Transformer winding high-frequency resonance monitoring method, device, equipment and medium

Abstract

The invention discloses a transformer winding high-frequency resonance monitoring method, device and equipment and a medium. The method comprises the steps that signals collected in real time are real-time voltage and current time domain signals output by the secondary side of the mutual inductor; performing high-frequency correction on the real-time voltage and current time domain signals through a pre-calibrated amplitude correction coefficient and a phase delay correction parameter to obtain corrected voltage and current signals; carrying out improved short-time Fourier transform on the corrected voltage and current signals, and respectively extracting a voltage high-frequency component frequency spectrum and a current high-frequency component frequency spectrum; determining a local resonant frequency library of a transformer winding based on a transformer distributed parameter equivalent model established by a finite element method; matching the voltage high-frequency component frequency spectrum and the current high-frequency component frequency spectrum with resonant frequencies in a local resonant frequency library to determine a frequency matching degree; and determining a risk degree evaluation result of the transformer winding state according to the frequency matching degree.

Description

Technical Field

[0001] This invention relates to the field of transformer monitoring technology, and in particular to a method, device, equipment and medium for monitoring high-frequency resonance of transformer windings. Background Technology

[0002] During the operation of a power system, the transformer, as a core transmission device, directly affects the reliable operation of the entire power grid. Over long-term operation, transformers may be affected by external high-frequency disturbances (such as switching operations, lightning strikes, and harmonic interference) as well as internal insulation aging and changes in winding structure, leading to local resonance phenomena. Once the local resonance frequency approaches the frequency of the external intrusion high-frequency components, it often causes a sudden increase in voltage and current amplitude in the local windings, thereby accelerating insulation breakdown or triggering partial discharge, ultimately potentially leading to equipment failure or even shutdown.

[0003] Currently, the power industry primarily uses voltage transformers (PTs) and current transformers (CTs) with power frequency optimized designs for high-frequency transformer monitoring. These transformers exhibit good transfer characteristics around 50Hz, but their amplitude attenuates significantly in the kHz to MHz frequency range, and they also experience phase delays. This leads to lower measured values ​​of high-frequency components and frequency peak drift, directly affecting the accuracy of critical frequency matching. Furthermore, existing signal analysis typically employs short-time Fourier transforms (STFTs) with fixed time windows, which presents a challenge in simultaneously achieving both frequency and time resolution in high-frequency transient detection, making it difficult to accurately capture short-time high-frequency resonance events.

[0004] To identify the local resonant frequencies of transformers, some studies have attempted to use the finite element method to extract distributed parameters and build models. However, these models are mostly calculated offline and do not consider the resonant frequency drift caused by structural changes during equipment operation, leading to the risk of missed or false alarms in frequency matching during long-term monitoring. In addition, existing resonance risk assessments are mostly based on high-amplitude signals, failing to combine amplitude and resonant frequency proximity in the evaluation, making it difficult to comprehensively reflect the degree of danger.

[0005] Therefore, there is an urgent need for a monitoring method that can accurately correct the high-frequency characteristics of transformers, take into account both high-frequency transient detection and long-term resonant frequency adaptive updates, in order to improve the accuracy of dangerous frequency matching of transformers and maintain the effectiveness of risk assessment throughout the entire life cycle. Summary of the Invention

[0006] Based on this, it is necessary to propose a method, device, equipment, and medium for high-frequency resonance monitoring of transformer windings to address the above problems.

[0007] A method for monitoring high-frequency resonance in transformer windings, the method comprising: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer; The real-time voltage and current time-domain signals are corrected at high frequency by using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals. An improved short-time Fourier transform is performed on the corrected voltage and current signals, and the high-frequency component spectra of the voltage and the high-frequency component spectra of the current are extracted respectively. The local resonant frequency library of transformer windings is determined based on the equivalent model of distributed parameters of transformers established by the finite element method. The frequency matching degree is determined by matching the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library. The risk assessment result of the transformer winding condition is determined based on the frequency matching degree.

[0008] Preferably, the pre-calibrated amplitude correction coefficient and phase delay correction parameter specifically include: A sweep frequency signal of known amplitude is applied to the input terminals of the voltage transformer and the current transformer, respectively; Simultaneously measure the voltage and current transformers and the secondary side output of the current transformer with a known amplitude secondary side signal; Determine the amplitude ratio and phase difference between the input and output voltage and current waveforms based on the known amplitude sweep frequency signal and the known amplitude secondary side signal; The amplitude correction coefficient is determined based on the amplitude ratio between the input and output voltage and current waveforms. The phase delay correction parameter is determined based on the phase difference between the input and output voltage and current waveforms.

[0009] Preferably, the step of performing high-frequency correction on the voltage and current signals using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals specifically includes: The spectrum is obtained by performing Fourier decomposition on the collected voltage and current transformers and the secondary side signals of the current transformers. The spectrum is compensated by the voltage and current amplitude correction coefficients; The phase portion of the spectrum is delayed and compensated, and then the corrected time-domain signal is obtained by inverse Fourier transform.

[0010] Preferably, the step of performing an improved short-time Fourier transform on the corrected voltage and current signals and extracting the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current respectively includes: preprocessing the corrected voltage and current signals by adding a Hamming window function; The preprocessed voltage and current signals are subjected to a short-time Fourier transform in the low-frequency band to obtain the complex spectrum in the low-frequency band. A frequency band adaptive short-time Fourier transform method is used to perform a high-frequency short-time Fourier transform on the complex spectrum of the low-frequency band to obtain the complex spectrum of the high-frequency band. The complex spectrum of the low-frequency band and the complex spectrum of the high-frequency band are fused to obtain a fused spectrum; The fused spectrum is extracted to obtain the high-frequency component spectra of voltage and current.

[0011] Preferably, the determination of the local resonant frequency library of the transformer windings using the transformer distributed parameter equivalent model established based on the finite element method specifically includes: The windings of the transformer are discretized according to their insulation structure, resulting in several discs. The local inductance and capacitance values ​​of each pie are determined by the finite element method, and the parameter set of each pie is used to form a distributed parameter list. A multi-degree-of-freedom oscillation circuit is formed based on the distributed parameter list; Construct the system matrix based on the multi-degree-of-freedom oscillation circuit; The resonant frequency corresponding to each pie is determined based on the system matrix, and the resonant frequencies of all pie pieces form a resonant frequency library.

[0012] Preferably, the step of matching the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree specifically includes: according to Determine the frequency matching degree, where, f rk The resonant frequency, F r This is a library of resonant frequencies.

[0013] Preferably, determining the risk assessment result of the transformer winding condition based on the frequency matching degree specifically includes: based on Determine the risk assessment results; among which, α and β These represent the amplitude and frequency weights, respectively. A max This is the maximum safe amplitude.

[0014] A transformer winding condition monitoring device based on high-frequency resonance, the device comprising: The high-speed acquisition unit is used to acquire real-time voltage and current time-domain signals output from the secondary side of the transformer. The real-time voltage and current time-domain signals are then subjected to high-frequency correction using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals. The signal analysis unit is used to perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current, respectively. The resonance matching and hazard calculation unit is used to determine the local resonant frequency library of the transformer winding based on the equivalent model of the transformer distributed parameters established by the finite element method; to match the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree; and to determine the hazard evaluation result of the transformer winding state based on the frequency matching degree.

[0015] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer; The real-time voltage and current time-domain signals are corrected at high frequency by using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals. An improved short-time Fourier transform is performed on the corrected voltage and current signals, and the high-frequency component spectra of the voltage and the high-frequency component spectra of the current are extracted respectively. The local resonant frequency library of transformer windings is determined based on the equivalent model of distributed parameters of transformers established by the finite element method. The frequency matching degree is determined by matching the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library. The risk assessment result of the transformer winding condition is determined based on the frequency matching degree.

[0016] A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the following steps: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer; The real-time voltage and current time-domain signals are corrected at high frequency by using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals. An improved short-time Fourier transform is performed on the corrected voltage and current signals, and the high-frequency component spectra of the voltage and the high-frequency component spectra of the current are extracted respectively. The local resonant frequency library of transformer windings is determined based on the equivalent model of distributed parameters of transformers established by the finite element method. The frequency matching degree is determined by matching the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library. The risk assessment result of the transformer winding condition is determined based on the frequency matching degree.

[0017] The embodiments of the present invention have the following beneficial effects: This invention performs high-frequency correction on the real-time signal of the secondary side of the instrument transformer, eliminating the amplitude attenuation and phase delay of traditional instrument transformers in the 1kHz~1MHz frequency band, significantly improving the measurement accuracy of high-frequency components. Furthermore, it employs an improved short-time Fourier transform, using an adaptive window length to balance frequency and time resolution, sensitively capturing short-time resonant transient events. A resonant frequency library based on finite element modeling provides an accurate benchmark, and combined with risk assessment using frequency matching degree and amplitude weighting, it avoids misjudgment of high amplitude and omission of low amplitude. Simultaneously, the resonant frequency adaptive update mechanism dynamically adjusts the library data through online impedance spectrum, ensuring monitoring reliability throughout the entire lifecycle. This method ultimately achieves high-precision, high-sensitivity risk assessment of transformer winding conditions (such as insulation aging and partial discharge), providing closed-loop protection for power grid safety. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] in: Figure 1 This is a flowchart of a method for monitoring high-frequency resonance of transformer windings, provided as an embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention provides a method for monitoring high-frequency resonance in transformer windings, such as... Figure 1 As shown, the method includes: Step 101: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer. Specifically, the acquisition unit is a high-speed oscilloscope with a sampling rate ≥1MHz, acquiring the secondary output signals of voltage transformers and current transformers.

[0022] Step 102: Perform high-frequency correction on the real-time voltage and current time-domain signals using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals; Specifically, before the signal enters the frequency analysis module, the amplitude correction coefficient is obtained through the frequency response calibration of the voltage and current transformers. K u ( f ), K i ( f ) and phase correction delay parameter t u ( f ), t i ( f The amplitude and phase of the acquired voltage and current signals are compensated to eliminate measurement errors of the transformer in the high-frequency range and ensure the accuracy of high-frequency component analysis.

[0023] A sweep frequency signal of known amplitude is applied to the input terminals of the voltage transformer and the current transformer, respectively. U in ( f ), I in ( f The signal frequency range covers the target analysis frequency band (1kHz~1MHz). The voltage and current transformers and the secondary side output of the current transformer with known amplitudes are measured simultaneously. U out ( f ), I out ( f ); Based on the known amplitude sweep signal U in ( f ), I in ( f ) and secondary side signals with known amplitudes U out ( f ), I out ( f Determine the amplitude ratio between the input and output voltage and current waveforms. H ( f and phase difference F ( f ); The amplitude correction coefficient is determined based on the amplitude ratio between the input and output voltage and current waveforms. Among them, according to , Determine the voltage-to-current amplitude ratio H magU ( f ), H magI ( f ); according to , Determine the amplitude correction factors for voltage and current. K U ( f ), K I ( f ).

[0024] In the formula, K U ( f )and K I ( f ) indicates frequency f The voltage and current amplitude correction factors are dimensionless. H magU ( f )and H magI ( f ) indicates frequency f The ratio of the magnitude of the voltage and current.

[0025] The phase delay correction parameter is determined based on the phase difference between the input and output voltage and current waveforms.

[0026] Among them, according to , Determine the phase difference between voltage and current F U ( f ), F I ( f ); according to , Determine the phase delay correction parameters for voltage and current. t U ( f ), t I ( f ); In the formula, f U ( f )and f I ( f ) indicates frequency f The phase deviation between voltage and current under the given conditions, expressed in radians; t U ( f )and t I ( f () represents the time delay corresponding to the phase deviation of voltage and current at frequency f, in seconds.

[0027] The acquired voltage and current transformers and the secondary side signals of the current transformers u m ( t ), i m ( t Fourier decomposition was performed to obtain the spectrum. u m ( f ), i m ( f ); Among them, according to , For secondary side signals u m ( t ), i m ( t Perform spectral transformation to obtain the spectrum. u m ( f ), i m ( f ); The spectrum is adjusted using the voltage and current amplitude correction coefficients. u m ( f ), i m ( f Compensation will be provided. Among them, according to , For the spectrum u m ( f ), i m ( f To receive compensation U ( f ), I ( f ); The phase portion of the spectrum is delayed and compensated for, and then the corrected time-domain signal is obtained through inverse Fourier transform. u ( t ), i ( t ).

[0028] Among them, according to , Perform delay compensation to obtain the delayed compensation result. U ( f ), I ( f ); Then according to , The time-domain corrected signal is obtained by performing an inverse Fourier transform. u ( t ), i ( t ).

[0029] Step 103: Perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage respectively. A u ( t , f ) and the spectrum of high-frequency components of current A I ( t , f ); Specifically, the corrected voltage and current signals are preprocessed by adding a Hamming window function; In order to reduce spectral leakage and prevent the main frequency energy of voltage and current signals from masking high-frequency details, according to , Time-domain corrected signal u ( t ), i ( t Add a window function; In the formula, u ( v )and i ( v This indicates the port voltage and current at the sampling point after amplitude and phase correction by the current transformer. v The value of v; v is the sampling point index of the time series, which is related to the start time t of the current analysis, v=0,1,...,N-1; oh ( v () represents the Hamming window function, and its formula is: ; N This represents the number of points used in a conventional short-time Fourier transform.

[0030] Perform a short-time Fourier transform on the preprocessed voltage and current signals in the low-frequency band to obtain the complex spectrum in the low-frequency band. U ( t , f ),I ( t , f ); Among them, according to , Perform a short-time Fourier transform on the low-frequency band to obtain the complex spectrum of the low-frequency band. U ( t , f ), I ( t , f ); In the formula, t Indicates the start time of the current analysis. f Indicates frequency, F s Indicates the frequency used. U ( t , f )and I ( t , f () represent time respectively t Corresponding frequency f The complex spectrum of voltage and current.

[0031] The complex spectrum of the low-frequency band is obtained by using a frequency band adaptive short-time Fourier transform method. U ( t , f ), I ( t , f Perform a short-time Fourier transform on a high frequency band to obtain the complex spectrum in the high-frequency range. U hf ( t , f ), I hf ( t , f ); Among them, since the high-frequency band and the low-frequency band switch the window length within the same short-time Fourier transform framework, specific improvements have been made for the high-frequency resonance monitoring scenario of transformers. This ensures both the frequency resolution of the low frequency and the transient response capability of the high frequency, thereby improving the sensitivity of high-frequency resonance monitoring of transformers.

[0032] Therefore, this invention employs a frequency band adaptive short-time Fourier transform method, the calculation method of which is as follows: When the frequency is above a certain threshold f cut When using compressed window length: , N hf Indicates the length of the high-frequency analysis window. k This represents the compression factor, which is typically between 2 and 4.

[0033] The corresponding time window length is: ;△ T This indicates the length of the routine analysis time window.

[0034] Therefore, according to , Perform a short-time Fourier transform at high frequencies to obtain the complex spectrum in the high-frequency band. U hf ( t , f ), I hf ( t , f ).

[0035] For the complex spectrum of the low frequency band U ( t , f ), I ( t , f ), high-frequency complex spectrum U hf ( t , f ), I hf ( t , f The fusion process is performed to obtain a fused spectrum; Among them, according to , Perform fusion to obtain a fused spectrum U ’ ( t , f ), I ’ ( t , f ).

[0036] The fused spectrum is extracted to obtain the high-frequency component spectra of voltage and current. A u ( t , f ), A I ( t , f ).

[0037] Among them, according to , Determine the high-frequency component spectra of voltage and current A u ( t , f ), A I( t , f ).

[0038] Step 104: Determine the local resonant frequency library of the transformer windings based on the equivalent model of distributed parameters of the transformer established by the finite element method. F r ; Specifically, the windings of the transformer are discretized according to the insulation structure, one disc at a time, to obtain several discs; In this process, the transformer windings are discretized and modeled layer by layer according to the insulation structure, with each layer modeled as: distributed inductance. L i (Determined by the number of winding turns, geometry, and magnetic circuit), distributed capacitance C i,i+1 (between adjacent cakes) and C i,gnd (Between the disc and the ground).

[0039] The local inductance and capacitance values ​​of each pie are determined by the finite element method, and the parameter set of each pie is used to form a distributed parameter list. The finite element method is used to calculate the electric and magnetic field distributions of each cake, deriving the local inductance and capacitance values. The calculation formulas are as follows: , ; N i Indicates the first i Number of turns of the cake f i Indicates magnetic flux. I i This indicates the current passing through the cake. e 0 and e r This represents the vacuum permittivity and the relative permittivity of insulating materials. S ij This represents the area of ​​the effective electric field between adjacent cakes. d ij This indicates the distance between adjacent cakes.

[0040] Each cake is considered as ( L i , C i,i+1 , C i,gnd The network nodes are connected to form a multi-degree-of-freedom oscillating circuit.

[0041] A multi-degree-of-freedom oscillation circuit is formed based on the distributed parameter list; Construct the system matrix based on the multi-degree-of-freedom oscillation circuit; The system matrix form of the transformer distributed equivalent circuit is as follows: ,L For the inductor matrix, C For the capacitance matrix, q ( t ) represents the charge vector of each cake.

[0042] Solve for the eigenvalues ​​of the system matrix, i.e.: The natural angular frequencies ω of each mode are obtained. k ; The resonant frequency corresponding to each pie is determined based on the system matrix, and the resonant frequencies of all pie pie pie pie constitute a resonant frequency library. F r .

[0043] Among them, according to Determine the corresponding resonant frequency f rk ; All f rk Forming a resonant frequency library F r .

[0044] Step 105, the high-frequency component spectrum of the voltage A u ( t , f ) and the spectrum of high-frequency components of current A I ( t , f ) and local resonant frequency library F r The resonant frequencies in the circuit are matched to determine the frequency matching degree. Specifically, according to Determine the frequency matching degree.

[0045] Step 106: Determine the risk assessment result of the transformer winding condition based on the frequency matching degree.

[0046] Specifically, according to Determine the risk assessment results; among which, α and β These represent the amplitude and frequency weights, respectively. A max For the maximum safe amplitude, A c (f) For the current frequency f Voltage amplitude spectrum A u (t,f) or current amplitude spectrum A I (t,f) .

[0047] Considering that in actual operation, when transformer winding resonance occurs, the voltage and current amplitudes can amplify to more than 10 times their normal values, and that such abnormal amplitude changes only occur when the frequency is close to the local resonant frequency, a higher weight needs to be assigned to frequency matching in the hazard assessment. Based on on-site operation monitoring and engineering experience, [the following is taken as an example]. α= 0.2 ，β= 0.8, the hazard assessment criteria are given, as shown in Table 1.

[0048] Table 1 Transformer Hazard Assessment Standards

[0049] This invention performs high-frequency correction on the real-time signal of the secondary side of the instrument transformer, eliminating the amplitude attenuation and phase delay of traditional instrument transformers in the 1kHz~1MHz frequency band, significantly improving the measurement accuracy of high-frequency components. Furthermore, it employs an improved short-time Fourier transform, using an adaptive window length to balance frequency and time resolution, sensitively capturing short-time resonant transient events. A resonant frequency library based on finite element modeling provides an accurate benchmark, and combined with risk assessment using frequency matching degree and amplitude weighting, it avoids misjudgment of high amplitude and omission of low amplitude. Simultaneously, the resonant frequency adaptive update mechanism dynamically adjusts the library data through online impedance spectrum, ensuring monitoring reliability throughout the entire lifecycle. This method ultimately achieves high-precision, high-sensitivity risk assessment of transformer winding conditions (such as insulation aging and partial discharge), providing closed-loop protection for power grid safety.

[0050] This invention also provides a transformer winding condition monitoring device based on high-frequency resonance, the device comprising: The high-speed acquisition unit is used to acquire real-time voltage and current time-domain signals output from the secondary side of the transformer. The real-time voltage and current time-domain signals are then subjected to high-frequency correction using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals. The signal analysis unit is used to perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current, respectively. The resonance matching and hazard calculation unit is used to determine the local resonant frequency library of the transformer winding based on the equivalent model of the transformer distributed parameters established by the finite element method; to match the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree; and to determine the hazard evaluation result of the transformer winding state based on the frequency matching degree.

[0051] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the following steps: Step 101: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer. Step 102: Perform high-frequency correction on the real-time voltage and current time-domain signals using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals; Step 103: Perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current, respectively. Step 104: Determine the local resonant frequency library of the transformer winding based on the equivalent model of the transformer distributed parameters established by the finite element method; Step 105: Match the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree. Step 106: Determine the risk assessment result of the transformer winding condition based on the frequency matching degree.

[0052] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, causes the processor to perform the following steps: Step 101: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer. Step 102: Perform high-frequency correction on the real-time voltage and current time-domain signals using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals; Step 103: Perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current, respectively. Step 104: Determine the local resonant frequency library of the transformer winding based on the equivalent model of the transformer distributed parameters established by the finite element method; Step 105: Match the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree. Step 106: Determine the risk assessment result of the transformer winding condition based on the frequency matching degree.

[0053] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0054] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for monitoring high-frequency resonance in transformer windings, characterized in that, The method includes: The signals acquired in real time are the real-time voltage and current time-domain signals output from the secondary side of the transformer; The real-time voltage and current time-domain signals are corrected at high frequency by using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals. An improved short-time Fourier transform is performed on the corrected voltage and current signals, and the high-frequency component spectra of the voltage and the high-frequency component spectra of the current are extracted respectively. The local resonant frequency library of transformer windings is determined based on the equivalent model of distributed parameters of transformers established by the finite element method. The frequency matching degree is determined by matching the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library. The risk assessment result of the transformer winding condition is determined based on the frequency matching degree.

2. The method for monitoring high-frequency resonance of transformer windings according to claim 1, characterized in that, The pre-calibrated amplitude correction coefficient and phase delay correction parameter specifically include: A sweep frequency signal of known amplitude is applied to the input terminals of the voltage transformer and the current transformer, respectively; Simultaneously measure the voltage and current transformers and the secondary side output of the current transformer with a known amplitude secondary side signal; Determine the amplitude ratio and phase difference between the input and output voltage and current waveforms based on the known amplitude sweep frequency signal and the known amplitude secondary side signal; The amplitude correction coefficient is determined based on the amplitude ratio between the input and output voltage and current waveforms. The phase delay correction parameter is determined based on the phase difference between the input and output voltage and current waveforms.

3. The method for monitoring high-frequency resonance of transformer windings according to claim 1 or 2, characterized in that, The step of performing high-frequency correction on the voltage and current signals using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain corrected voltage and current signals specifically includes: The spectrum is obtained by performing Fourier decomposition on the collected voltage and current transformers and the secondary side signals of the current transformers. The spectrum is compensated by the voltage and current amplitude correction coefficients; The phase portion of the spectrum is delayed and compensated, and then the corrected time-domain signal is obtained by inverse Fourier transform.

4. The method for monitoring high-frequency resonance of transformer windings according to claim 3, characterized in that, The step of performing an improved short-time Fourier transform on the corrected voltage and current signals and extracting the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current respectively includes: adding a Hamming window function to the corrected voltage and current signals for preprocessing; The preprocessed voltage and current signals are subjected to a short-time Fourier transform in the low-frequency band to obtain the complex spectrum in the low-frequency band. A frequency band adaptive short-time Fourier transform method is used to perform a high-frequency short-time Fourier transform on the complex spectrum of the low-frequency band to obtain the complex spectrum of the high-frequency band. The complex spectrum of the low-frequency band and the complex spectrum of the high-frequency band are fused to obtain a fused spectrum; The fused spectrum is extracted to obtain the high-frequency component spectra of voltage and current.

5. The method for monitoring high-frequency resonance of transformer windings according to claim 4, characterized in that, The transformer distributed parameter equivalent model established based on the finite element method determines the local resonant frequency library of the transformer winding, specifically including: The windings of the transformer are discretized according to their insulation structure, resulting in several discs. The local inductance and capacitance values ​​of each pie are determined by the finite element method, and the parameter set of each pie is used to form a distributed parameter list. A multi-degree-of-freedom oscillation circuit is formed based on the distributed parameter list; Construct the system matrix based on the multi-degree-of-freedom oscillation circuit; The resonant frequency corresponding to each pie is determined based on the system matrix, and the resonant frequencies of all pie pieces form a resonant frequency library.

6. The method for monitoring high-frequency resonance of transformer windings according to claim 5, characterized in that, The step of matching the high-frequency component spectra of the voltage and current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree specifically includes: according to Determine the frequency matching degree, where, f rk The resonant frequency, F r This is a library of resonant frequencies.

7. The method for monitoring high-frequency resonance of transformer windings according to claim 6, characterized in that, The determination of the risk assessment result of the transformer winding condition based on the frequency matching degree specifically includes: based on Determine the risk assessment results; among which, α and β These represent the amplitude and frequency weights, respectively. A max This is the maximum safe amplitude.

8. A transformer winding condition monitoring device based on high-frequency resonance, characterized in that, The device includes: The high-speed acquisition unit is used to acquire real-time voltage and current time-domain signals output from the secondary side of the transformer. The real-time voltage and current time-domain signals are then subjected to high-frequency correction using pre-calibrated amplitude correction coefficients and phase delay correction parameters to obtain the corrected voltage and current signals. The signal analysis unit is used to perform an improved short-time Fourier transform on the corrected voltage and current signals and extract the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current, respectively. The resonance matching and hazard calculation unit is used to determine the local resonant frequency library of the transformer winding based on the equivalent model of the transformer distributed parameters established by the finite element method; to match the high-frequency component spectrum of the voltage and the high-frequency component spectrum of the current with the resonant frequencies in the local resonant frequency library to determine the frequency matching degree; and to determine the hazard evaluation result of the transformer winding state based on the frequency matching degree.

9. A computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the steps of the method as claimed in any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method as claimed in any one of claims 1 to 7.

Images