CN115436044A - On-load tap-changer mechanical fault diagnosis method and device and electronic equipment - Google Patents

Abstract

A method, device and electronic equipment for diagnosing a mechanical fault of an on-load tap-changer, the method comprising collecting vibration signals during the switching process of the on-load tap-changer to be diagnosed; extracting the time-frequency characteristics of the vibration signals through short-time Fourier transform; Dynamic time warping algorithm to calculate the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library. Different characteristic diagnosis sequences correspond to different on-load tap-changers Switch status; analyze the fault type and fault degree of the on-load tap-changer to be diagnosed according to the degree of difference. The invention adopts the dynamic time warping algorithm to calculate the difference degree between the time-frequency characteristics of the vibration signal of the on-load tap changer and the different characteristic diagnosis sequences in the mechanical fault diagnosis library, and can dynamically match the vibration signal to be diagnosed with the mechanical fault diagnosis The vibration signals in the library are locally most similarly matched, the fault type diagnosis accuracy is higher, and the fault degree can be analyzed.

Description

Method, device and electronic equipment for mechanical fault diagnosis of on-load tap changer

technical field

The invention relates to the technical field of fault diagnosis of transformer equipment, in particular to a mechanical fault diagnosis method, device and electronic equipment of an on-load tap changer.

Background technique

In the prior art, the on-load tap changer, as a voltage regulating device commonly used in power transformers, undertakes important tasks such as stabilizing the load center voltage and regulating reactive power flow in the power system. Due to frequent switching and complex working conditions, on-load tap-changers have gradually become one of the parts with the highest fault ratio in transformers, and mechanical faults are the main fault types of on-load tap-changers. Therefore, realizing accurate diagnosis of mechanical faults of on-load tap-changers is of great significance to the safe and stable operation of power systems. According to the switching principle of the on-load tap-changer, the switching process is completed by the action and separation of several pairs of contacts in sequence under the drive of the energy storage spring. The vibration signal generated in this process contains a large number of on-load tap-changer Therefore, the fault diagnosis of on-load tap-changers through vibration signal analysis is a widely used and feasible method at present.

However, there are two deficiencies in the existing on-load tap-changer mechanical fault diagnosis methods. First of all, as mentioned above, the vibration signal of the on-load tap-changer is generated by several pairs of contacts in sequence, which has obvious timing, but the existing methods do not take this characteristic into account, which reduces the fault diagnosis. accuracy. Secondly, the existing fault diagnosis methods can only realize the identification of the fault type, but cannot effectively characterize the severity of the fault. Therefore, it is necessary to provide a new method for diagnosing mechanical faults of on-load tap-changers to solve the problems of insufficient identification accuracy of fault types and failure to characterize the severity of faults in the prior art.

Contents of the invention

The technical problem to be solved by the present invention is to provide a method, device and electronic equipment for diagnosing mechanical faults of on-load tap-changers with higher accuracy and effectively characterizing the severity of faults.

In order to achieve the above purpose, based on one aspect of the present invention, a method for diagnosing mechanical faults of on-load tap-changers is provided, including:

Collect vibration signals during switching of on-load tap-changers to be diagnosed;

Extracting the time-frequency characteristics of the vibration signal by short-time Fourier transform;

Through the dynamic time warping algorithm, the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library is calculated, and the different characteristic diagnosis sequences correspond to Characterization of different on-load tap-changer states;

According to the degree of difference, the fault type and fault degree of the on-load tap-changer to be diagnosed are analyzed.

In one of the embodiments, the extracting the time-frequency features of the vibration signal through short-time Fourier transform includes:

Taking 10% to 20% of the length of the vibration signal of the on-load tap-changer to be diagnosed as the window length, taking 20% to 30% of the window length as the step size, according to the preset short-time Fourier The transformed window function intercepts the vibration signal;

The frequency band of each window signal of the intercepted vibration signal is divided into three frequency bands of low frequency, intermediate frequency and high frequency, and the signal energy size in each frequency band is calculated;

According to the calculated signal energy in each frequency band, the i-th window signal of the vibration signal is characterized by E _i =[E _{i low frequency} , E _{i intermediate frequency} , E _{i high frequency} ], and the vibration signal The time-frequency feature of is represented by an energy matrix with the number of rows equal to the total number of windows and the number of columns equal to three columns.

In one of the embodiments, in the mechanical fault diagnosis library, each of the characteristic diagnosis sequences is obtained by calculation of a plurality of vibration signal samples corresponding to the state of the on-load tap changer, and the calculation is based on dynamic time warping The algorithm calculates the center-of-gravity sequence of the time-frequency characteristic sequence of the plurality of vibration signal samples, and the dynamic time warping algorithm is used to calculate the time-frequency characteristic of the on-load tap-changer to be diagnosed and the preset mechanical fault diagnosis library The degree of difference between each characteristic diagnostic sequence in , including:

For any characteristic diagnosis sequence in the mechanical fault diagnosis library, calculate the characteristic diagnosis sequence C={c ₁ ,c ₂ ,...,ci ,...,c _{m }} that is related to the load tap to be diagnosed The Euclidean distance between any two elements in the feature sequence Y={y ₁ ,y ₂ ,…,y _j ,…,y _n } of the switch time-frequency feature to obtain the distance matrix d,

Among them, m represents the number of windows of the characteristic diagnosis sequence C, n represents the number of windows of the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed, and d(i, j) represents the i-th window signal of the characteristic diagnosis sequence C The distance between c _i and the _jth window signal yj of the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed,

Based on the calculated distance matrix d, calculate the cumulative distance matrix W, where any element W(i,j) in the cumulative distance matrix W is obtained by the following formula:

W(i,j)=d(i,j)+min(W(i-1,j-1),W(i,j-1),W(i-1,j));

According to the calculated cumulative distance matrix W, the difference DTW(C, Y) between the characteristic diagnostic sequence C and the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed is obtained:

DTW(C,Y)=W(m,n).

In one of the embodiments, the dynamic time warping algorithm is used to calculate the degree of difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library Previously, also included:

Perform multiple measurements on the vibration signals of the on-load tap-changer when switching in each state, so as to obtain multiple vibration signal samples corresponding to each state of the on-load tap-changer;

The plurality of vibration signal samples corresponding to each state of the on-load tap changer are respectively extracted by short-time Fourier transform, so as to obtain the time corresponding to the plurality of vibration signal samples in each state of the on-load tap-changer Frequency feature sequence;

The time-frequency characteristic sequence in each state of the on-load tap changer is calculated respectively by a dynamic time warping algorithm, so as to obtain the barycenter sequence corresponding to the time-frequency characteristic sequence in each state of the on-load tap-changer ;

Each of the calculated center-of-gravity sequences is associated with the corresponding on-load tap-changer state as a characteristic diagnosis sequence, and a mechanical fault diagnosis library of the on-load tap-changer is established.

In one of the embodiments, the time-frequency feature sequences of the multiple vibration signal samples representing the same on-load tap-changer state jointly constitute the corresponding vibration signal time-frequency feature set of the on-load tap-changer state, for any Each state of the on-load tap-changer has a corresponding time-frequency feature set of the vibration signal, and the dynamic time warping algorithm is used to calculate the time-frequency feature sequence in each state of the on-load tap-changer to obtain the active The center of gravity sequence corresponding to the time-frequency characteristic sequence under each state of the on-load tap changer includes:

For the vibration signal time-frequency feature set in any state of the on-load tap-changer, arbitrarily select a time-frequency feature sequence in the vibration signal time-frequency feature set as the initial center-of-gravity sequence;

Using a dynamic time warping algorithm to calculate a warped path sequence between the barycenter sequence and each time-frequency feature sequence in the vibration signal time-frequency feature set;

Using each obtained regularized path sequence to form a regularized path sequence set, the regularized path sequence in the regularized path sequence set corresponds to the time-frequency feature sequence in the vibration signal time-frequency feature set;

A new center-of-gravity sequence is calculated according to the regularized path sequence set, and any element in the new center-of-gravity sequence is calculated by the following formula:

Among them, c' _t represents the tth element of the new center of gravity sequence C', and s' _it represents the element of the tth match with the center of gravity sequence C in the ith regularized path sequence S' _i of the regularized path sequence set S' , N represents the total number of regularized path sequences in S', and α represents the total number of elements that perform the t-th match with the center of gravity sequence C in all N regularized path sequences of the regularized path sequence set S';

The new center-of-gravity sequence is used as the new center-of-gravity sequence, and iterative calculation is performed until the sum of the differences between the obtained new center-of-gravity sequence and each time-frequency feature sequence in the time-frequency feature set of the vibration signal is compared with the new center-of-gravity sequence and vibration The sum of the difference degrees of each time-frequency feature sequence in the signal time-frequency feature set, until the difference value does not exceed the preset ratio. This ratio is generally preset at 5% to 10%.

In one of the embodiments, the use of the dynamic time warping algorithm to separately calculate the warped path sequence between the center of gravity sequence and each time-frequency feature sequence in the vibration signal time-frequency feature set includes:

For any time-frequency feature sequence in the time-frequency feature set of the vibration signal, calculate the Euclidean distance between any two elements in the time-frequency feature sequence and the barycenter sequence to obtain a distance matrix;

Calculate a cumulative distance matrix based on the calculated distance matrix;

According to the calculated cumulative distance matrix, the normalized path sequence corresponding to the time-frequency feature sequence and the degree of difference between the time-frequency feature sequence and the center-of-gravity sequence are acquired.

In one of the embodiments, the analysis of the fault type and fault degree of the on-load tap-changer to be diagnosed according to the degree of difference includes:

Determining the state of the on-load tap-changer corresponding to the characteristic diagnosis sequence with the smallest time-frequency characteristic difference of the on-load tap-changer to be diagnosed as the current state of the on-load tap-changer to be diagnosed;

The fault degree of the on-load tap-changer to be diagnosed is determined according to the degree of difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequence representing the normal state of the on-load tap-changer.

Based on the same inventive concept, the present invention also provides a mechanical fault diagnosis device for on-load tap-changers, including:

The vibration signal collection module is used to collect vibration signals during the switching process of the on-load tap-changer to be diagnosed;

The time-frequency feature extraction module is used to extract the time-frequency feature of the vibration signal by short-time Fourier transform;

The difference degree calculation module is used to calculate the difference degree between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library through a dynamic time warping algorithm, different The characteristic diagnostic sequence corresponding to characterize different on-load tap-changer states;

The switch fault analysis module is used to analyze the fault type and fault degree of the on-load tap changer to be diagnosed according to the degree of difference.

Based on the same inventive concept, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor. When the processor executes the computer program, A method for diagnosing a mechanical fault of an on-load tap changer described in any one of the above items is realized.

Based on the same inventive concept, the present invention also provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to make the computer perform the mechanical failure of the on-load tap changer as described in any one of the above diagnosis method.

The on-load tap-changer mechanical fault diagnosis method, device and electronic equipment provided by the present invention use the center-of-gravity sequence corresponding to the vibration signal of the on-load tap-changer switching in each state calculated by the dynamic time warping algorithm as the mechanical fault The characteristic diagnosis sequence in the diagnostic library takes the timing of the vibration signal when the on-load tap-changer is switched into consideration, and uses the dynamic time warping algorithm to calculate the time-frequency characteristics of the vibration signal of the on-load tap-changer and the mechanical fault diagnosis library. The degree of difference between the characteristic diagnostic sequences of vibration signals representing different states of on-load tap-changers can be used to perform local most similar matching of vibration signals in a dynamic matching manner, thereby improving the accuracy of difference degree calculation; using dynamic time warping center of gravity The average algorithm calculates the center-of-gravity sequence corresponding to the vibration signal of the on-load tap-changer switching in each state, and uses the center-of-gravity sequence as the time-frequency characteristic parameter of the final diagnosis, which avoids the influence of random errors in single-sample measurement and improves the fault diagnosis sample rate. The accuracy rate of the sequence; and, on the basis of the fault type diagnosis of the on-load tap-changer, through the quantitative index of difference degree, the severity of the fault can also be analyzed, and more fault information can be provided.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

Fig. 1 is a vibration signal diagram under the normal state of the on-load tap-changer in an embodiment of the present invention;

Fig. 2 is a schematic flowchart of a method for diagnosing a mechanical fault of an on-load tap-changer according to an embodiment of the present invention;

Fig. 3 is a time-frequency energy characteristic diagram of a vibration signal of an on-load tap-changer in a normal state in an embodiment of the present invention;

Fig. 4 is a time-frequency energy characteristic diagram of the vibration signal of the on-load tap-changer to be diagnosed extracted by short-time Fourier transform in an embodiment of the present invention;

Fig. 5 is a structural schematic diagram of an embodiment of an on-load tap-changer mechanical fault diagnosis device of the present invention;

FIG. 6 is a schematic diagram of an electronic device in an embodiment of the present invention.

detailed description

In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

The invention provides a method for diagnosing mechanical faults of on-load tap-changers. The vibration signal of the on-load tap-changer switching process under normal conditions is shown in Figure 1. It can be seen that, corresponding to the switching principle of the on-load tap-changer, its There are multiple sequentially arranged vibration peaks in the vibration signal. Based on this, as shown in Figure 2, in this embodiment, the method for diagnosing mechanical faults of the on-load tap-changer includes the following steps:

S1: Collect vibration signals during switching of on-load tap-changers to be diagnosed.

Specifically, in the actual operation process, the on-load tap-changer includes normal state and mechanical fault state, and the mechanical fault of the on-load tap-changer includes but not limited to: typical mechanical faults such as loose contacts, worn contacts, insufficient spring energy storage, etc. , when the on-load tap-changer is in different fault states, the vibration signal during the switching process is also different. When measuring the vibration signal of the on-load tap-changer, it is preferable to arrange the measuring sensor on the top flange of the on-load tap-changer position, to avoid the difference in the installation position of the on-load tap-changer in the transformer oil tank from affecting the vibration signal measurement.

S2: Extract the time-frequency features of the vibration signal through short-time Fourier transform.

In this embodiment, extracting the time-frequency characteristics of the vibration signal through short-time Fourier transform specifically includes taking 10% to 20% of the length of the vibration signal of the on-load tap-changer to be diagnosed as the window length, and using 20% of the window length % to 30% as the step size, the vibration signal is intercepted according to the short-time Fourier transform, and then, as shown in Figure 3, the frequency band of each window signal of the intercepted vibration signal is divided into low frequency, intermediate frequency and high frequency. frequency band, calculate the signal energy size in each frequency band, then, according to the signal energy size in each frequency band calculated, use the i-th window signal of the vibration signal with E _i =[E _{i low frequency} , E _{i intermediate frequency} , E _{i High frequency} ] for characterization, the time-frequency characteristics of the vibration signal are characterized by an energy matrix P whose number of rows is equal to the number of windows and the number of columns is equal to three columns:

Wherein, T represents the total number of windows of the vibration signal, E _{T, low frequency} represents the low frequency band signal energy of the T window signal, E _{T, intermediate frequency} represents the mid frequency band signal energy of the T window signal, E _{T, high frequency} represents The high-frequency band signal energy of the T-th window signal.

In this embodiment, the short-time Fourier transform is used to obtain the frequency distribution of the vibration signal over time, and the energy of the distribution feature in different frequency bands is calculated over time, so as to realize the compression of the feature dimension and the time of the vibration signal sample. The frequency characteristics are represented by time-frequency energy characteristics. The time-frequency energy characteristics of the vibration signal of the on-load tap-changer under normal working conditions are shown in Figure 3, and the time-frequency energy characteristics of the vibration signal of the on-load tap-changer to be diagnosed are shown in Figure 4. .

S3: Through the dynamic time warping algorithm, calculate the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library. Different characteristic diagnosis sequences correspond to different load tap-changer status.

In this embodiment, in the mechanical fault diagnosis library, each characteristic diagnosis sequence is obtained by calculating a plurality of vibration signal samples corresponding to the state of the on-load tap changer, and the calculation is to use a dynamic time warping algorithm to calculate a plurality of vibration signal samples The barycentric sequence of the time-frequency feature sequence of .

Specifically, before calculating the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library through the dynamic time warping algorithm, it also includes establishing the on-load tap-changer Mechanical fault diagnosis library.

Firstly, the vibration signals of the on-load tap-changer when switching in each state are measured several times, so as to obtain a plurality of vibration signal samples corresponding to each state of the on-load tap-changer. In this embodiment, the state of the on-load tap-changer includes four types: normal state, loose contact, wear of the contact, and insufficient spring energy storage. Multiple vibration signal samples for each of the four states of loose head, worn contact, and insufficient spring energy storage. By measuring the vibration signal of the same state of the on-load tap changer multiple times, random factors in a single measurement can be avoided, and influence on subsequent diagnosis can be avoided.

Then, the multiple vibration signal samples corresponding to each state of the on-load tap-changer are extracted respectively by short-time Fourier transform, so as to obtain the time-frequency feature sequence corresponding to the multiple vibration signal samples under each state of the on-load tap-changer, The time-frequency feature sequences of a plurality of vibration signal samples representing the same on-load tap-changer state jointly constitute a time-frequency feature set of vibration signal corresponding to the on-load tap-changer state. In this embodiment, for the extraction of the time-frequency characteristics of multiple vibration signal samples, the same as the extraction of the time-frequency characteristics of the on-load tap-changer to be diagnosed, the time-frequency characteristics of the vibration signal are extracted by short-time Fourier transform, Divide the frequency band of the intercepted vibrating window signal into three frequency bands of low frequency, intermediate frequency and high frequency, calculate the signal energy in each frequency band, and then divide the vibration signal according to the calculated signal energy in each frequency band The time-frequency feature is represented by an energy matrix with the number of rows equal to the total number of windows and the number of columns equal to three columns (low frequency, intermediate frequency, high frequency).

The time-frequency characteristic sequence in each state of the on-load tap-changer is calculated by the dynamic time warping algorithm to obtain the barycentric sequence corresponding to the time-frequency characteristic sequence in each state of the on-load tap-changer.

Specifically, the time-frequency feature sequences of multiple vibration signal samples representing the same on-load tap-changer state together constitute the corresponding time-frequency feature set of the vibration signal of the on-load tap-changer state, for any on-load tap-changer state There are corresponding vibration signal time-frequency feature sets. In this embodiment, the status of the on-load tap-changer includes four types: normal status, loose contact, worn contact, and insufficient spring energy storage. Therefore, there are four corresponding vibration signals. frequency feature set.

The time-frequency characteristic sequence of the on-load tap-changer in the normal state, loose contact, wear of the contact, and insufficient spring energy storage is calculated by the dynamic time warping algorithm, so as to obtain the The barycentric sequence corresponding to the time-frequency feature sequence of , specifically includes:

First, for the vibration signal time-frequency feature set S={S ₁ , S ₂ ,…,S _i ,…,S _N } in any state of the on-load tap-changer, N is the total number of time-frequency feature sequences in S, Randomly select one of the time-frequency feature sequences S _i ={s _i1 ,s _i2 ,…,s _ik ,…,s _iT } as the initial barycentric sequence C={c ₁ ,c ₂ ,…,c _t ,…, c _T }. Based on the extraction method of time-frequency features of multiple vibration signal samples, it can be known that for any element s _ik in S _i , it is expressed as E _ik =[E _{ik low frequency} , E _{ik intermediate frequency} , E _{ik high frequency} ].

Then, use the dynamic time warping algorithm to calculate the warped path sequence S' _i ={s' _i1 ,s' _i2 ,...,s ' _ik ,...,s' _iT }.

Then, use each obtained regularized path sequence S' _i to form a regularized path sequence set S'={S' ₁ , S' ₂ ,...,S' _i ,...,S' _N }, in the regularized path sequence set S' The regular path sequence S' _i of the vibration signal corresponds to the time-frequency feature sequence S _i in the time-frequency feature set S of the vibration signal;

Then, calculate the new center-of-gravity sequence C'={c' ₁ ,c' ₂ ,...,c' _t ,...,c' _T } according to the obtained regular path sequence set S', any element in the new center-of-gravity sequence C' Calculated by the following formula:

Among them, c' _t represents the tth element of the new center of gravity sequence C', and s' _it represents the element of the tth match with the center of gravity sequence C in the ith regularized path sequence S' _i of the regularized path sequence set S' , N represents the total number of regularized path sequences in S', and α represents the total number of elements that perform the t-th match with the center of gravity sequence C in all N regularized path sequences of the regularized path sequence set S';

The new center-of-gravity sequence C' is used as the new center-of-gravity sequence C, and the iterative calculation is performed until the sum of the differences between the obtained new center-of-gravity sequence C' and each time-frequency feature sequence S _i in the vibration signal time-frequency feature set S, compared to The sum of the differences between the new centroid sequence of the last iteration and each time-frequency feature sequence Si in the time-frequency feature set S of the vibration signal, _until the difference does not exceed 5%.

Finally, each calculated gravity center sequence is associated with the corresponding on-load tap-changer state as a characteristic diagnosis sequence, and a mechanical fault diagnosis library of on-load tap-changer is established.

In this embodiment, the dynamic time warping algorithm is used to calculate the _{warped path sequence S' i ={s' i1 , s} ' _i2 ,…,s' _iT }, including:

For any time-frequency feature sequence S _i in the time-frequency feature set of the vibration signal, calculate the time-frequency feature sequence S _i ={s _i1 ,s _i2 ,…,s _ik ,…,s _iT } and the centroid sequence C={c ₁ ,c ₂ ,…,c _t ,…,c _T } the Euclidean distance between any two elements in order to obtain the distance matrix D,

Among them, D(k, t) represents the distance between the k-th window signal s _ik of the feature sequence S _i and the t-th window signal c _t of the initial center-of-gravity sequence C.

Based on the calculated distance matrix D, calculate the cumulative distance matrix W', where any element W'(k,t) in the cumulative distance matrix W' is obtained by the following formula:

W'(k,t)=D(k,t)+min(W'(k-1,t-1), W'(k,t-1), W'(k-1,t));

According to the calculated cumulative distance matrix W', the regularized path sequence S' _i and the difference dtw(S _i , C) between the time-frequency feature sequence S _i and the barycenter sequence C are obtained.

In this embodiment, the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library is calculated through the dynamic time warping algorithm, including:

For any characteristic diagnosis sequence in the mechanical fault diagnosis library, calculate the characteristics of the time-frequency characteristics of the on-load tap-changer to be diagnosed in the characteristic diagnosis sequence C={c ₁ ,c ₂ ,…,ci ,…,c _{m }} The Euclidean distance between any two elements in the sequence Y={y ₁ ,y ₂ ,…,y _j ,…,y _n } to obtain the distance matrix d,

Among them, m represents the number of windows of the characteristic diagnosis sequence C, n represents the number of windows of the characteristic sequence Y of the time-frequency characteristics of the on-load tap-changer to be diagnosed, d(i, j) represents the i-th window signal c _i of the characteristic diagnosis sequence C The distance between the jth window signal y _j of the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed,

Based on the calculated distance matrix d, calculate the cumulative distance matrix W, where any element W(i, j) in the cumulative distance matrix W is obtained by the following formula:

W(i,j)=d(i,j)+min(W(i-1,j-1),W(i,j-1),W(i-1,j));

According to the calculated cumulative distance matrix W, the difference degree DTW(C, Y) between the characteristic diagnosis sequence C and the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed is obtained:

DTW(C,Y)=W(m,n).

S4: Analyze the fault type and fault degree of the on-load tap-changer to be diagnosed according to the degree of difference.

Specifically, the state of the on-load tap-changer corresponding to the characteristic diagnosis sequence with the smallest time-frequency characteristic difference degree of the on-load tap-changer to be diagnosed is determined as the current state of the on-load tap-changer to be diagnosed, and then, according to the Diagnose the degree of difference between the time-frequency characteristics of the on-load tap-changer and the characteristic diagnosis sequence representing the normal state of the on-load tap-changer, and determine the fault degree of the on-load tap-changer to be diagnosed.

Compare the time-frequency characteristics of the vibration signal of the on-load tap-changer to be diagnosed with the fault diagnosis sequence corresponding to each state of the on-load tap-changer in the mechanical fault diagnosis library according to the above-mentioned difference calculation method, and calculate the The degree of difference between the time-frequency characteristics of the vibration signal of the on-load tap-changer and the time-frequency characteristics (fault diagnosis sequence) corresponding to the state of each on-load tap-changer in the mechanical fault diagnosis library, and compare the size of the different degrees of difference, and the degree of difference is the smallest The state corresponding to the fault diagnosis sequence is the state of the on-load tap-changer to be diagnosed, that is, the corresponding state in the mechanical fault diagnosis library according to the fault diagnosis sequence with the smallest time-frequency characteristic difference with the on-load tap-changer to be diagnosed , to determine the fault type of the on-load tap-changer to be diagnosed. At the same time, according to the difference between the vibration signal of the on-load tap-changer to be diagnosed and the vibration signal of the on-load tap-changer in the normal working state, the severity of the fault of the on-load tap-changer to be diagnosed is analyzed. increase, the difference value will also increase, that is, according to the time-frequency characteristics of the on-load tap-changer to be diagnosed and the time-frequency characteristics corresponding to the normal working state of the on-load tap-changer in the preset mechanical fault diagnosis library (fault diagnosis sequence) The magnitude of the difference degree determines the fault degree of the on-load tap-changer to be diagnosed, so as to realize the type diagnosis and fault degree analysis of the mechanical fault of the on-load tap-changer.

The method for diagnosing mechanical faults of on-load tap-changers provided by the present invention uses the center-of-gravity sequence corresponding to the vibration signal of the on-load tap-changer switched in each state calculated by the dynamic time warping algorithm as a feature in the mechanical fault diagnosis library Diagnosis sequence, taking into account the timing of the vibration signal when the on-load tap-changer is switched, the dynamic time warping algorithm is used to calculate the time-frequency characteristics of the vibration signal of the on-load tap-changer, which is different from the representation in the mechanical fault diagnosis library. The degree of difference between the characteristic diagnosis sequences of the vibration signal in the switch state can be dynamically matched to the local most similar matching of the vibration signal, thereby improving the accuracy of the difference degree calculation; the dynamic time warping center of gravity average algorithm is used to calculate the load The center-of-gravity sequence corresponding to the vibration signal switched by the tap changer in each state, the center-of-gravity sequence is used as the time-frequency characteristic parameter of the final diagnosis, which avoids the influence of random error in single sample measurement and improves the accuracy of fault diagnosis; and, On the basis of the type diagnosis of on-load tap-changer faults, through the quantitative index of difference degree, the severity of faults can also be analyzed, and more fault information can be provided.

As shown in Figure 5, based on the same inventive concept, and corresponding to the methods in the above embodiments, an embodiment of the present invention also provides a mechanical fault diagnosis device for on-load tap-changers, which is used to realize the active On-load tap-changer mechanical fault diagnosis method, including:

The vibration signal collection module 10 is used to collect vibration signals during the switching process of the on-load tap-changer to be diagnosed;

Time-frequency feature extraction module 20, for extracting the time-frequency feature of vibration signal by short-time Fourier transform;

The difference degree calculation module 30 is used to calculate the difference degree between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library through the dynamic time warping algorithm. Different characteristic diagnosis Sequence correspondence characterizes different on-load tap-changer states;

The switch fault analysis module 40 is configured to analyze the fault type and fault degree of the on-load tap changer to be diagnosed according to the degree of difference.

The apparatuses in the foregoing embodiments are used to implement the corresponding methods in the foregoing embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

Based on the same inventive concept, and corresponding to the method in any of the above embodiments, 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 operable on the processor. When the processor executes the program, the method for diagnosing the mechanical fault of the on-load tap changer described in the above embodiment is realized.

FIG. 6 shows a more specific hardware schematic diagram of an electronic device provided by this embodiment. The device may include: a processor 100 , a memory 200 , an input/output interface 300 , a communication interface 400 and a bus 500 . The processor 100, the memory 200, the input/output interface 300, the communication interface 400, and the bus 500 implement communication connections among each other within the device.

The processor 100 may be implemented by a general-purpose CPU (Central Processing Unit, central processing unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute related programs to realize the technical solutions provided by the embodiments of the present invention.

The memory 200 may be implemented in the form of ROM (Read Only Memory, read only memory), RAM (Random Access Memory, random access memory), static storage device, dynamic storage device, and the like. The memory 200 can store an operating system and other application programs. When implementing the technical solutions provided by the embodiments of the present invention through software or firmware, related program codes are stored in the memory 200 and invoked by the processor 100 for execution.

The input/output interface 300 is used to connect the input/output module to realize information input and output. The input/output/module can be configured in the device as a component (not shown in the figure), or can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, mouse, touch screen, microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, and the like.

The communication interface 400 is used to connect a communication module (not shown in the figure), so as to realize the communication interaction between the device and other devices. The communication module can realize communication through wired means (such as USB, network cable, etc.), and can also realize communication through wireless means (such as mobile network, WIFI, Bluetooth, etc.).

Bus 500 includes a path for transferring information between various components of the device (eg, processor 100, memory 200, input/output interface 300, and communication interface 400).

It should be noted that although the above device only shows the processor 100, the memory 200, the input/output interface 300, the communication interface 400, and the bus 500, in the specific implementation process, the device may also include other components. In addition, those skilled in the art can understand that the above-mentioned device may only include components necessary to implement the solutions of the embodiments of this specification, and does not necessarily include all the components shown in the figure.

Based on the same inventive concept, corresponding to the method in any of the above embodiments, an embodiment of the present invention further provides a computer-readable storage medium, the computer-readable storage medium stores computer instructions, and the computer instructions are used to make the computer Execute the method for diagnosing mechanical faults of the on-load tap changer as described in the above-mentioned embodiments.

The computer-readable storage medium in this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be realized by any method or technology; the information can be computer-readable instructions, data structures, program modules or other data. Examples of computer readable 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 Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridge, magnetic tape disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information that can be accessed by computer equipment.

The computer instructions stored in the computer storage medium of the above embodiment are used to make the computer execute the method for diagnosing mechanical faults of the on-load tap changer described in the above embodiment, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

Those of ordinary skill in the art should understand that: the discussion of any of the above embodiments is exemplary only, and is not intended to imply that the scope of the present invention is limited to these examples; under the idea of the present invention, the techniques in the above embodiments or different embodiments Features may also be combined, steps may be performed in any order, and there are many other variations of the different aspects of the embodiments of the invention as described above, which are not presented in detail for the sake of brevity.

The embodiments of the present invention are intended to cover all such alternatives, modifications and variations that fall within the broad scope of the present invention. Therefore, any omission, modification, equivalent replacement, improvement, etc. made within the spirit and principles of the embodiments of the present invention shall fall within the protection scope of the present invention.

Claims (10)

1. A method for diagnosing mechanical faults of an on-load tap changer, characterized in that it comprises: Collect vibration signals during switching of on-load tap-changers to be diagnosed; Extracting the time-frequency characteristics of the vibration signal by short-time Fourier transform; Through the dynamic time warping algorithm, the difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library is calculated, and the different characteristic diagnosis sequences correspond to Characterization of different on-load tap-changer states; According to the degree of difference, the fault type and fault degree of the on-load tap-changer to be diagnosed are analyzed. 2. The on-load tap-changer mechanical fault diagnosis method as claimed in claim 1, wherein said extracting the time-frequency characteristics of said vibration signal by short-time Fourier transform comprises: Taking 10% to 20% of the length of the vibration signal of the on-load tap-changer to be diagnosed as the window length, taking 20% to 30% of the window length as the step size, according to the preset short-time Fourier The transformed window function intercepts the vibration signal; The frequency band of each window signal of the intercepted vibration signal is divided into three frequency bands of low frequency, intermediate frequency and high frequency, and the signal energy size in each frequency band is calculated; According to the calculated signal energy in each frequency band, the i-th window signal of the vibration signal is characterized by E _i =[E _{i low frequency} , E _{i intermediate frequency} , E _{i high frequency} ], and the vibration signal The time-frequency feature of is represented by an energy matrix with the number of rows equal to the total number of windows and the number of columns equal to three columns. 3. The method for diagnosing mechanical faults of on-load tap-changers as claimed in claim 1, wherein, in the database of mechanical fault diagnosis, each of the characteristic diagnostic sequences is composed of multiples of its corresponding on-load tap-changer states. Vibration signal samples are obtained by calculation, the calculation is to use a dynamic time warping algorithm to calculate the center-of-gravity sequence of the time-frequency feature sequence of the plurality of vibration signal samples, and the dynamic time warping algorithm is used to calculate the load points to be diagnosed The degree of difference between the time-frequency characteristics of the switch and each characteristic diagnosis sequence in the preset mechanical fault diagnosis library, including: For any characteristic diagnosis sequence in the mechanical fault diagnosis library, calculate the characteristic diagnosis sequence C={c ₁ ,c ₂ ,...,ci ,...,c _{m }} that is related to the load tap to be diagnosed The Euclidean distance between any two elements in the feature sequence Y={y ₁ ,y ₂ ,…,y _j ,…,y _n } of the switch time-frequency feature to obtain the distance matrix d, Among them, m represents the number of windows of the characteristic diagnosis sequence C, n represents the number of windows of the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed, and d(i, j) represents the i-th window signal of the characteristic diagnosis sequence C The distance between c _i and the _jth window signal yj of the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed,

Based on the calculated distance matrix d, calculate the cumulative distance matrix W, where any element W(i, j) in the cumulative distance matrix W is obtained by the following formula: W(i,j)=d(i,j)+min(W(i-1,j-1),W(i,j-1),W(i-1,j)); According to the calculated cumulative distance matrix W, the difference DTW(C, Y) between the characteristic diagnostic sequence C and the characteristic sequence Y of the time-frequency characteristic of the on-load tap-changer to be diagnosed is obtained: DTW(C,Y)=W(m,n). 4. The method for diagnosing mechanical faults of on-load tap-changers as claimed in claim 1, wherein the time-frequency characteristic and preset Before the degree of difference between each feature diagnosis sequence in the mechanical fault diagnosis library, it also includes: Perform multiple measurements on the vibration signals of the on-load tap-changer when switching in each state, so as to obtain multiple vibration signal samples corresponding to each state of the on-load tap-changer; The plurality of vibration signal samples corresponding to each state of the on-load tap changer are respectively extracted by short-time Fourier transform, so as to obtain the time corresponding to the plurality of vibration signal samples in each state of the on-load tap-changer Frequency feature sequence; The time-frequency characteristic sequence in each state of the on-load tap changer is calculated respectively by a dynamic time warping algorithm, so as to obtain the barycenter sequence corresponding to the time-frequency characteristic sequence in each state of the on-load tap-changer ; Each of the calculated center-of-gravity sequences is associated with the corresponding on-load tap-changer state as a characteristic diagnosis sequence, and a mechanical fault diagnosis library of the on-load tap-changer is established. 5. The method for diagnosing mechanical faults of on-load tap-changers as claimed in claim 4, wherein the time-frequency characteristic sequences representing the multiple vibration signal samples of the same on-load tap-changer state jointly constitute a corresponding on-load tap-changer The vibration signal time-frequency feature set of the tap changer state has a corresponding vibration signal time-frequency feature set for any on-load tap-changer state. The following time-frequency characteristic sequence is calculated to obtain the barycenter sequence corresponding to the time-frequency characteristic sequence in each state of the on-load tap-changer, including: For the vibration signal time-frequency feature set in any state of the on-load tap-changer, arbitrarily select a time-frequency feature sequence in the vibration signal time-frequency feature set as the initial center-of-gravity sequence; Using a dynamic time warping algorithm to calculate a warped path sequence between the barycenter sequence and each time-frequency feature sequence in the vibration signal time-frequency feature set; Using each obtained regularized path sequence to form a regularized path sequence set, the regularized path sequence in the regularized path sequence set corresponds to the time-frequency feature sequence in the vibration signal time-frequency feature set; A new center-of-gravity sequence is calculated according to the regularized path sequence set, and any element in the new center-of-gravity sequence is calculated by the following formula:

Among them, c't represents the tth element of the new center of gravity sequence C', and s'it represents the element of the tth match with the center of gravity sequence C in the ith regularized path sequence S' _i of the regularized path sequence set S' , N represents the total number of regularized path sequences in S', and α represents the total number of elements that perform the t-th match with the center of gravity sequence C in all N regularized path sequences of the regularized path sequence set S'; The new center-of-gravity sequence is used as the new center-of-gravity sequence, and iterative calculation is performed until the sum of the differences between the obtained new center-of-gravity sequence and each time-frequency feature sequence in the time-frequency feature set of the vibration signal is compared with the new center-of-gravity sequence and vibration The sum of the difference degrees of each time-frequency feature sequence in the signal time-frequency feature set, until the difference value does not exceed the preset ratio. 6. The method for diagnosing mechanical faults of on-load tap-changers as claimed in claim 5, wherein the use of a dynamic time warping algorithm calculates each time-frequency of the center-of-gravity sequence and the vibration signal time-frequency feature set respectively A sequence of regular paths between feature sequences, including: For any time-frequency feature sequence in the time-frequency feature set of the vibration signal, calculate the Euclidean distance between any two elements in the time-frequency feature sequence and the barycenter sequence to obtain a distance matrix; Calculate a cumulative distance matrix based on the calculated distance matrix; According to the calculated cumulative distance matrix, the normalized path sequence corresponding to the time-frequency feature sequence and the degree of difference between the time-frequency feature sequence and the center-of-gravity sequence are acquired. 7. The method for diagnosing mechanical faults of on-load tap-changers according to any one of claims 1-6, wherein the analysis of the fault type and fault degree of the on-load tap-changer to be diagnosed according to the degree of difference, include: Determining the state of the on-load tap-changer corresponding to the characteristic diagnosis sequence with the smallest time-frequency characteristic difference of the on-load tap-changer to be diagnosed as the current state of the on-load tap-changer to be diagnosed; The fault degree of the on-load tap-changer to be diagnosed is determined according to the degree of difference between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequence representing the normal state of the on-load tap-changer. 8. A mechanical fault diagnosis device for an on-load tap changer, characterized in that it comprises: The vibration signal collection module is used to collect the vibration signal during the switching process of the on-load tap-changer to be diagnosed; The time-frequency feature extraction module is used to extract the time-frequency feature of the vibration signal by short-time Fourier transform; The difference degree calculation module is used to calculate the difference degree between the time-frequency characteristics of the on-load tap-changer to be diagnosed and the characteristic diagnosis sequences in the preset mechanical fault diagnosis library through a dynamic time warping algorithm, different The characteristic diagnostic sequence corresponding to characterize different on-load tap-changer states; The switch fault analysis module is used to analyze the fault type and fault degree of the on-load tap changer to be diagnosed according to the degree of difference. 9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, characterized in that, when the processor executes the computer program, the computer program according to claim 1 is realized. The method for diagnosing mechanical faults of on-load tap-changers described in any one of 1 to 7. 10. A computer-readable storage medium, the computer-readable storage medium storing computer instructions, characterized in that the computer instructions are used to make a computer execute the loaded Diagnosis method for tap changer mechanical faults.

Images