1. A partial discharge fault diagnosis method for a voltage transformer is characterized by comprising the following steps:

s1: collecting partial discharge data I of various defect models under a withstand voltage test, and acquiring characteristic data of each defect model according to the partial discharge data I;

s2: establishing a feature identification model according to the feature data and the defect type of the defect model;

s3: collecting partial discharge data II of the voltage transformer to be detected under the withstand voltage test;

s4: and acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the feature recognition model.

2. The method of claim 1, wherein the defect model comprises: a point discharge model, an insulation defect discharge model, an air gap discharge model and a particle discharge model.

3. The method according to claim 1, wherein the step S1 specifically includes the steps of:

s1.1, connecting the defect model and a standard voltage transformer in parallel and then connecting the defect model and the standard voltage transformer into a test circuit of the withstand voltage test;

s1.2, acquiring an initial discharge voltage Ua and a breakdown voltage Ub of the defect model, and setting an experimental voltage as Um, wherein Ua is less than Um and less than Ub;

s1.3, keeping the experimental voltage stable, collecting the partial discharge data I, and carrying out normalization processing on the partial discharge data I to obtain the characteristic data, wherein the characteristic data comprises the partial discharge data I subjected to normalization processing and the current experimental voltage Um;

s1.4, storing the characteristic data, and adding a corresponding label to the characteristic data, wherein the label is the defect type of the current defect model;

s1.5, adjusting the experimental voltage Um and repeating the steps S1.3-S1.4;

s1.6, replacing the defect model and repeating the steps S1.1-S1.5.

4. The method according to claim 3, wherein the step S2 specifically comprises the steps of:

s2.1, dividing all the stored characteristic data into two parts, wherein one part is learning data and the other part is evaluation data;

s2.2, fitting the learning data by adopting a random gradient descent algorithm to form the feature recognition model;

s2.3, judging the accuracy of the feature recognition model through the evaluation data, and finishing the establishment of the feature recognition model if the accuracy reaches the standard.

5. Method according to claim 4, characterized in that in step S2.1: the number ratio of the learning data to the evaluation data was 8: 2.

6. The method according to claim 4, wherein the step S3 is specifically:

s3.1, connecting the voltage transformer to be detected into a test circuit of the withstand voltage test;

s3.2, acquiring an initial discharge voltage Ua and a breakdown voltage Ub of the voltage transformer to be detected, and setting an experimental voltage as Um, wherein Ua is less than Um and less than Ub;

and S3.3, adjusting the experiment voltage Um until the partial discharge value is larger than a set value, keeping the experiment voltage stable, and collecting the partial discharge data II.

7. The method according to claim 6, wherein the step S4 is specifically:

and acquiring the partial discharge data II and the current experimental voltage Um after normalization processing, substituting the partial discharge data II and the current experimental voltage Um into the characteristic identification model, and acquiring the defect type of the voltage transformer to be detected.

8. A voltage transformer partial discharge fault diagnosis device, comprising:

the first acquisition module is used for acquiring partial discharge data I of various defect models under a withstand voltage test and acquiring characteristic data of each defect model according to the partial discharge data I;

the establishing module is used for correspondingly establishing a characteristic identification model according to the characteristic data and the defect type of the defect model;

the acquisition module I is used for acquiring partial discharge data II of the voltage transformer to be detected under the withstand voltage test;

and the acquisition module is used for acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the feature identification model.

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the diagnostic method of any one of claims 1-7.

10. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the diagnostic method of any one of claims 1-7.

Background

Partial Discharges (PDs) of electrical equipment refer to a phenomenon that when insulation defects exist in the electrical equipment, if voltage is applied to the electrical equipment, the insulation defects of the electrical equipment are repeatedly broken down and extinguished. In the operation process of electrical equipment, breakdown cannot be caused immediately when the release amount is not large, but insulation defects are more deteriorated after long-time operation, and finally breakdown is caused. Because the partial discharge of the electrical equipment is greatly related to the insulation condition, the accident probability of the electrical equipment can be greatly reduced by regularly detecting and analyzing the partial discharge of the electrical equipment, particularly high-voltage electrical equipment.

At present, in factory detection of a voltage transformer, partial discharge is an item which needs to be detected, and the partial discharge level can effectively reflect the insulation defect condition of a tested product. At present, when a manufacturer leaves a factory and tests, if the partial discharge result is abnormal, the manufacturer generally needs to dissect a tested product, the operation is complicated, and great cost waste is caused.

Disclosure of Invention

In view of this, embodiments of the present invention provide a method and an apparatus for diagnosing a partial discharge fault of a voltage transformer, and an electronic device, which analyze a partial discharge signal tested by a partial discharge detector and diagnose a type of an insulation defect, so as to specifically maintain a defective product, facilitate maintenance operation, and save cost.

In a first aspect, an embodiment of the present invention provides a voltage transformer partial discharge fault diagnosis method, including the following steps:

s1: collecting partial discharge data I of various defect models under a withstand voltage test, and acquiring characteristic data of each defect model according to the partial discharge data I;

s2: establishing a feature identification model according to the feature data and the defect type of the defect model;

s3: collecting partial discharge data II of the voltage transformer to be detected under the withstand voltage test;

s4: and acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the feature recognition model.

The beneficial effects of the above embodiment are: the diagnosis method can conveniently and quickly determine the defect type of the voltage transformer to be detected, and is convenient for manufacturers to maintain defective products in a targeted manner, so that the cost is saved.

According to a specific implementation manner of the embodiment of the present invention, the defect model includes: a point discharge model, an insulation defect discharge model, an air gap discharge model and a particle discharge model. The defect model is artificially manufactured and contains common defect types, the defect model can be replaced by a defect voltage transformer with a determined type, and the determined defect model is used for acquiring characteristic data corresponding to the defect type.

According to a specific implementation manner of the embodiment of the present invention, the step S1 specifically includes the following steps:

s1.1, connecting the defect model and a standard voltage transformer in parallel and then connecting the defect model and the standard voltage transformer into a test circuit of the withstand voltage test;

s1.2, acquiring an initial discharge voltage Ua and a breakdown voltage Ub of the defect model, and setting an experimental voltage as Um, wherein Ua is less than Um and less than Ub;

s1.3, keeping the experimental voltage stable, collecting the partial discharge data I, and carrying out normalization processing on the partial discharge data I to obtain the characteristic data, wherein the characteristic data comprises the partial discharge data I subjected to normalization processing and the current experimental voltage Um;

s1.4, storing the characteristic data, and adding a corresponding label to the characteristic data, wherein the label is the defect type of the current defect model;

s1.5, adjusting the experimental voltage Um and repeating the steps S1.3-S1.4;

s1.6, replacing the defect model and repeating the steps S1.1-S1.5.

The characteristic data corresponding to the determined defect model is collected in the above mode, so that a sample library is formed, and support is provided for subsequently establishing a characteristic identification model.

According to a specific implementation manner of the embodiment of the present invention, the step S2 specifically includes the following steps:

s2.1, dividing all the stored characteristic data into two parts, wherein one part is learning data and the other part is evaluation data;

s2.2, fitting the learning data by adopting a random gradient descent algorithm to form the feature recognition model;

s2.3, judging the accuracy of the feature recognition model through the evaluation data, and finishing the establishment of the feature recognition model if the accuracy reaches the standard.

And forming a feature recognition model of feature data corresponding to the label of the feature data by the learning data through a neural network, substituting the feature data in the evaluation data into the feature recognition model, comparing the output defect type with the actual label, if the accuracy rate exceeds a set value, determining that the feature recognition model is correct, and if the accuracy rate is lower than the set value, adjusting the parameters to establish the feature recognition model again.

According to a specific implementation manner of the embodiment of the present invention, in step S2.1: the number ratio of the learning data to the evaluation data was 8: 2.

According to a specific implementation manner of the embodiment of the present invention, the step S3 specifically includes:

s3.1, connecting the voltage transformer to be detected into a test circuit of the withstand voltage test;

s3.2, acquiring an initial discharge voltage Ua and a breakdown voltage Ub of the voltage transformer to be detected, and setting an experimental voltage as Um, wherein Ua is less than Um and less than Ub;

and S3.3, adjusting the experiment voltage Um until the partial discharge value is larger than a set value, keeping the experiment voltage stable, and collecting the partial discharge data II.

According to a specific implementation manner of the embodiment of the present invention, the step S4 specifically includes:

and acquiring the partial discharge data II and the current experimental voltage Um after normalization processing, substituting the partial discharge data II and the current experimental voltage Um into the characteristic identification model, and acquiring the defect type of the voltage transformer to be detected.

In a second aspect, an embodiment of the present invention provides a voltage transformer partial discharge fault diagnosis apparatus, including:

the first acquisition module is used for acquiring partial discharge data I of various defect models under a withstand voltage test and acquiring characteristic data of each defect model according to the partial discharge data I;

the establishing module is used for correspondingly establishing a characteristic identification model according to the characteristic data and the defect type of the defect model;

the acquisition module I is used for acquiring partial discharge data II of the voltage transformer to be detected under the withstand voltage test;

and the acquisition module is used for acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the feature identification model.

In a third aspect, an embodiment of the present invention further provides an electronic device, where the electronic device includes:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the diagnostic method of any one of the preceding first aspects or any implementation manner of the first aspect.

In a fourth aspect, the embodiments of the present invention also provide a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the diagnostic method of the first aspect or any implementation manner of the first aspect.

The voltage transformer partial discharge fault diagnosis method and device, the electronic equipment and the non-transient computer readable storage medium provided by the embodiment of the invention provide a mode for determining the fault type of the voltage transformer to be detected without decomposition for a user.

Drawings

In order to more clearly illustrate the detailed description of the invention or the technical solutions in the prior art, the drawings that are needed in the detailed description of the invention or the prior art will be briefly described below. Throughout the drawings, like elements or portions are generally identified by like reference numerals. In the drawings, elements or portions are not necessarily drawn to scale.

Fig. 1 is a flowchart illustrating a partial discharge fault diagnosis method for a voltage transformer according to an embodiment of the present invention;

FIG. 2 shows a test circuit for defect model access in an embodiment of the invention;

FIG. 3 shows a test circuit for connecting a voltage transformer to be tested in the embodiment of the invention;

fig. 4 shows a block diagram of a partial discharge fault diagnosis apparatus for a voltage transformer according to an embodiment of the present invention;

fig. 5 is a schematic structural diagram illustrating an electronic device for diagnosing partial discharge faults of a voltage transformer according to an embodiment of the present invention;

Detailed Description

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings. The following examples are only for illustrating the technical solutions of the present invention more clearly, and therefore are only examples, and the protection scope of the present invention is not limited thereby.

It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which the invention pertains.

Fig. 1 is a flowchart illustrating steps of a partial discharge fault diagnosis method for a voltage transformer according to an embodiment of the present invention, and referring to fig. 1, the method includes the following steps:

s1: and collecting the partial discharge data I of each defect model under the withstand voltage test, and acquiring the characteristic data of each defect model according to the partial discharge data I.

Wherein the defect model comprises: a point discharge model, an insulation defect discharge model, an air gap discharge model and a particle discharge model. The defect model can be made manually or replaced by a defect voltage transformer of a determined type.

The method specifically comprises the following steps:

s1.1, connecting a defect model and a standard voltage transformer in parallel and then connecting the defect model and the standard voltage transformer into a test circuit of a withstand voltage test, as shown in figure 2;

s1.2, gradually increasing the experimental voltage by adopting a step-up method, determining and obtaining the initial discharge voltage Ua and the breakdown voltage Ub of the defect model, and setting the experimental voltage as Um, wherein Ua is less than Um and less than Ub;

s1.3, keeping the experimental voltage stable, and collecting and recording discharge signal data through JFD-1C;

examples are as follows: collecting PRPD (Phase Resolved Partial Discharge) data of a frame of Partial Discharge waveform, firstly normalizing the PRPD data, wherein the normalized data are 360 floating point numbers, and storing the normalized data and experimental voltage Um data (1 floating point number) into a MySql database system; since the partial discharge data has periodicity, only one frame of data needs to be collected, and 361 floating point numbers saved at this time are feature data.

S1.4, storing the characteristic data, and adding a corresponding label to the characteristic data, wherein the label is the defect type of the current defect model;

marking the stored characteristic data sample records in a database, and writing corresponding type data according to the defect type of the defect model; the specific way is to mark the TYPE field information of the record as data of the following defect TYPEs: a) air gap discharge AirGap; b) particle discharge, Granule; c) a Needle discharge; d) insulation defect WeekInsulantion.

S1.5, gradually adjusting the experimental voltage Um and repeating the steps S1.3-S1.4;

s1.6, replacing defect models of different defect types and repeating the steps S1.1-S1.5.

S2: and establishing a feature identification model according to the feature data and the defect type of the defect model.

S2.1, dividing all stored characteristic data into two parts, wherein one part is learning data and the other part is evaluation data;

reading all sample data from the database, and performing the following steps on all sample data according to the weight ratio of 8: the ratio of 2 was randomly divided into two parts, one for learning data and one for evaluation data. Recording characteristic data and mark type fields of different defect models in a database, and carrying out One-Hot coding (One-Hot) on the mark type field data to be used as a Label value (Label); the specific coding mode is as follows: a) AirGap: [1,0,0,0 ]; b) the weight ratio of Granule: [0,1,0,0 ]; c) a Needle: [0,0,1,0 ]; d) WeekInsulation: [0,0,0,1].

S2.2, fitting the learning data by adopting a random gradient descent algorithm to form a feature identification model;

fitting the learning data by adopting an SGD (Stochastic Gradient Descent) algorithm to form a data model; the whole fitting process is as follows:

now, there are several sample data in the database, whose characteristic values are 361 data and the label value is 4 data, and a function is found by the neural network algorithm, so that:

(y₁，y₂，y₃，y₄)＝f(x₁，x₁，…x₃₆₁) (1)

can be simplified as follows:

Y＝f(X) (2)

wherein X is characteristic data, Y is label data, and the data types are numerical value matrixes.

At the beginning of the fitting, we preset all parameter values as random numbers conforming to the gaussian distribution, then bring k sample data (e.g. 100) into the function, and calculate the cumulative error E:

in the formula (3), y_kCalculating a value, t, for the function in equation (2)_kIs the correct tag value. Gradually fine-tuning internal parameters of the neural network algorithm to gradually reduce the accumulated error, repeating the process (such as 2000 times), and taking the function with the minimum accumulated error as the final data model which represents X rotationThe calculation procedure and parameter values for Y.

And S2.3, judging the accuracy of the feature recognition model through the evaluation data, and finishing the establishment of the feature recognition model if the accuracy reaches the standard.

The evaluation method is that the evaluation data (the data do not participate in fitting) are brought into a data model (function) one by one, the output of the data model is compared with the actual label data, then the overall accuracy of the model is judged, and if the accuracy exceeds 90%, the model can be adopted to diagnose in a generation environment.

S3: and collecting partial discharge data II of the voltage transformer to be detected under the voltage withstand test.

S3.1, connecting the voltage transformer to be detected into a test circuit of a withstand voltage test, as shown in figure 3;

s3.2, acquiring an initial discharge voltage Ua and a breakdown voltage Ub of the voltage transformer to be detected, and setting an experimental voltage as Um, wherein Ua is less than Um and less than Ub;

and S3.3, adjusting the experimental voltage Um until the partial discharge value is larger than a set value (the partial discharge value reaches a more obvious level and is generally larger than 50 pC), keeping the experimental voltage stable, and collecting partial discharge data II.

S4: and acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the characteristic identification model.

And acquiring the partial discharge data II and the current experimental voltage Um after normalization processing, substituting the partial discharge data II and the current experimental voltage Um into the characteristic identification model obtained in the step S2, and calculating the defect type of the voltage transformer to be detected.

In the later period, if the user acquires the accurate fault type of the voltage transformer to be detected through other modes (such as dissection, X-ray fluoroscopy and the like), the data tested at this time can be marked according to the fault type, so that the number of sample libraries is increased. And after the sample library is updated, the step S2 is executed again to update the data model.

It should be noted that, the modules are arranged according to a streaming layout, which is only one embodiment of the present invention, and may also be arranged in other manners, and the present invention is not limited to this.

Fig. 4 is a block diagram of a partial discharge fault diagnosis apparatus for a voltage transformer according to an embodiment of the present invention, where the apparatus includes:

the first acquisition module is used for acquiring partial discharge data I of various defect models under a withstand voltage test and acquiring characteristic data of each defect model according to the partial discharge data I;

the establishing module is used for correspondingly establishing a characteristic identification model according to the characteristic data and the defect type of the defect model;

the acquisition module I is used for acquiring partial discharge data II of the voltage transformer to be detected under the withstand voltage test;

and the acquisition module is used for acquiring the defect type of the voltage transformer to be detected according to the partial discharge data II and the feature identification model.

The functions of the modules in the embodiment of fig. 4 correspond to the contents in the corresponding method embodiment, and are not described again here.

Fig. 5 shows a schematic structural diagram of the electronic device 50 according to an embodiment of the present invention, where the electronic device 50 includes at least one processor 501 (e.g., a CPU), at least one input/output interface 504, a memory 502, and at least one communication bus 503, which are used to implement connection communication between these components. The at least one processor 501 is configured to execute computer instructions stored in the memory 502 to enable the at least one processor 501 to perform any of the previously described embodiments of the diagnostic method. The Memory 502 is a non-transitory Memory (non-transitory Memory), which may include a volatile Memory such as a high-speed Random Access Memory (RAM) or a non-volatile Memory such as at least one disk Memory. A communication connection with at least one other device or unit is made via at least one input-output interface 504, which may be a wired or wireless communication interface.

In some embodiments, the memory 502 stores a program 5021, and the processor 501 executes the program 5021 to perform any of the above-described embodiments of the table splitting method.

The electronic device may exist in a variety of forms, including but not limited to:

(1) a mobile communication device: such devices are characterized by mobile communications capabilities and are primarily targeted at providing voice, data communications. Such terminals include: smart phones (e.g., iphones), multimedia phones, functional phones, and low-end phones, among others.

(2) Ultra mobile personal computer device: the equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include: PDA, MID, and UMPC devices, etc., such as ipads.

(3) A portable entertainment device: such devices can display and play multimedia content. This type of device comprises: audio, video players (e.g., ipods), handheld game consoles, electronic books, and smart toys and portable car navigation devices.

(4) The specific server: the device for providing the computing service comprises a processor, a hard disk, a memory, a system bus and the like, and the server is similar to a general computer architecture, but has higher requirements on processing capacity, stability, reliability, safety, expandability, manageability and the like because of the need of providing high-reliability service.

(5) And other electronic equipment with data interaction function.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments.

In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof.

The above description is only for the specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.