Abnormal dielectric loss diagnosis method of capacitor voltage transformer guided by group phase error

Abstract

The present invention provides a group phase error-guided abnormal dielectric loss diagnosis method for capacitor voltage transformers, which collects the initial phase data of the secondary voltage of the same-phase capacitor voltage transformer (CVT) of the same voltage level in the substation; extracts features from the initial phase data of the secondary voltage, and constructs dielectric loss phase offset feature quantities based on wavelet decomposition reconstruction and time series decomposition algorithms; performs density clustering on the dielectric loss phase offset feature quantities to locate abnormal CVTs; uses the initial phase data of the secondary voltage of normal CVTs to construct dielectric loss phase error feature quantities, and calculates the dielectric loss factor of abnormal CVTs. The dielectric loss abnormality diagnosis method proposed by the present invention can get rid of the dependence on periodic power outages, and can realize the positioning and abnormality degree monitoring of abnormal CVTs with dielectric loss without the need for additional sensors, and can evaluate the insulation degradation state of the CVT capacitor unit in real time, avoid the further development of potential insulation faults, and ensure the safe and stable operation of CVTs and power systems.

Description

Technical Field

[0001] The present invention relates to the field of power equipment status assessment, and more specifically, to a group phase error-guided abnormal dielectric loss diagnosis method for a capacitor voltage transformer. Background Art

[0002] Capacitor Voltage Transformer (CVT) is an important electric energy metering and protection device in the power system. It mainly consists of two parts: a capacitor voltage divider unit and an electromagnetic unit. The dielectric loss reflects the degree of moisture and aging of the insulation of the CVT capacitor voltage divider unit, and is an important characteristic parameter of the internal insulation performance. Abnormal increase in CVT dielectric loss will cause partial discharge, resulting in CVT capacitor unit breakdown and other faults, endangering the stable operation of the power system. Therefore, it is very important to monitor the CVT dielectric loss status, locate abnormal CVT and perform timely operation and maintenance.

[0003] The monitoring of the abnormal degree of dielectric loss of CVT depends on the measurement of dielectric loss factor. At present, the mainstream method for measuring dielectric loss of CVT capacitor voltage divider unit in high-voltage power grid is insulation preventive test. There are three test methods: positive connection method, reverse connection method and self-excitation method. This method needs to be coordinated with planned power outages. Its fixed detection cycle leads to the coexistence of "under-repair" and "over-repair" problems, and it is impossible to grasp the dielectric loss status of CVT in the cycle in real time.

[0004] In response to the above shortcomings, online monitoring technology has gained attention for its advantage of non-stop operation. This technology is mainly divided into two categories: "hardware method" and "software method". The "hardware method" uses the time difference between the zero-crossing points of the voltage and current signals to measure the dielectric loss. The measurement accuracy of this method greatly depends on the filter performance and the stability and anti-interference performance of the hardware circuit. With the development of data analysis technology, in recent years, the data-driven idea of ​​using CVT's own measurement data to diagnose capacitor breakdown faults has gained widespread attention. The "software method" relies on software and uses certain mathematical analysis methods to achieve online monitoring of the dielectric loss factor. Conventional methods mostly use signal reconstruction schemes to diagnose dielectric loss anomalies based on the relationship between the dielectric loss angle and the voltage and current phasors. Patent CN106093593B proposes an online detection method and system for dielectric loss of capacitive power equipment, which uses ground current and secondary voltage combined with Fourier analysis to calculate the dielectric loss factor of the capacitor. The document "Duan DP, ZengY, Huang CJ, et al. Digital algorithm based on orthogonal decomposition for measurement of dielectric loss factor[J]. IET Generation, Transmission&Distribution, 2008, 2(6): 868" proposes a digital algorithm for dielectric loss measurement based on orthogonal decomposition, which uses ground current to calculate the ratio of the norm of active current to reactive current, and then calculates the dielectric loss factor of the capacitor. However, both of the above methods require additional current sensors to collect CVT ground current signals, which are costly and pose safety risks, and the algorithm performance is affected by the performance of the collection equipment.

[0005] In summary, insulation preventive tests need to be coordinated with planned power outages, and there are problems of "under-repair" and "over-repair"; the performance of the existing dielectric loss online monitoring technology algorithm depends on the performance of the acquisition equipment and additional sensors need to be installed. Summary of the invention

[0006] In view of the technical problems existing in the prior art, the present invention provides a method for diagnosing abnormal dielectric loss of a capacitor voltage transformer guided by a group phase error, so that online diagnosis of abnormal CVT dielectric loss can be achieved without installing additional detection devices.

[0007] According to a first aspect of the present invention, a method for diagnosing abnormal dielectric loss of a capacitor voltage transformer guided by a group phase error is provided, comprising:

[0008] Step S1: For the busbars or lines of the same voltage level in parallel operation in the substation, the same-phase capacitor voltage transformers CVT are used to form a group, and the initial phase data of the secondary voltage of the group is collected. , =[ , ....， ，...， ], n represents the number of CVTs, m represents the number of sampling points, j represents the index of CVTs, and i represents the index of sampling points;

[0009] Step S2: Initial phase data of group secondary voltage Perform feature extraction and construct group phase feature , based on the population phase characteristic Calculate the population phase characteristic offset , =[ , , ..., ,..., ;

[0010] Step S3: Reconstruct the pair using wavelet decomposition To remove noise, denoted as , using the time series decomposition algorithm Decompose and remove the periodic fluctuations to obtain , construct the population dielectric loss phase shift characteristic , =[ , , ..., ,..., ];

[0011] Step S4: Based on the density clustering method, the phase shift characteristic of the group dielectric loss Clustering will be Divide into different groups, the CVT dielectric loss corresponding to the largest group is normal, the CVT dielectric loss corresponding to other groups is abnormal, and locate the CVT with abnormal dielectric loss;

[0012] Step S5: Constructing a reference phase characteristic for a normal CVT , based on the reference phase characteristic Calculate the reference phase characteristic offset , =[ , , ..., ,..., , based on the denoising algorithm and the timing decomposition algorithm, the reference phase characteristic offset Remove noise and periodic fluctuations to construct dielectric loss phase error characteristics , =[ , , ..., ,..., ];

[0013] Step S6: Calculate the dielectric loss factor of the abnormal CVT based on the mapping relationship between the dielectric loss phase error characteristic quantity and the dielectric loss.

[0014] The present invention provides a group phase error-guided abnormal dielectric loss diagnosis method for capacitor voltage transformers, which collects the initial phase data of the secondary voltage of the same-phase capacitor voltage transformer (CVT) of the same voltage level in the substation; extracts features from the initial phase data of the secondary voltage, and constructs dielectric loss phase offset feature quantities based on wavelet decomposition reconstruction and time series decomposition algorithms; performs density clustering on the dielectric loss phase offset feature quantities to locate abnormal CVTs; uses normal CVT secondary voltage initial phase data to construct dielectric loss phase error feature quantities, and calculates the dielectric loss factor of abnormal CVTs. The dielectric loss abnormality diagnosis method proposed by the present invention can get rid of the dependence on periodic power outages, and can realize the positioning and abnormality degree monitoring of abnormal CVTs with dielectric loss without the need for additional sensors, and can evaluate the insulation degradation state of CVT capacitor units in real time to avoid the further development of potential insulation faults and ensure the safe and stable operation of CVTs and power systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a topological diagram of a connection structure of five voltage transformers on a 220kV bus in a substation provided by an embodiment of the present invention;

[0016] Figure 2 A flow chart of a method for diagnosing abnormal dielectric loss of a capacitor voltage transformer guided by a group phase error provided by the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. In addition, the technical features in the various embodiments or single embodiments provided by the present invention can be arbitrarily combined with each other to form a feasible technical solution. This combination is not subject to the constraints of the sequence of steps and / or the structural composition mode, but must be based on the ability of ordinary technicians in this field to achieve. When the combination of technical solutions is contradictory or cannot be achieved, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0018] Five groups of capacitor voltage transformers on the same bus of a 220kV substation were selected to implement a method for diagnosing abnormal dielectric loss of capacitor voltage transformers proposed in this method. The wiring structure topology is shown in the figure below: Figure 1 shown.

[0019] Figure 2 A flow chart of a method for diagnosing abnormal dielectric loss of a capacitor voltage transformer guided by a group phase error provided by the present invention is shown in FIG. Figure 2 As shown, the method includes:

[0020] Step S1: For the busbars or lines of the same voltage level in parallel operation in the substation, the same-phase capacitor voltage transformers CVT are used to form a group, and the initial phase data of the secondary voltage of the group is collected. , =[ , ....， ，...， ], n represents the number of CVTs, m represents the number of sampling points, j represents the index of CVT, and i represents the index of sampling points.

[0021] Specifically, multiple groups of CVTs are connected to the bus or line of the same voltage level in the substation, and the same-phase capacitor voltage transformers (CVTs) of the same voltage level are constructed into a group to collect phase data under their operating status. , =[ , ....， ，...， ],in Respectively represent Group Phase CVT secondary voltage phase data, is the number of sampling points, that is, the secondary voltage phase data of n CVTs of the same phase at m sampling points are collected.

[0022] Step S2: Initial phase data of group secondary voltage Perform feature extraction and construct group phase feature , based on the population phase characteristic Calculate the population phase characteristic offset , =[ , , ..., ,..., .

[0023] It is understandable that the phase consistency of the primary voltage of the same-phase voltage transformer in the parallel operation state is used to calculate the initial phase data of the secondary voltage of the group. Perform feature extraction and construct group phase feature , further calculate the population phase characteristic offset , eliminating the change of phase acquisition data caused by the inconsistency between the primary voltage frequency and the acquisition frequency device frequency.

[0024] Specifically, record the initial phase data of the in-phase CVT secondary voltage , =[ , ....， ，...， ],in is the number of CVTs, is the number of sampling points, j is the index of CVT, and i is the index of the sampling point.

[0025] Calculate the average value of the initial phase data of the secondary voltage at each sampling point of the same-phase CVT to form the group phase characteristic , the average value is calculated as follows:

[0026] ;

[0027] calculate and The deviation between them is used to construct the population phase characteristic offset :

[0028] ;

[0029] in, represents the population phase characteristic of the j-th CVT at the i-th sampling point, Represents the population phase characteristic offset of the j-th CVT at the i-th sampling point.

[0030] Step S3: Reconstruct the pair using wavelet decomposition To remove noise, denoted as , using the time series decomposition algorithm Decompose and remove the periodic fluctuations to obtain , construct the population dielectric loss phase shift characteristic , =[ , , ..., ,..., ].

[0031] It is understandable that the phase characteristic offset of each CVT at each sampling point is calculated in step S2. In this step, each CVT is taken as a unit, and the phase characteristic offset of each CVT is processed to construct a group dielectric loss phase offset characteristic.

[0032] Phase characteristic offset for the jth CVT , using wavelet decomposition and reconstruction, To remove noise, denoted as ; Further, using the time series decomposition algorithm Decompose and remove the periodic fluctuations caused by factors such as environment and load to obtain , and then construct the dielectric loss phase shift characteristic , =[ , , ..., ,..., ], and obtain the relative relationship between the phases of the same-phase CVT group.

[0033] More specifically, for each transformer, wavelet decomposition and reconstruction method is used to Perform wavelet threshold denoising to form a denoised deviation data set Threshold The selection method is as follows: each CVT corresponds to Consistent:

[0034] ;

[0035] In the formula, is the signal length, is the standard deviation of the noise signal. .

[0036] Use the ACF autocorrelation function to determine the period of time series decomposition ;right Decompose the time series into trend terms, periodic terms and residuals. Subtract the periodic data to obtain , and then construct the dielectric loss phase shift characteristic , =[ , , ..., ,..., ].

[0037] Step S4: Based on the density clustering method, the phase shift characteristic of the group dielectric loss Clustering will be Divided into different groups, the CVT dielectric loss corresponding to the largest group is normal, and the CVT dielectric loss corresponding to other groups is abnormal, and the CVT with abnormal dielectric loss is located.

[0038] Among them, based on the density clustering method, the dielectric loss phase offset characteristic of all CVTs at all sampling points is Clustering is performed to divide all dielectric loss phase shift feature quantities into different groups, and the dielectric loss phase shift feature quantities in the largest group are marked as normal dielectric loss, and the dielectric loss phase shift feature quantities in other groups are marked as abnormal dielectric loss.

[0039] It can be understood that the dielectric loss phase shift feature quantities collected from n CVTs at m sampling points are clustered together to obtain multiple clustered groups, the dielectric loss phase shift feature quantities in the largest group are marked as normal dielectric loss, and the dielectric loss phase shift feature quantities in other groups are marked as abnormal dielectric loss.

[0040] In the clustering process, according to the physical structure of CVT, the mapping relationship between CVT measurement phase error and dielectric loss is derived as follows:

[0041] ;

[0042] In the formula, is the absolute change in CVT phase error, in units of , is the number of capacitors on the high voltage side, N is the number of capacitors on the medium voltage side, is the change in dielectric loss factor on the high-voltage side, is the change in dielectric loss factor on the medium voltage side.

[0043] According to GB / T 20840.5-2013, the dielectric loss factor of film-paper composite insulation capacitors (film-paper-film or paper-film-paper) at rated power frequency voltage should be less than 0.15%. The CVT measurement phase error threshold is as follows:

[0044] ;

[0045] When clustering the dielectric loss phase offset characteristics of n CVTs collected at m sampling points, the clustering neighborhood radius is During the clustering process, the phase data of a CVT deviates from the normal group by more than When , it is identified as an anomaly, so the neighborhood radius Depend on Determine as follows:

[0046] .

[0047] After clustering, record the number of cluster groups according to the clustering results And the amount of data contained in each cluster group. , then all transformer dielectric losses are normal; if , the cluster group with the largest amount of data is regarded as normal data, and the other groups are regarded as abnormal data. In the data, set labels [0] and [1], label [0] marks normal data, and label [1] marks abnormal data.

[0048] The abnormality of the dielectric loss phase shift characteristic value of each sampling point of each CVT is marked, and the number of sampling points of the dielectric loss phase shift characteristic value marked as normal dielectric loss and the number of sampling points of the dielectric loss phase shift characteristic value marked as abnormal dielectric loss in each CVT are counted, and the proportion of the number of sampling points of the dielectric loss phase shift characteristic value marked as abnormal dielectric loss in each CVT is calculated. ,like If the value is greater than the set threshold, the CVT is a CVT with abnormal dielectric loss. When the value is greater than 95%, the CVT is identified as an abnormal CVT, and the number of abnormal CVTs is recorded. , which corresponds to The phase CVT numbers are .

[0049] Step S5: construct a reference phase characteristic for a normal CVT , based on the reference phase characteristic Calculate the reference phase characteristic offset , =[ , , ..., ,..., , based on the denoising algorithm and the timing decomposition algorithm, the reference phase characteristic offset Remove noise and periodic fluctuations to construct dielectric loss phase error characteristics , =[ , , ..., ,..., ].

[0050] It can be understood that, according to the abnormal CVT positioning result in step S4, a reference phase characteristic quantity is constructed for the normal CVT therein. , and calculate the reference phase characteristic offset , eliminate the influence of abnormal CVT, then use the denoising algorithm and timing decomposition algorithm to remove noise and periodic fluctuations, and construct the dielectric loss phase error characteristic .

[0051] Specifically, the numbers of all abnormal CVTs in the same phase are ,exist Select the normal CVT secondary voltage initial phase data and calculate The average phase data of the normal CVT at each sampling point is used to construct the reference phase characteristic , the calculation formula is as follows:

[0052] ;

[0053] in, represents the phase characteristic of the j-th CVT at the i-th sampling point, Indicates The phase characteristic of an abnormal CVT at the i-th sampling point, is the number of abnormal CVTs.

[0054] calculate and The deviation between them is used to construct the reference phase characteristic offset , the calculation formula is as follows:

[0055] ;

[0056] in, Represents the reference phase characteristic offset of the j-th CVT at the ith sampling point.

[0057] Then, the reference phase feature offset is calculated based on the denoising algorithm and the timing decomposition algorithm. Remove noise and periodic fluctuations to construct dielectric loss phase error characteristics , =[ , , ..., ,..., ]. For specific denoising and time series decomposition, please refer to step 3 and will not be repeated here.

[0058] Step S6: Calculate the dielectric loss factor of the abnormal CVT based on the mapping relationship between the dielectric loss phase error characteristic quantity and the dielectric loss.

[0059] Among them, based on the constructed dielectric loss phase error characteristic , obtain the abnormal CVT dielectric loss phase error characteristic =[ ], is the number of the abnormal CVT.

[0060] Calculation of dielectric loss phase error characteristic of abnormal CVT The average As shown below:

[0061] ;

[0062] in is the number of sampling points.

[0063] Calculate the dielectric loss phase error characteristic of each CVT , based on dielectric loss phase error characteristic The mapping relationship with the dielectric loss is used to calculate the dielectric loss factor of the abnormal CVT, and finally the positioning and abnormal degree monitoring of the CVT with abnormal dielectric loss can be achieved.

[0064] The calculation formula for the dielectric loss factor of abnormal CVT is:

[0065] ;

[0066] in, The unit is radian.

[0067] The abnormal dielectric loss diagnosis method of a capacitor voltage transformer provided by the present invention is described below with a specific example, which mainly includes the following steps:

[0068] (1) Taking the A-phase CVT group on the 220kV side as an example, there are a total of 5 CVTs, all of which have undergone periodic inspection of dielectric loss and are recorded as , , , , , the dielectric loss factor is shown in Table 1, where The dielectric loss on the medium voltage side is abnormal, and other CVTs are in normal condition.

[0069] Table 1 A-phase CVT dielectric loss

[0070]

[0071] The secondary voltage phase data of each group of CVTs to be evaluated in each state is collected for one evaluation cycle, where the evaluation cycle is set to 7 days. The output frequency of the CVT initial phase data collection sample in this embodiment is 1min / time. ,by Taking as an example for analysis, part of the secondary voltage initial phase data set is extracted as shown in Table 2.

[0072] Table 2 Secondary voltage initial phase

[0073]

[0074] (2) Calculation The average value of phase data at each moment is used to construct the phase characteristic of the group ,calculate and The deviation between them, that is, the population phase characteristic offset Using wavelet decomposition and reconstruction method, Perform wavelet threshold denoising to form a post-noise deviation data set ; Using the ACF autocorrelation function, determine The change cycle ;right Perform STL timing decomposition, using Subtract the periodic term data to construct the dielectric loss phase shift characteristic . Cut part As shown in Table 3, some As shown in Table 4.

[0075] Table 3 Group phase characteristic offset

[0076]

[0077] Table 4 Dielectric loss phase shift characteristic quantity

[0078]

[0079] (3) Yes Perform DBSCAN cluster analysis and record the number of cluster groups And the amount of data contained in each cluster group. The clustering results are shown in Table 5. According to the clustering results, ,right Data setting label, calculate the proportion of abnormal data in each CVT The results show that ,Right now Abnormal dielectric loss, - It is in normal state and consistent with the actual situation.

[0080] Table 5 DBSCAN clustering results

[0081]

[0082] (4) Based on the abnormal identification result of S3, a reference phase characteristic offset is constructed for the normal CVT secondary voltage initial phase ,right Perform denoising and de-periodic processing to construct dielectric loss phase error feature 。 Cut part and As shown in Table 6 and Table 7.

[0083] Table 6 Reference phase characteristic offset

[0084]

[0085] Table 7 Dielectric loss phase error characteristic quantity

[0086]

[0087] Calculate the abnormal CVT dielectric loss phase error characteristic quantity during the evaluation period The average value of , the dielectric loss factor of the abnormal CVT is further calculated as follows:

[0088] , representing CVT 5-A The dielectric loss of the capacitor on the medium voltage side of the capacitor voltage divider unit is abnormal. For the 220kV side CVT, , satisfying the following formula:

[0089] .

[0090] At this time, relative to the actual measurement result, the absolute error between the method and the actual situation is:

[0091] .

[0092] The above method is used to diagnose abnormal CVT dielectric loss. The results show that CVT 5-A The dielectric loss on the medium-voltage side of the capacitor voltage divider unit is abnormal, which is consistent with the actual situation; the dielectric loss is , the absolute error of the method , that is, this method can realize the positioning and abnormality degree monitoring of CVT with abnormal dielectric loss.

[0093] The embodiment of the present invention provides a group phase error-guided abnormal dielectric loss diagnosis method for capacitor voltage transformers, which collects the initial phase data of the secondary voltage of the same-phase capacitor voltage transformers (CVTs) of the same voltage level in the substation; extracts features from the initial phase data of the secondary voltage, and constructs dielectric loss phase offset feature quantities based on wavelet decomposition reconstruction and time series decomposition algorithms; performs density clustering on the dielectric loss phase offset feature quantities to locate abnormal CVTs; uses normal CVT secondary voltage initial phase data to construct dielectric loss phase error feature quantities, and calculates the dielectric loss factor of abnormal CVTs. The dielectric loss abnormality diagnosis method proposed by the present invention can get rid of the dependence on periodic power outages, and can realize the positioning and abnormality degree monitoring of abnormal CVTs with dielectric loss without the need for additional sensors, and can evaluate the insulation degradation state of CVT capacitor units in real time to avoid the further development of potential insulation faults, thereby ensuring the safe and stable operation of CVTs and power systems.

[0094] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0095] 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.

[0096] 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 computer, or other programmable data processing device to produce 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.

[0097] 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.

[0098] 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 A step that specifies a function in one or more boxes.

[0099] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for diagnosing abnormal dielectric loss of capacitor voltage transformer guided by group phase error, characterized in that: include: Step S1: For the buses or lines of the same voltage level in parallel operation in the substation, the same-phase capacitor voltage transformers CVT are used to form a group, and the initial phase data of the secondary voltage of the group is collected. , =[ , ....， ，...， ], n represents the number of CVTs, m represents the number of sampling points, j represents the index of CVTs, and i represents the index of sampling points; Step S2: Initial phase data of group secondary voltage Extract features and construct group phase features , based on the population phase characteristic Calculate the population phase characteristic offset , =[ , , ..., ,..., ; Step S3: Reconstruct the pair using wavelet decomposition To remove noise, denoted as , using the time series decomposition algorithm Decompose and remove the periodic fluctuations to obtain , construct the population dielectric loss phase shift characteristic , =[ , , ..., ,..., ]; Step S4: Based on the density clustering method, the phase shift characteristic of the group dielectric loss Clustering will be Divide into different groups, the CVT dielectric loss corresponding to the largest group is normal, the CVT dielectric loss corresponding to other groups is abnormal, and locate the CVT with abnormal dielectric loss; Step S5: Constructing a reference phase characteristic for a normal CVT , based on the reference phase characteristic Calculate the reference phase characteristic offset , =[ , , ..., ,..., , based on the denoising algorithm and the timing decomposition algorithm, the reference phase characteristic offset Remove noise and periodic fluctuations to construct dielectric loss phase error characteristics , =[ , , ..., ,..., ]; Step S6: Calculate the dielectric loss factor of the abnormal CVT based on the mapping relationship between the dielectric loss phase error characteristic quantity and the dielectric loss.

2. The abnormal dielectric loss diagnosis method of capacitor voltage transformer according to claim 1, characterized in that: The step S2 comprises: Record the initial phase data of the secondary voltage of the same phase CVT , =[ , ....， ，...， ],in is the number of CVTs, is the number of sampling points, j is the index of CVT, and i is the index of the sampling point; Calculate the average value of the initial phase data of the secondary voltage at each sampling point of the same-phase CVT to form the group phase characteristic : ； calculate and The deviation between them is used to construct the population phase characteristic offset : ； in, represents the population phase characteristic of the j-th CVT at the i-th sampling point, Represents the population phase characteristic offset of the j-th CVT at the i-th sampling point.

3. The abnormal dielectric loss diagnosis method of capacitor voltage transformer according to claim 1, characterized in that: The step S3 comprises: For each CVT, wavelet decomposition and reconstruction method is used to Perform wavelet threshold denoising to form a denoised deviation data set ; Use the ACF autocorrelation function to determine the period of time series decomposition , based on the cycle right Decompose the time series into trend terms, periodic terms and residuals. Subtract the periodic term and we get , construct dielectric loss phase shift characteristic , =[ , , ..., ,..., ].

4. The abnormal dielectric loss diagnosis method of a capacitor voltage transformer according to claim 1, characterized in that: The step S4 includes: Based on the density clustering method, the dielectric loss phase offset characteristics of all CVTs at all sampling points are calculated. Clustering is performed to divide all dielectric loss phase shift feature quantities into different groups, and the dielectric loss phase shift feature quantities in the largest group are marked as normal dielectric loss, and the dielectric loss phase shift feature quantities in other groups are marked as abnormal dielectric loss; Among them, in the phase shift characteristic of the group dielectric loss When clustering, the cluster neighborhood radius The calculation method is as follows: ； in, is the number of CVTs on the high-pressure side among all CVTs, N is the number of CVTs on the medium-pressure side, is the dielectric loss threshold, is the abnormal phase error threshold, in units of .

5. The abnormal dielectric loss diagnosis method of a capacitor voltage transformer according to claim 4, characterized in that: The step S4 further comprises: Count the number of sampling points of dielectric loss phase shift characteristic quantity marked as normal dielectric loss and the number of sampling points of dielectric loss phase shift characteristic quantity marked as abnormal dielectric loss in each CVT, and calculate the proportion of the number of sampling points of dielectric loss phase shift characteristic quantity marked as abnormal dielectric loss in each CVT ,like If it is greater than the set threshold, the CVT is a CVT with abnormal dielectric loss.

6. The abnormal dielectric loss diagnosis method of capacitor voltage transformer according to claim 1, characterized in that: The step S5 comprises: The numbers of all abnormal CVTs in the same phase are ,exist Select the normal CVT secondary voltage initial phase data and calculate The average phase data of the normal CVT at each sampling point is used to construct the reference phase characteristic : ； in, represents the phase characteristic of the j-th CVT at the i-th sampling point, Indicates The phase characteristic of an abnormal CVT at the i-th sampling point, is the number of abnormal CVTs; calculate and The deviation between them is used to construct the reference phase characteristic offset : ； in, Represents the reference phase characteristic offset of the j-th CVT at the i-th sampling point.

7. The abnormal dielectric loss diagnosis method of capacitor voltage transformer according to claim 1, characterized in that: The step S6 comprises: Based on the constructed dielectric loss phase error characteristic , obtain the abnormal CVT dielectric loss phase error characteristic =[ ], is the number of the abnormal CVT; Calculation of dielectric loss phase error characteristic of abnormal CVT The average , as shown below: ； in is the number of sampling points; based on The mapping relationship between the dielectric loss and the dielectric loss factor of the abnormal CVT is calculated: ； in is the number of capacitors on the high voltage side, N is the number of capacitors on the medium voltage side, The unit is radian.

Images