A method for processing transient acoustic vibration signals of transformer

Abstract

The present invention relates to the technical field of fault diagnosis of electrical equipment, and in particular to a method for processing transient acoustic vibration signals of transformers. The present invention provides a method for processing transient acoustic vibration signals of transformers, including collecting transient acoustic vibration signals of transformers; performing fast Fourier transform on the obtained transient acoustic vibration signals of transformers; and performing time-frequency analysis on the transient acoustic vibration signals of transformers after Fourier transform using a wavelet transform algorithm to obtain time domain information of the signal and corresponding frequency domain information. The present invention can describe the frequency components existing in the transient acoustic vibration signals of transformers, and can also display the time when each frequency component appears, and can also avoid the problem that the relative amplitudes existing in wavelet time-frequency analysis are not intuitive.

Description

Technical Field

[0001] The invention relates to the technical field of electrical equipment fault diagnosis, and in particular to a method for processing transient acoustic vibration signals of a transformer. Background Art

[0002] When a transformer is short-circuited externally, the current flowing through the winding surges, which generates a huge electromagnetic impact force. The damage to the transformer may be caused by excessive electric force, which causes winding deformation and insulation damage at one time, or electric force acts on the winding multiple times, gradually accumulating and causing structural changes such as loosening of winding pads and winding deformation. Therefore, it is very important to study acoustic vibration signals. Traditional acoustic vibration signal processing methods have background noise, and non-steady-state signals have spectrum leakage due to the presence of mutation signals, resulting in errors in spectrum information, low frequency resolution and time resolution, and non-intuitive relative amplitudes. Summary of the invention

[0003] In order to solve the above problems, the present invention provides a method for processing transient acoustic vibration signals of transformers. When the mechanical state of the winding is intact, the influence of the nonlinear factors of the system is small, and the transient acoustic vibration signal of the winding is relatively simple, reflecting that the nonlinear vibration of the winding has different characteristics at different times. The specific technical solution is as follows:

[0004] A method for processing transient acoustic vibration signals of a transformer comprises the following steps:

[0005] Step S1, collecting transient acoustic vibration signals of the transformer;

[0006] Step S2, performing fast Fourier transform on the obtained transformer transient acoustic vibration signal;

[0007] Step S3, using a wavelet transform algorithm to perform time-frequency analysis on the transient acoustic vibration signal of the transformer after Fourier transform, to obtain the time domain information of the signal and the corresponding frequency domain information.

[0008] Preferably, in step S3, the transformer transient acoustic vibration signal is assumed to be w(t); its continuous wavelet transform is:

[0009]

[0010] a is the expansion factor, b is the translation factor;

[0011] The function ψ(a,b)(x) is a continuous wavelet function that depends on the parameter pair (a,b) and is generated by the wavelet mother function ψ(t), referred to as wavelet;

[0012]

[0013] Preferably, the wavelet coefficients are corrected, and the following correction function can be obtained by combining the scale transformation property of the continuous wavelet transform and the corresponding wavelet basis function parameters:

[0014]

[0015] Where W′ w is the corrected wavelet coefficient, W w is the wavelet coefficient before correction, f is the signal frequency, f s is the signal sampling frequency, f c is the center frequency of the wavelet basis function.

[0016] The beneficial effects of the present invention are as follows: the present invention provides a method for processing transient acoustic vibration signals of transformers, including collecting transient acoustic vibration signals of transformers; performing fast Fourier transform on the obtained transient acoustic vibration signals of transformers; and using a wavelet transform algorithm to perform time-frequency analysis on the transient acoustic vibration signals of transformers after Fourier transform, to obtain the time domain information of the signal and the corresponding frequency domain information. The present invention can describe the frequency components existing in the transient acoustic vibration signals of transformers, and can also display the time when each frequency component appears, and can also avoid the problem that the relative amplitude in wavelet time-frequency analysis is not intuitive. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0018] Figure 1 Current, vibration and sound pressure signals when the transformer is short-circuited;

[0019] Figure 2 is the time domain waveform of the analog signal, where Figure 2 (a) is the time domain waveform of the steady-state analog signal. Figure 2 (b) is the time domain waveform of the non-steady-state analog signal;

[0020] Figure 3 is the FFT spectrum of the analog signal, where Figure 3 (a) is the FFT spectrum of the steady-state analog signal. Figure 3 (b) is the FFT spectrum of the non-steady-state analog signal;

[0021] Figure 4 is the time-frequency diagram of the wavelet transform of the analog signal, where Figure 4 (a) is the time-frequency diagram of the wavelet transform of the steady-state analog signal. Figure 4(b) is the time-frequency diagram of the wavelet transform of the non-steady-state analog signal;

[0022] Figure 5 is the time-frequency diagram of the analog signal after wavelet transform correction, where Figure 5 (a) is the time-frequency diagram of the steady-state analog signal after wavelet transform correction. Figure 5 (b) is the time-frequency diagram of the non-steady-state analog signal after wavelet transformation correction;

[0023] Figure 6 Time-frequency diagram of transient acoustic vibration signal; Figure 6 (a) is the time-frequency diagram of the vibration signal. Figure 6 (b) is the time-frequency diagram of the acoustic signal. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 of 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.

[0025] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0026] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0027] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0028] A specific embodiment of the present invention provides a method for processing transient acoustic vibration signals of a transformer, characterized in that it includes the following steps:

[0029] Step S1, collecting transient acoustic and vibration signals of the transformer. When the transformer is subjected to an external short-circuit shock, the current flowing through the winding includes a periodic component and an exponential decay component. Therefore, the short-circuit current will first reach a maximum, then decay and tend to stabilize until the circuit breaker is activated to cut off the power supply. The vibration or acoustic signal contains constant terms, decay terms, current frequency periodic decay terms (50Hz), and double current frequency periodic terms (100Hz). Taking a 70% short-circuit shock on a 110kV transformer A phase high-to-center as an example, the time domain waveforms of the short-circuit current, vibration, and sound pressure signals are as follows: Figure 1 As shown, the short-circuit impact did not cause any damage to the winding. It can be found that:

[0030] (1) By controlling the closing phase, the asymmetric short-circuit current reaches its maximum value at the first peak. Under 70% of the specified current loading, it can reach 4424A, which is 16.89 times the rated current of 262A. Then it gradually decays to the symmetrical short-circuit current value of 1633, which is 6.23 times the rated current. After 0.24 seconds, the power supply is cut off and the current returns to zero. Under 100% of the specified current loading, the first peak can reach 6321A, which will have a huge impact on the electrical performance, thermal performance, mechanical performance, etc. of the winding;

[0031] (2) Since the electromotive force is proportional to the square of the current, the electromotive force on the winding during the entire short-circuit process can reach 39-285 times the electromotive force under steady-state conditions. Under 100% of the specified current loading, the maximum electromotive force can even reach 582 times. In this case, the vibration of the winding will be much greater than other vibration sources (iron core, cooling device, etc.). Therefore, the vibration and acoustic signals during the short-circuit impact can be regarded as signals emitted by the winding alone;

[0032] (3) Compared with the vibration signal in steady state, the vibration acceleration caused by short circuit impact can reach 10g level (g is a gravitational acceleration, about 10m / s 2 ), the maximum value of the vibration does not appear at the first peak of the current, but increases first and then decreases, presenting a "shuttle shape", and after the short-circuit current is cut off at 0.24 seconds, it does not disappear immediately, but turns from forced vibration to free vibration and then slowly decays, eventually running out of energy, lasting about 0.5 seconds;

[0033] (4) The acoustic pressure signal, as the propagation of vibration in the air, has a good correlation with the vibration of the oil tank surface, and the change trend is consistent with the vibration signal. In addition, it can be observed that the acoustic signal is not smooth before and after the short-circuit impact, and there is background noise, but the sound pressure value during the impact is much larger than the background noise, so these interferences can be ignored. Although noise is not suitable for fault diagnosis of transformer windings under steady-state conditions, it has a high utilization value during short-circuit impact. For convenience, in subsequent studies, the vibration acoustic signal of the transformer when it is subjected to a short-circuit impact is referred to as a transient acoustic vibration signal.

[0034] Step S2, perform fast Fourier transform on the obtained transformer transient acoustic vibration signal; after obtaining the time domain signal of transient acoustic vibration, the information contained therein is not intuitive, so the signal needs to be further processed. In traditional transformer vibration signal analysis, fast Fourier transform (FFT) is most commonly used for signal processing. However, for non-steady-state acoustic vibration signals, FFT has great limitations. To intuitively represent this limitation, the following two signals are simulated respectively:

[0035] x(t)=cos50πt+cos100πt+cos150πt+cos200πt; (1)

[0036]

[0037] Wherein, equation (1) is a steady signal, and equation (2) is a non-steady signal. Although both contain signals of four frequencies, the steady signal runs through the entire time domain, while the non-steady signal appears with different frequencies in sequence over time. Its time domain waveform is as follows: Figure 2 As shown, the sampling rate is 10240S / s and the number of sampling points is 10240. Perform FFT analysis on the two analog signals, as shown in Figure 3 The figure shows the FFT spectrum of an analog signal. For steady-state signals, FFT can obtain very ideal results. However, due to the presence of mutation signals in non-steady-state signals, spectrum leakage and other phenomena occur, which causes errors in spectrum information. In addition, FFT can only obtain the frequency components contained in a signal as a whole, but cannot obtain the specific time when each component appears. This also leads to the situation where different frequencies appear at different times but the spectrum information is consistent (for example, Figure 3 If the signal in (a) appears in sequence at 200Hz, 150Hz, 100Hz, and 50Hz, the spectrum diagram will be Figure 3 To avoid this situation, it is necessary to perform time-frequency analysis on the signal.

[0038] Step S3, using the wavelet transform algorithm to perform time-frequency analysis on the transient acoustic vibration signal of the transformer after Fourier transform, to obtain the time domain information of the signal and the corresponding frequency domain information. Among the signal time-frequency analysis methods, wavelet transform has been widely used due to its perfect theoretical basis and flexible wavelet function. Assume that the wavelet mother function is ψ(t), then its function space L 2 (R) satisfies formula (3):

[0039]

[0040] In the formula, R * represents all non-zero real numbers, is the Fourier transform of ψ(x).

[0041] Then the function ψ(a, b)(x) represented in equation (4) is called a continuous wavelet function that depends on the parameter pair (a, b) and is generated by the wavelet mother function ψ(t), or simply wavelet.

[0042]

[0043] Where a is called the expansion factor and b is called the translation factor.

[0044] For a signal w(x), its continuous wavelet transform (CWT) is defined as:

[0045]

[0046] The wavelet function cmor (complex Morlet wavelet) suitable for processing unstable mutation signals is selected as the wavelet basis. In order to obtain higher frequency resolution and time resolution, the center frequency is set to 5 and the bandwidth parameter is set to 1. The time-frequency diagram of the above simulation signal is as follows: Figure 4 As shown. It can be found that, whether it is a steady-state or non-steady-state signal, the wavelet time-frequency diagram can not only describe the frequency components that exist, but also show the time when each frequency component appears, which is very necessary for transient acoustic vibration signals. However, this method still has certain defects. The amplitude of each frequency component in the analog signal is all 1, but the wavelet coefficient in the time-frequency diagram is not the same as the vibration amplitude. In addition, since the energy of high frequency is dispersed in a wider frequency band, the higher the frequency, the smaller the wavelet coefficient, which is very unfavorable for transient acoustic vibration signals that focus on the signal amplitude, making the relative amplitude between different frequencies not intuitive.

[0047] In order to avoid the problem of unintuitive relative amplitude in wavelet time-frequency analysis, it is necessary to correct the wavelet coefficients. Combining the scale transformation properties of continuous wavelet transform and the corresponding wavelet basis function parameters, the following correction function can be obtained:

[0048]

[0049] Where W w ′ is the corrected wavelet coefficient, W w is the wavelet coefficient before correction, f is the signal frequency, f s is the signal sampling frequency, f c is the center frequency of the wavelet basis function.

[0050] The corrected wavelet time-frequency diagram is shown in Figure 5 As shown, it can be found that the wavelet coefficients of different frequencies can reflect their true amplitudes.

[0051] right Figure 1The transient acoustic vibration signal in is corrected by continuous wavelet transform, and its time-frequency diagram can be obtained as follows Figure 6 As shown, we can find that:

[0052] (1) The acoustic vibration signal mainly contains 50Hz and 100Hz components, which is consistent with the theoretical analysis. In addition, there are also smaller high-order harmonic components such as 150Hz and 200Hz, which are related to the nonlinearity of the system (superharmonic vibration, subharmonic vibration or parametric vibration). In addition, when the mechanical state of the winding is intact, the nonlinear factors of the system have little influence, and the transient acoustic vibration signal of the winding is relatively simple;

[0053] (2) The appearance time of different frequency components in the acoustic vibration signal is not the same. The time-frequency diagram can intuitively reflect the development process of each frequency during the entire short-circuit impact process. The first component to appear is the 100Hz vibration, followed by other components such as 50Hz, and the duration and amplitude appear at different times. This is because the mechanical state and force of the winding are dynamically changing during the entire short-circuit impact process, so the nonlinear vibration of the winding has different manifestations at different times;

[0054] (3) The acoustic signal and the vibration signal are similar, but the acoustic signal is relatively richer. This is because the acoustic sensor collects the signal radiated outward from the entire transformer tank, which can be regarded as a concentrated expression of the vibration information on the tank surface, while the vibration signal is a single-point signal.

[0055] The present invention provides a method for processing transient acoustic vibration signals of transformers, including collecting transient acoustic vibration signals of transformers; performing fast Fourier transform on the obtained transient acoustic vibration signals of transformers; and performing time-frequency analysis on the transient acoustic vibration signals of transformers after Fourier transform using a wavelet transform algorithm to obtain time domain information of the signal and corresponding frequency domain information. The present invention can describe the frequency components existing in the transient acoustic vibration signals of transformers, and can also display the time when each frequency component appears, and can also avoid the problem that the relative amplitude in wavelet time-frequency analysis is not intuitive.

[0056] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0057] In the embodiments provided in the present application, it should be understood that the division of units is merely a logical function division, and there may be other division methods in actual implementation, for example, multiple units may be combined into one unit, one unit may be split into multiple units, or some features may be ignored.

[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for processing transient acoustic vibration signals of a transformer, characterized in that: The following steps are involved: Step S1, collecting transient acoustic vibration signals of the transformer; Step S2, performing fast Fourier transform on the obtained transformer transient acoustic vibration signal; Step S3, using a wavelet transform algorithm to perform time-frequency analysis on the transient acoustic vibration signal of the transformer after Fourier transform, to obtain time domain information of the signal and corresponding frequency domain information; In step S3, the transformer transient acoustic vibration signal is set to ; Its continuous wavelet transform is: ； a is the stretch factor, b is the translation factor; function The wavelet mother function is The generated parameters depend on The continuous wavelet function of , referred to as wavelet; ； It also includes the correction of wavelet coefficients. Combining the scale transformation properties of continuous wavelet transform and the corresponding wavelet basis function parameters, the following correction function can be obtained: ； In the formula is the corrected wavelet coefficient, is the wavelet coefficient before correction, is the signal frequency, is the signal sampling frequency, is the center frequency of the wavelet basis function.

Images