wherein:

[0154] The long cycle division module is used to divide each long cycle within the sampling period.

[0155] The phase angle association module is used to associate each extreme point with its corresponding phase angle.

[0156] The local discharge event aggregation module is used to cluster the extreme value sequences within a long period according to the amplitude and phase angle of the extreme value points, and then determine the local discharge events occurring within the long period.

[0157] The joint analysis module is used to analyze each long-period partial discharge event and combine it with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding.

[0158] Optionally, each long-period partial discharge event is analyzed and combined with the amplitude statistical parameter analysis results of each target short period in the sampling period to determine the insulation state of the stator winding. The specific process is as follows:

[0159] Analyze the changing trend of the amplitude statistical parameters of each target short period within the sampling period;

[0160] Analyze the changing trend of the number of partial discharge events in each long period within the sampling period;

[0161] The variation trend of the amplitude statistical parameters of each target short period and the variation trend of the number of partial discharge events of each long period within the sampling period are combined and analyzed to determine the insulation state of the stator winding.

[0162] Optionally, a combined analysis is performed on the variation trend of the amplitude statistical parameter of each target short period and the variation trend of the number of partial discharge events of each long period within the sampling period, so as to determine the insulation state of the stator winding. The specific process is as follows:

[0163] If the change trend of the amplitude statistical parameters of each target short period within the sampling period is an increasing trend, and the change trend of the number of partial discharge events of each long period within the sampling period is an increasing trend, it is used to determine that an insulation fault has occurred in the stator winding.

[0164] Optionally, according to the amplitude and phase angle of the extreme value point, the extreme value sequence in the long period is clustered to determine the local discharge event occurring in the long period. The specific process is as follows:

[0165] Calculate the Euclidean distance between any two extreme value points in the extreme value sequence in the two-dimensional vector space composed of amplitude and phase angle. If the Euclidean distance is less than the first threshold, it is determined that the current two extreme value points are the same event, otherwise they are not the same event.

[0166] The total number of extreme value points that have been determined to be the same event is counted, and if the total number of extreme value points exceeds a second threshold, it is determined to be a partial discharge event.

[0167] Optionally, there are multiple process object links to which the periodic pulse sequence within the sampling period belongs, the extreme value sequence includes multiple extreme value subsequences, and each extreme value subsequence is divided into each process object link one by one. The extreme value sampling module is also used to perform the following process:

[0168] If the change of process object link A causes a delay in the response change of process object link B, the time distance between a certain extreme value point in the extreme value subsequence corresponding to process object link A and each extreme value point in the extreme value subsequence corresponding to process object link B is calculated respectively, and the minimum value is taken to obtain the minimum time distance. Similarly, the minimum time distances corresponding to each extreme value point in the extreme value subsequence corresponding to process object link A are obtained, and all minimum time distances form a set;

[0169] Count the number of occurrences of each minimum time interval in the set to obtain the mode of the set, and determine the mode as the delayed time interval. When associating each extreme point in the extreme value subsequence corresponding to the process object link B with its corresponding occurrence time value, the occurrence time value is the difference between the time value of the pulse position corresponding to the extreme point and the time interval.

[0170] On the other hand, an embodiment of the present invention further provides a processor, which implements any one of the ultra-high frequency partial discharge stability testing methods for a hydro-turbine generator as described in Embodiments 1 to 3 when executing at least one instruction, or at least one program, or a code set, or an instruction set.

[0171] Optionally, the processor includes an FPGA processing chip and an ARM chip.

[0172] The FPGA processing chip is used to sample the extreme value points within the sampling period of the acquired periodic pulse sequence and generate an extreme value sequence, which is used to output to the ARM chip.

[0173] The ARM chip is used to: divide the sampling period into long and short periods; determine the target short period based on the partial discharge stability index of the amplitude statistical parameters in the short period, smooth out abnormal fluctuations, and determine the partial discharge events in the long period based on the clustering algorithm; analyze the amplitude statistical parameters in the target short period to obtain the first analysis result; analyze the number of partial discharge events in the long period to obtain the second analysis result; establish a mapping relationship between the first analysis result and the second analysis result and the insulation state of the stator winding.

[0174] On the other hand, an embodiment of the present invention further provides a machine-readable storage medium, on which instructions are stored, and when the instructions are executed by a processor, the processor is configured to execute any one of the ultra-high frequency partial discharge stability test methods for a hydro-turbine generator described in embodiments one to three.

[0175] Finally, an embodiment of the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any one of the ultra-high frequency partial discharge stability test methods for a hydro-turbine generator as described in Embodiments 1 to 3.

[0176] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0177] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0178] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0179] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0180] In a typical configuration, an electronic device includes one or more processors (CPU), input / output interfaces, network interfaces, memory, and storage. The memory may include non-permanent memory in a machine-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a machine-readable medium.

[0181] Machine-readable media include permanent and non-permanent, removable and non-removable media, and can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0182] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0183] The above are only embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A method for testing the stability of ultra-high frequency partial discharge of a hydro-generator, characterized in that: The method comprises: Obtaining a periodic pulse sequence formed by discretizing the partial discharge signal; Sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, and associating the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value; Divide each short period within the sampling period; Determine the amplitude statistical parameters of the short period according to the extreme value sequence in the short period, and determine the stability index of the partial discharge amount of the short period according to the amplitude statistical parameters; When the partial discharge stability index of a short cycle meets the preset requirements, the short cycle is determined as the target short cycle; Analyze the amplitude statistical parameters of each target short period within the sampling period; Divide each long period within the sampling period; Associating each extreme point in the extreme value sequence with its corresponding phase angle; According to the amplitude and phase angle of the extreme point, the extreme value sequence in the long period is clustered to determine the local discharge events occurring in the long period; The partial discharge events of each long period are analyzed, and the analysis results of the partial discharge events of each long period are combined with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding.

2. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 1, characterized in that: The step of sampling the extreme value points within the sampling period of the periodic pulse sequence to generate an extreme value sequence, and associating the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value, comprises: During the sampling period, the amplitude of each pulse in the periodic pulse sequence is differentially calculated; Determine the extreme point based on the difference operation result; An extreme value sequence is composed of all extreme value points. In the extreme value sequence, the amplitude of each extreme value point is associated with its corresponding occurrence time value.

3. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 2, characterized in that: Determining the extreme point according to the difference operation result includes: If the amplitude of a pulse position in the periodic pulse sequence is greater than 0, and the differential operation result for the pulse position satisfies the first determination condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point; the first determination condition is that the amplitude of the pulse position is greater than the amplitude of the pulse position adjacent to it; If the amplitude of a pulse position in a periodic pulse sequence is less than 0, and the differential operation result of the pulse amplitude meets the second judgment condition, it is determined that an extreme point appears at the pulse position, and the pulse amplitude is determined as the amplitude of the extreme point; the second judgment condition is that the amplitude of the pulse position is smaller than the amplitude of the pulse position adjacent to it.

4. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 1, characterized in that: The amplitude statistical parameter includes at least one of a maximum value, a 95% value and an average value.

5. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 4, characterized in that: The partial discharge stability index includes at least one of a partial discharge fluctuation value index and a 95% value statistical deviation index; The partial discharge fluctuation value index represents the variation range of the maximum value compared with the average value in a short period; The 95% value statistical deviation index represents the variation range of the maximum value of the 95% value in each periodic time period in the short cycle compared with the minimum value of the 95% value in each periodic time period.

6. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 1, characterized in that: The analyzing of each long-period partial discharge event and combining the analysis results of each long-period partial discharge event with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding includes: Analyze the changing trend of the amplitude statistical parameters of each target short period within the sampling period; Analyze the changing trend of the number of partial discharge events in each long period within the sampling period; The variation trend of the amplitude statistical parameters of each target short period within the sampling period and the variation trend of the number of partial discharge events of each long period are combined and analyzed to determine the insulation state of the stator winding.

7. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 6, characterized in that: The combined analysis of the change trend of the amplitude statistical parameter of each target short period and the change trend of the number of partial discharge events of each long period within the sampling period to determine the insulation state of the stator winding includes: If the change trend of the amplitude statistical parameters of each target short period within the sampling period is an increasing trend, and the change trend of the number of partial discharge events of each long period within the sampling period is an increasing trend, it is used to determine that an insulation fault has occurred in the stator winding.

8. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 1, characterized in that: The method of clustering the extreme value sequence in a long period according to the amplitude and phase angle of the extreme value point to determine the local discharge event occurring in the long period includes: Calculate the Euclidean distance between any two extreme value points in the extreme value sequence in the two-dimensional vector space composed of amplitude and phase angle. If the Euclidean distance is less than the first threshold, it is determined that the current two extreme value points are the same event, otherwise they are not the same event. The total number of extreme value points that have been determined to be the same event is counted, and if the total number of extreme value points exceeds a second threshold, it is determined to be a partial discharge event.

9. The method for testing the stability of ultra-high frequency partial discharge of a hydro-generator according to claim 1, characterized in that: There are multiple process object links to which the periodic pulse sequence within the sampling period belongs, the extreme value sequence includes multiple extreme value subsequences, and each extreme value subsequence is divided into each process object link in a one-to-one correspondence. The method also includes: If the change of process object link A causes a delay in the response of process object link B, the time distance between a certain extreme value point in the extreme value subsequence corresponding to process object link A and each extreme value point in the extreme value subsequence corresponding to process object link B is calculated respectively, and the minimum value is taken to obtain the minimum time distance. Similarly, the minimum time distances corresponding to each extreme value point in the extreme value subsequence corresponding to process object link A are obtained, and all minimum time distances form a set; Count the number of occurrences of each minimum time interval in the set to obtain the mode of the set, and determine the mode as the time interval of the delay; when associating each extreme point in the extreme value subsequence corresponding to the process object link B with its corresponding occurrence time value, the occurrence time value is the difference between the time value of the pulse position corresponding to the extreme point and the time interval.

10. A hydro-generator ultra-high frequency partial discharge stability test device, characterized in that: The device comprises: An acquisition module, used to acquire a periodic pulse sequence formed after the partial discharge signal is discretized; An extreme value sampling module is used to sample extreme value points within a sampling period of a periodic pulse sequence to generate an extreme value sequence, and associate the amplitude of each extreme value point in the extreme value sequence with its corresponding occurrence time value; A short cycle division module is used to divide each short cycle within the sampling period; A partial discharge stability index determination module is used to determine the amplitude statistical parameters of the short period according to each extreme value sequence in the short period, and determine the partial discharge stability index of the short period according to the amplitude statistical parameters; A target short cycle determination module is used to determine the short cycle as a target short cycle when the partial discharge stability index of the short cycle meets the preset requirements; The insulation state mapping module is used to analyze the amplitude statistical parameters of each target short period within the sampling period; The long cycle division module is used to divide each long cycle within the sampling period; A phase angle association module is used to associate each extreme point with its corresponding phase angle; A partial discharge event aggregation module is used to cluster the extreme value sequence in a long period according to the amplitude and phase angle of the extreme value point and determine the partial discharge events occurring in the long period; The joint analysis module is used to analyze each long-period partial discharge event and combine it with the amplitude statistical parameter analysis results of each target short period within the sampling period to determine the insulation state of the stator winding.

11. A machine-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by a processor, the processor is configured to execute the ultra-high frequency partial discharge stability test method for a hydro-turbine generator according to any one of claims 1 to 9.

12. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the ultra-high frequency partial discharge stability testing method for a hydro-turbine generator according to any one of claims 1 to 9 is implemented.