System for diagnosing deformation of transformer winding based on invasive wave and vibration signals

Abstract

The application discloses a system for diagnosing and fusing transformer winding deformation based on invasive wave and vibration signals, an invasive wave signal collection unit collects invasive wave signals and transmits the signals to an invasive wave characteristic value extraction unit, a wave characteristic value extraction module extracts the invasive wave signals to perform invasive wave characteristic value extraction processing; a vibration signal collection unit collects transformer vibration signals and transmits the signals to a vibration characteristic value extraction unit, the vibration characteristic value extraction unit extracts the transformer vibration signals to perform vibration characteristic value extraction processing; a truncated normal fitting unit performs characteristic value diagnosis function calculation processing on the invasive wave characteristic value extraction processing result and the vibration characteristic value extraction processing result; a characteristic value weight calculation unit performs characteristic value weight calculation processing based on the characteristic value diagnosis function calculation processing result; and data fusion processing is performed by the data fusion unit based on the characteristic value weight calculation processing result, so as to determine whether the transformer winding is deformed and the deformation degree.

Description

Technical Field

[0001] This invention relates to the field of power equipment testing technology, and in particular to a system for determining transformer winding deformation based on the fusion of intrusion wave and vibration signal diagnosis. Background Technology

[0002] As a crucial piece of equipment in the power transmission and distribution system, transformers are costly to construct, and any failure can have adverse effects on the power grid. Therefore, the safe and stable operation of transformers is of profound significance for ensuring power grid security. Real-time sensing and diagnosis of transformer operating status can make power system operation safer, more reliable, more efficient, and more intelligent, laying a solid foundation for the digital transformation of the power grid. Transformer winding deformation diagnosis is an important component of transformer fault diagnosis. Currently, there are two main types of methods for online assessment of winding deformation: the first is to invert the operational characteristics exhibited by the winding deformation; the second is to measure factors that may cause winding deformation and then estimate the probability of transformer winding deformation.

[0003] Most existing studies separate the two types of assessment methods and evaluate them from only one aspect, which leads to high rates of misdiagnosis and missed diagnosis, and makes it difficult to achieve multi-feature fusion diagnosis. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, this invention provides a transformer winding deformation determination system based on the diagnostic fusion of intrusion wave and vibration signal. This system enables the determination of whether the transformer winding is deformed and the degree of deformation by performing diagnostic fusion using intrusion wave signal and transformer vibration signal.

[0005] To achieve the above-mentioned objectives, the present invention adopts the following technical solution: A transformer winding deformation diagnosis and determination system based on the fusion of intrusion wave and vibration signal diagnostics includes an intrusion wave signal acquisition unit, an intrusion wave feature value extraction unit, a truncated normal fitting unit, a feature value weight calculation unit, a vibration signal acquisition unit, a vibration feature value extraction unit, and a data fusion unit. The intrusion wave signal acquisition unit is connected to the intrusion wave feature value extraction unit, the vibration signal acquisition unit is connected to the vibration feature value extraction unit, the truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit, and the feature value weight calculation unit, and the feature value weight calculation unit is connected to the data fusion unit. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave signal and transmits it to the truncated normal fitting unit. The system consists of a truncated normal fitting unit and a vibration signal acquisition unit. The vibration signal acquisition unit collects the transformer vibration signal and transmits it to the vibration feature value extraction unit. The vibration feature value extraction unit extracts the vibration feature values ​​from the transformer vibration signal and transmits them to the truncated normal fitting unit. The truncated normal fitting unit calculates the diagnostic functions for each feature value based on the extracted wave feature value and vibration feature value, and transmits the results to the feature value weight calculation unit. The feature value weight calculation unit calculates the feature value weights based on the results of the feature value diagnostic function calculation and transmits the results to the data fusion unit. The data fusion unit performs data fusion based on the results of the feature value weight calculation to determine whether the transformer windings are deformed and the degree of deformation.

[0006] Furthermore, the intrusion wave signal acquisition unit includes a voltage transformer module and a current transformer module. The voltage transformer module acquires the intrusion wave voltage signal and transmits it to the intrusion wave feature value extraction unit, while the current transformer module acquires the intrusion wave current signal and transmits it to the intrusion wave feature value extraction unit.

[0007] Furthermore, the vibration signal acquisition unit includes a vibration signal acquisition module, which acquires vibration signals and transmits them to the vibration feature value extraction unit.

[0008] Furthermore, the vibration signal acquisition module includes one or more acceleration sensors, which are installed on both sides of the transformer tank wall.

[0009] Optionally, the installation method of the acceleration sensor of the vibration signal acquisition module is to install the vibration signal acquisition module on both sides of the transformer tank wall. Two vibration signal acquisition points are set for each phase of the transformer winding. The specific installation positions are one-third and two-thirds of the same height on the side of the transformer tank.

[0010] Optionally, the acceleration sensor of the vibration signal acquisition module is a capacitive acceleration sensor.

[0011] Furthermore, the intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave signal, specifically including: performing intrusion wave feature value extraction processing on the intrusion wave signal, and calculating the intrusion wave voltage amplitude, intrusion wave current amplitude, intrusion wave voltage spectrum main frequency, intrusion wave current spectrum main frequency, intrusion wave voltage spectrum entropy value, and intrusion wave current spectrum entropy value.

[0012] Furthermore, the vibration feature extraction unit performs vibration feature extraction processing on the transformer vibration signal, specifically including: extracting vibration feature values ​​from the transformer vibration signal, and calculating the low-frequency proportion of the vibration spectrum, the total harmonic distortion rate of the vibration, and the vibration entropy.

[0013] Furthermore, the truncated normal fitting unit performs diagnostic function calculations for each feature value in the extracted and processed results of the intrusive wave and vibration features. Specifically, this includes: The truncated normal fitting unit sets the domain of each eigenvalue, performs standard normal fitting, and obtains bell curves for each eigenvalue pair; the bell curves for each eigenvalue pair are truncated according to the domain of the eigenvalues, and the truncated curves are normalized; the truncated normal curves of normal data are compared with the truncated normal curves of fault data to obtain the diagnostic functions for each eigenvalue.

[0014] Furthermore, the eigenvalue weight calculation unit performs eigenvalue weight calculation based on the eigenvalue diagnostic function calculation results. Specifically, the eigenvalue weight calculation unit performs eigenvalue variation coefficient calculation on the intrusion wave eigenvalue extraction results and vibration eigenvalue extraction results, and uses the relative proportion of each eigenvalue variation coefficient calculation result as the eigenvalue weight of the corresponding eigenvalue diagnostic function calculation results.

[0015] Furthermore, the data fusion unit performs data fusion processing based on the weight calculation results of each feature value, specifically including: Based on the calculation and processing results of the diagnostic function of each characteristic value of the intrusion wave and the calculation and processing results of the weight of each characteristic value of the intrusion wave, the intrusion wave fusion processing is performed to obtain the intrusion wave evaluation result of the transformer winding deformation. Vibration fusion processing is performed based on the calculation results of the diagnostic function of each vibration characteristic value and the calculation results of the weight of each vibration characteristic value to obtain the vibration assessment results of the transformer winding deformation. The evaluation results are fused based on the intrusion wave assessment results and the vibration assessment results to obtain the fused diagnostic results of the intrusion wave and vibration on the transformer winding deformation.

[0016] Furthermore, the transformer winding deformation determination system based on the diagnostic fusion of intrusion wave and vibration signal also includes a Wiener filtering unit and a vibration denoising unit. The Wiener filtering unit is connected to the intrusion wave signal acquisition unit and the intrusion wave feature value extraction unit, respectively, and the vibration denoising unit is connected to the vibration signal acquisition unit and the vibration feature value extraction unit, respectively. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the Wiener filtering unit, which filters the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration denoising unit, which denoises the transformer vibration signal and transmits it to the vibration feature value extraction unit.

[0017] Furthermore, the Wiener filter unit performs filtering processing on the intrusion wave voltage signal and the intrusion wave current signal, specifically including: performing high-frequency noise filtering processing on the intrusion wave voltage signal and the intrusion wave current signal through the filter of the Wiener filter unit.

[0018] Furthermore, in the transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis, the vibration signal acquisition unit also includes a vibration noise acquisition module, which acquires noise signals and transmits them to the vibration denoising unit.

[0019] Furthermore, the vibration denoising unit performs denoising processing on the vibration signal and noise signal, specifically by performing differential denoising processing on the vibration signal and noise signal.

[0020] Furthermore, the transformer winding deformation diagnosis and determination system based on the fusion of intrusion wave and vibration signal diagnosis also includes an anomaly detection unit, which is used to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal.

[0021] Furthermore, the anomaly detection unit is used to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal, including the following steps: The vibration signal waveform is standardized based on the transformer voltage during vibration measurement; The standardized vibration signal is compared with historical vibration data under normal operation in the database; If the standardized vibration amplitude is greater than the historical vibration amplitude threshold, it is determined that the transformer vibration signal is abnormal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into working state. If the standardized vibration amplitude is less than or equal to the historical vibration amplitude threshold, it is determined that there is no abnormality in the transformer vibration signal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into a dormant state.

[0022] Furthermore, the vibration signal waveform is standardized based on the transformer voltage during vibration measurement. The standardization calculation formula is shown below: A2 = A1 / U 2 Where A1 is the transformer vibration signal before standardization, A2 is the transformer vibration signal after standardization, and U is the voltage during vibration measurement.

[0023] The beneficial effects of this invention are: it enables the diagnosis and fusion of intrusion wave signals and transformer vibration signals to determine whether the transformer windings are deformed and the degree of deformation.

[0024] By leveraging the domain of eigenvalues, the error of the diagnostic function can be reduced, thereby improving the accuracy of diagnostic fusion in determining transformer winding deformation.

[0025] The system achieves noise reduction of intrusion waves and vibration signals through Wiener filtering and vibration denoising units, which greatly improves the accuracy of diagnostic fusion judgment of transformer winding deformation.

[0026] The system enables the anomaly detection unit to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal. Based on whether there is an anomaly in the transformer vibration signal, the system enables the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit to be either in a dormant state or in an active state, which greatly reduces the operating load of the monitoring equipment. Attached Figure Description

[0027] 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.

[0028] Figure 1 This is a schematic diagram of the structure of the transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis of the present invention. Detailed Implementation

[0029] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0030] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0031] Example 1: A transformer winding deformation diagnosis and determination system based on the fusion of intrusion wave and vibration signal diagnostics includes an intrusion wave signal acquisition unit, an intrusion wave feature value extraction unit, a truncated normal fitting unit, a feature value weight calculation unit, a vibration signal acquisition unit, a vibration feature value extraction unit, and a data fusion unit. The intrusion wave signal acquisition unit is connected to the intrusion wave feature value extraction unit, the vibration signal acquisition unit is connected to the vibration feature value extraction unit, the truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit, and the feature value weight calculation unit, and the feature value weight calculation unit is connected to the data fusion unit. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave signal and transmits it to the truncated normal fitting unit. The system consists of a truncated normal fitting unit and a vibration signal acquisition unit. The vibration signal acquisition unit collects the transformer vibration signal and transmits it to the vibration feature value extraction unit. The vibration feature value extraction unit extracts the vibration feature values ​​from the transformer vibration signal and transmits them to the truncated normal fitting unit. The truncated normal fitting unit calculates the diagnostic functions for each feature value based on the extracted wave feature value and vibration feature value, and transmits the results to the feature value weight calculation unit. The feature value weight calculation unit calculates the feature value weights based on the results of the feature value diagnostic function calculation and transmits the results to the data fusion unit. The data fusion unit performs data fusion based on the results of the feature value weight calculation to determine whether the transformer windings are deformed and the degree of deformation.

[0032] The intrusion wave signal acquisition unit is connected to the intrusion wave feature value extraction unit. The intrusion wave signal acquisition unit includes a voltage transformer module and a current transformer module. The voltage transformer module acquires the intrusion wave voltage signal and transmits it to the intrusion wave feature value extraction unit. The current transformer module acquires the intrusion wave current signal and transmits it to the intrusion wave feature value extraction unit. The intrusion wave feature value extraction unit receives the intrusion wave voltage signal transmitted by the voltage transformer module and the intrusion wave current signal transmitted by the current transformer module. The intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave voltage signal and the intrusion wave current signal. The intrusion wave feature value extraction processing unit transmits the intrusion wave feature value extraction processing to the truncated normal fitting unit.

[0033] The vibration signal acquisition unit is connected to the vibration feature value extraction unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration feature value extraction unit. The vibration signal acquisition unit includes a vibration signal acquisition module, which acquires the vibration signal and transmits it to the vibration feature value extraction unit; that is, the transformer vibration signal includes the vibration signal itself. The vibration feature value extraction unit receives the vibration signal transmitted by the vibration signal acquisition module and performs vibration feature value extraction processing on the vibration signal. The vibration feature value extraction unit then transmits the vibration feature value extraction processing result to the truncated normal fit unit.

[0034] The vibration signal acquisition module includes one or more accelerometers. The module acquires vibration signals and transmits them to the vibration feature value extraction unit. The specific number of accelerometers in the vibration signal acquisition module can be selected according to actual usage. The accelerometers of the vibration signal acquisition module are installed on both sides of the transformer tank wall to acquire transformer vibration signals. The accelerometers in the vibration signal acquisition module can be capacitive accelerometers; that is, the type of accelerometer is not specifically defined in this application. The goal is to ensure that all acquired transformer vibration signals fall within the protection scope of this application. The vibration signal acquisition module is installed on both sides of the transformer tank wall to facilitate the acquisition of vibration acceleration waveforms. For example, with the vibration signal acquisition module installed on both sides of the transformer tank wall, six vibration signal acquisition points can be set on each side of the transformer tank wall, i.e., two vibration signal acquisition points are set on each phase of the longitudinal transformer winding. The specific installation positions in the longitudinal direction can be at one-third or two-thirds of the height of the transformer tank side.

[0035] The intrusion wave feature value extraction unit is connected to both the intrusion wave signal acquisition unit and the truncated normal fitting unit. The intrusion wave signal acquisition unit transmits the intrusion wave signal to the intrusion wave feature value extraction unit. The intrusion wave signal includes an intrusion wave voltage signal and an intrusion wave current signal. The intrusion wave feature value extraction unit receives the intrusion wave signal transmitted by the intrusion wave signal acquisition unit, specifically the intrusion wave voltage signal transmitted by the voltage transformer module and the intrusion wave current signal transmitted by the current transformer module. The intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave voltage signal and the intrusion wave current signal. Specifically, the intrusion wave feature value extraction processing includes: extracting intrusion wave feature values ​​from the intrusion wave signal and calculating the intrusion wave voltage amplitude, intrusion wave current amplitude, intrusion wave voltage spectrum dominant frequency, intrusion wave current spectrum dominant frequency, intrusion wave voltage spectrum entropy value, and intrusion wave current spectrum entropy value. The intrusion wave feature value extraction unit transmits the intrusion wave feature value extraction processing results to the truncated normal fitting unit.

[0036] The vibration signal acquisition unit is connected to the vibration feature value extraction unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration feature value extraction unit. The transformer vibration signal includes vibration signals. The vibration signal acquisition module acquires the vibration signals and transmits them to the vibration feature value extraction unit. The vibration feature value extraction unit receives the vibration signals transmitted by the vibration signal acquisition module and performs vibration feature value extraction processing on the vibration signals. Specifically, the vibration feature value extraction unit performs vibration feature value extraction processing on the transformer vibration signals and calculates the low-frequency proportion of the vibration spectrum, the total harmonic distortion rate of the vibration, and the vibration entropy.

[0037] The truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit, and the feature value weight calculation unit, respectively. The intrusion wave feature value extraction unit transmits the intrusion wave feature value extraction processing results to the truncated normal fitting unit, and the vibration feature value extraction unit transmits the vibration feature value extraction processing results to the truncated normal fitting unit. The truncated normal fitting unit receives the intrusion wave feature value extraction processing results transmitted by the intrusion wave feature value extraction unit, and the truncated normal fitting unit receives the vibration feature value extraction processing results transmitted by the vibration feature value extraction unit. In other words, the truncated normal fitting unit receives both the intrusion wave feature value extraction processing results and the vibration feature value extraction processing results. The truncated normal fitting unit performs diagnostic function calculations for each feature value in the intrusion wave feature value extraction processing results and the vibration feature value extraction processing results. Specifically, the diagnostic function calculations for each feature value in the truncated normal fitting unit include: The truncated normal fitting unit sets the domain of each eigenvalue, performs standard normal fitting, and obtains bell-shaped curves for each eigenvalue pair. The bell-shaped curves are then truncated according to the eigenvalue domain range, and the truncated curves are normalized. The truncated normal curves of normal data are compared with those of fault data to obtain the corresponding diagnostic functions for each eigenvalue. The truncated normal fitting unit transmits the diagnostic function calculation results to the eigenvalue weight calculation unit. The eigenvalue weight calculation unit receives the diagnostic function calculation results from the truncated normal fitting unit and performs eigenvalue weight calculation based on these results. Specifically, the eigenvalue weight calculation unit calculates the coefficient of variation for each eigenvalue extracted from the intrusive wave and vibration eigenvalues, and uses the relative proportion of each coefficient of variation calculation result as the eigenvalue weight of the corresponding diagnostic function calculation result.

[0038] The eigenvalue weight calculation unit is connected to the data fusion unit. The eigenvalue weight calculation unit transmits the processing results of each eigenvalue weight calculation to the data fusion unit. The data fusion unit receives the processing results of each eigenvalue weight calculation transmitted by the eigenvalue weight calculation unit and performs data fusion processing based on these results to determine whether the transformer windings are deformed and the degree of deformation. The data fusion processing performed by the data fusion unit based on the processing results of each eigenvalue weight calculation specifically includes: Based on the calculation and processing results of the diagnostic function of each characteristic value of the intrusion wave and the calculation and processing results of the weight of each characteristic value of the intrusion wave, the intrusion wave fusion processing is performed to obtain the intrusion wave evaluation result of the transformer winding deformation. Vibration fusion processing is performed based on the calculation results of the diagnostic function of each vibration characteristic value and the calculation results of the weight of each vibration characteristic value to obtain the vibration assessment results of the transformer winding deformation. The evaluation results are fused based on the intrusion wave assessment results and the vibration assessment results to obtain the fused diagnostic results of the intrusion wave and vibration on the transformer winding deformation.

[0039] Example 2: The difference from Embodiment 1 is that the transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis also includes a Wiener filter unit and a vibration denoising unit. The Wiener filtering unit is connected to the intrusion wave signal acquisition unit and the intrusion wave feature value extraction unit, respectively; the vibration denoising unit is connected to the vibration signal acquisition unit and the vibration feature value extraction unit, respectively; the truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit and the feature value weight calculation unit, respectively; and the feature value weight calculation unit is connected to the data fusion unit. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the Wiener filtering unit. The Wiener filtering unit filters the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The intrusion wave feature value extraction unit extracts the intrusion wave feature values ​​from the high-frequency noise filtering result and transmits it to the truncated normal fitting unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration denoising unit. The vibration denoising unit denoises the transformer vibration signal and transmits it to the vibration feature value extraction unit. The vibration feature value extraction unit extracts vibration feature values ​​from the denoising result and transmits it to the truncated normal fitting unit. The truncated normal fitting unit performs diagnostic function calculations on the intrusion wave feature value extraction results and the vibration feature value extraction results, and transmits the feature value diagnostic function calculation results to the feature value weight calculation unit. The feature value weight calculation unit performs feature value weight calculation based on the feature value diagnostic function calculation results, and transmits the feature value weight calculation results to the data fusion unit. The data fusion unit performs data fusion processing based on the feature value weight calculation results to determine whether the transformer winding has deformation and the degree of deformation.

[0040] An intrusion wave signal acquisition unit is connected to a Wiener filtering unit. The acquisition unit includes a voltage transformer module and a current transformer module. The voltage transformer module acquires the intrusion wave voltage signal and transmits it to the Wiener filtering unit, while the current transformer module acquires the intrusion wave current signal and transmits it to the Wiener filtering unit. The Wiener filtering unit receives the intrusion wave voltage and current signals from both modules and performs filtering processing on them. Specifically, the Wiener filtering unit performs high-frequency noise filtering on the intrusion wave voltage and current signals using its filter. The Wiener filtering unit then transmits the filtering result to an intrusion wave feature value extraction unit. This unit receives the filtering result, extracts intrusion wave feature values, and transmits the extracted feature values ​​to a truncated normal fit unit.

[0041] The vibration signal acquisition unit is connected to the vibration denoising unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration denoising unit. The vibration signal acquisition unit includes a vibration signal acquisition module and a vibration noise acquisition module. The vibration signal acquisition module acquires the vibration signal and transmits it to the vibration denoising unit, and the vibration noise acquisition module acquires the noise signal and transmits it to the vibration denoising unit. In other words, the transformer vibration signal includes both vibration and noise signals. The vibration denoising unit receives the vibration signal transmitted from the vibration signal acquisition module and the noise signal transmitted from the vibration noise acquisition module, and performs denoising processing on the vibration and noise signals. Specifically, the denoising processing of the vibration and noise signals includes subtraction denoising. The vibration denoising unit transmits the denoising processing result to the vibration feature value extraction unit, which receives the denoising processing result from the vibration denoising unit. The vibration feature value extraction unit then performs vibration feature value extraction processing on the denoising result and transmits the vibration feature value extraction result to the truncated normal fitting unit.

[0042] The vibration signal acquisition module includes one or more accelerometers. The module acquires vibration signals and transmits them to the vibration denoising unit. The specific number of accelerometers in the acquisition module can be selected based on actual usage. The accelerometers are installed on both sides of the transformer tank wall to collect transformer vibration signals. The accelerometers can be capacitive accelerometers; the type of accelerometer is not specifically defined in this application. The goal is to ensure that all acquired transformer vibration signals fall within the protection scope of this application. The vibration signal acquisition module is installed on both sides of the transformer tank wall to facilitate the acquisition of vibration acceleration waveforms. For example, six vibration signal acquisition points can be set on each side of the transformer tank wall, meaning two vibration signal acquisition points are set on each phase of the longitudinal transformer winding. The specific installation positions in the longitudinal direction can be at one-third or two-thirds of the height of the transformer tank side.

[0043] The vibration and noise acquisition module includes one or more accelerometers. The vibration and noise acquisition module acquires noise signals and transmits them to the vibration denoising unit. The specific number of accelerometers in the vibration and noise acquisition module can be selected according to the actual application. The accelerometers of the vibration and noise acquisition module are installed on the surface of the transformer's cooling system. The accelerometers of the vibration and noise acquisition module can be capacitive accelerometers, but this application does not specify a particular type.

[0044] The Wiener filtering unit is connected to both the intrusion wave signal acquisition unit and the intrusion wave feature value extraction unit. The intrusion wave signal acquisition unit transmits the intrusion wave signal, which includes both voltage and current signals, to the Wiener filtering unit. The Wiener filtering unit receives the intrusion wave signal transmitted from the acquisition unit, specifically the voltage signal from the voltage transformer module and the current signal from the current transformer module. The Wiener filtering unit performs high-frequency noise filtering on both the voltage and current signals. This high-frequency noise filtering specifically involves using the Wiener filtering unit's filter to perform high-frequency noise filtering on both the voltage and current signals. The Wiener filtering unit then transmits the filtering result to the intrusion wave feature value extraction unit.

[0045] The intrusion wave feature value extraction unit is connected to both the Wiener filtering unit and the truncated normal fitting unit. The intrusion wave feature value extraction unit receives the filtering results transmitted by the Wiener filtering unit and performs intrusion wave feature value extraction processing on the filtering results. Specifically, this processing includes extracting intrusion wave feature values ​​from the filtering results and calculating the intrusion wave voltage amplitude, intrusion wave current amplitude, dominant frequency of the intrusion wave voltage spectrum, dominant frequency of the intrusion wave current spectrum, entropy value of the intrusion wave voltage spectrum, and entropy value of the intrusion wave current spectrum. The intrusion wave feature value extraction unit then transmits the extracted feature values ​​to the truncated normal fitting unit.

[0046] The vibration denoising unit is connected to both the vibration signal acquisition unit and the vibration feature value extraction unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration denoising unit. The transformer vibration signal includes both vibration and noise signals. The vibration signal acquisition module acquires the vibration signal and transmits it to the vibration denoising unit, and the vibration noise acquisition module acquires the noise signal and transmits it to the vibration denoising unit. The vibration denoising unit receives both the vibration signal and noise signals transmitted from the vibration signal acquisition module and performs denoising processing on them. The vibration denoising unit then transmits the denoising results to the vibration feature value extraction unit. The vibration feature value extraction unit receives the denoising results from the vibration denoising unit and performs vibration feature value extraction processing on the denoising results. Specifically, this includes extracting vibration feature values ​​from the denoising results and calculating the low-frequency proportion of the vibration spectrum, the total harmonic distortion rate, and the vibration entropy.

[0047] The truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit, and the feature value weight calculation unit, respectively. The intrusion wave feature value extraction unit transmits the intrusion wave feature value extraction processing results to the truncated normal fitting unit, and the vibration feature value extraction unit transmits the vibration feature value extraction processing results to the truncated normal fitting unit. The truncated normal fitting unit receives the intrusion wave feature value extraction processing results transmitted by the intrusion wave feature value extraction unit, and the truncated normal fitting unit receives the vibration feature value extraction processing results transmitted by the vibration feature value extraction unit. In other words, the truncated normal fitting unit receives both the intrusion wave feature value extraction processing results and the vibration feature value extraction processing results. The truncated normal fitting unit performs diagnostic function calculations for each feature value in the intrusion wave feature value extraction processing results and the vibration feature value extraction processing results. Specifically, the diagnostic function calculations for each feature value in the truncated normal fitting unit include: The truncated normal fitting unit sets the domain of each eigenvalue, performs standard normal fitting, and obtains bell-shaped curves for each eigenvalue pair. The bell-shaped curves are then truncated according to the eigenvalue domain range, and the truncated curves are normalized. The truncated normal curves of normal data are compared with those of fault data to obtain the corresponding diagnostic functions for each eigenvalue. The truncated normal fitting unit transmits the diagnostic function calculation results to the eigenvalue weight calculation unit. The eigenvalue weight calculation unit receives the diagnostic function calculation results from the truncated normal fitting unit and performs eigenvalue weight calculation based on these results. Specifically, the eigenvalue weight calculation unit calculates the coefficient of variation for each eigenvalue extracted from the intrusive wave and vibration eigenvalues, and uses the relative proportion of each coefficient of variation calculation result as the eigenvalue weight of the corresponding diagnostic function calculation result.

[0048] The eigenvalue weight calculation unit is connected to the data fusion unit. The eigenvalue weight calculation unit transmits the processing results of each eigenvalue weight calculation to the data fusion unit. The data fusion unit receives the processing results of each eigenvalue weight calculation transmitted by the eigenvalue weight calculation unit and performs data fusion processing based on these results to determine whether the transformer windings are deformed and the degree of deformation. The data fusion processing performed by the data fusion unit based on the processing results of each eigenvalue weight calculation specifically includes: Based on the calculation and processing results of the diagnostic function of each characteristic value of the intrusion wave and the calculation and processing results of the weight of each characteristic value of the intrusion wave, the intrusion wave fusion processing is performed to obtain the intrusion wave evaluation result of the transformer winding deformation. Vibration fusion processing is performed based on the calculation results of the diagnostic function of each vibration characteristic value and the calculation results of the weight of each vibration characteristic value to obtain the vibration assessment results of the transformer winding deformation. The evaluation results are fused based on the intrusion wave assessment results and the vibration assessment results to obtain the fused diagnostic results of the intrusion wave and vibration on the transformer winding deformation.

[0049] Example 3: The difference from Embodiment 2 is that the transformer winding deformation diagnosis and determination system based on the fusion of intrusion wave and vibration signal diagnosis further includes an anomaly detection unit. This anomaly detection unit is used to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal. The anomaly detection unit determines whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal by including the following steps: The vibration signal waveform is standardized based on the transformer voltage during vibration measurement. The standardization calculation formula is shown below: A2 = A1 / U 2 Where A1 is the transformer vibration signal before standardization, A2 is the transformer vibration signal after standardization, and U is the voltage during vibration measurement.

[0050] The standardized vibration signal is compared with historical vibration data under normal operation in the database; If the standardized vibration amplitude is greater than the historical vibration amplitude threshold, it is determined that the transformer vibration signal is abnormal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into working state. If the standardized vibration amplitude is less than or equal to the historical vibration amplitude threshold, it is determined that there is no abnormality in the transformer vibration signal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into a dormant state.

[0051] The terms "first," "second," and "third," etc., used in this application's specification and the foregoing drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0052] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A system for diagnosing and determining transformer winding deformation based on the fusion of intrusion wave and vibration signal diagnostics, characterized in that, The system includes an intrusion wave signal acquisition unit, an intrusion wave feature value extraction unit, a truncated normal fitting unit, a feature value weight calculation unit, a vibration signal acquisition unit, a vibration feature value extraction unit, a data fusion unit, a Wiener filtering unit, and a vibration denoising unit. The intrusion wave signal acquisition unit is connected to the intrusion wave feature value extraction unit, the vibration signal acquisition unit is connected to the vibration feature value extraction unit, the truncated normal fitting unit is connected to the intrusion wave feature value extraction unit, the vibration feature value extraction unit, and the feature value weight calculation unit, and the feature value weight calculation unit is connected to the data fusion unit. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The intrusion wave feature value extraction unit performs intrusion wave feature value extraction processing on the intrusion wave signal and transmits it to the truncated normal fitting unit. The unit comprises: a vibration signal acquisition unit that acquires transformer vibration signals and transmits them to a vibration feature value extraction unit; a vibration feature value extraction unit that extracts vibration feature values ​​from the transformer vibration signals and transmits them to a truncated normal fitting unit; a truncated normal fitting unit that calculates diagnostic functions for each feature value based on the extracted wave feature value and vibration feature value, and transmits the diagnostic function calculation results to a feature value weight calculation unit; a feature value weight calculation unit that calculates feature value weights based on the diagnostic function calculation results and transmits the result to a data fusion unit; and a data fusion unit that performs data fusion based on the feature value weight calculation results to determine whether the transformer windings are deformed and the degree of deformation. Specifically, the truncated normal fitting unit performs diagnostic function calculations on the intrusion wave feature value extraction results and vibration feature value extraction results, including: The truncated normal fitting unit sets the domain of each eigenvalue, performs standard normal fitting, and obtains bell curves for each eigenvalue pair; the bell curves for each eigenvalue pair are truncated according to the domain of the eigenvalues, and the truncated curves are normalized; the truncated normal curves of normal data are compared with the truncated normal curves of fault data to obtain the corresponding diagnostic functions for each eigenvalue. Specifically, the eigenvalue weight calculation unit performs eigenvalue weight calculation based on the eigenvalue diagnostic function calculation results. This includes: the eigenvalue weight calculation unit calculates the coefficient of variation of each eigenvalue for the intrusion wave eigenvalue extraction and vibration eigenvalue extraction results, and uses the relative proportion of each eigenvalue coefficient of variation calculation result as the eigenvalue weight of the corresponding eigenvalue diagnostic function calculation result. It also includes an anomaly detection unit, which is used to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal; The anomaly detection unit is used to determine whether there is an anomaly in the transformer vibration signal based on the transformer voltage signal and the transformer vibration signal, including the following steps: The vibration signal waveform is standardized based on the transformer voltage during vibration measurement; The standardized vibration signal is compared with historical vibration data under normal operation in the database; If the standardized vibration amplitude is greater than the historical vibration amplitude threshold, it is determined that the transformer vibration signal is abnormal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into working state. If the standardized vibration amplitude is less than or equal to the historical vibration amplitude threshold, it is determined that there is no abnormality in the transformer vibration signal, and the intrusion wave feature value extraction unit, truncated normal fitting unit, feature value weight calculation unit, vibration feature value extraction unit, data fusion unit, Wiener filtering unit and vibration denoising unit are put into a dormant state.

2. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to claim 1, characterized in that, The intrusion wave signal acquisition unit includes a voltage transformer module and a current transformer module. The voltage transformer module acquires the intrusion wave voltage signal and transmits it to the intrusion wave feature value extraction unit. The current transformer module acquires the intrusion wave current signal and transmits it to the intrusion wave feature value extraction unit.

3. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to claim 1, characterized in that, The intrusion wave feature value extraction unit specifically includes the following steps for the intrusion wave signal to be processed: extracting intrusion wave feature values ​​from the intrusion wave signal and calculating the intrusion wave voltage amplitude, intrusion wave current amplitude, intrusion wave voltage spectrum main frequency, intrusion wave current spectrum main frequency, intrusion wave voltage spectrum entropy value, and intrusion wave current spectrum entropy value.

4. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to claim 1, characterized in that, The vibration feature extraction unit performs vibration feature extraction processing on the transformer vibration signal, specifically including: extracting vibration feature values ​​from the transformer vibration signal, and calculating the low-frequency proportion of the vibration spectrum, the total harmonic distortion rate of the vibration, and the vibration entropy.

5. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to claim 1, characterized in that, The data fusion unit performs data fusion processing based on the weight calculation results of each feature value, specifically including: Based on the calculation and processing results of the diagnostic function of each characteristic value of the intrusion wave and the calculation and processing results of the weight of each characteristic value of the intrusion wave, the intrusion wave fusion processing is performed to obtain the intrusion wave evaluation result of the transformer winding deformation. Vibration fusion processing is performed based on the calculation results of the diagnostic function of each vibration characteristic value and the calculation results of the weight of each vibration characteristic value to obtain the vibration assessment results of the transformer winding deformation. The evaluation results are fused based on the intrusion wave assessment results and the vibration assessment results to obtain the fused diagnostic results of the intrusion wave and vibration on the transformer winding deformation.

6. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to claim 1, characterized in that, The vibration signal acquisition unit includes a vibration signal acquisition module, which acquires vibration signals and transmits them to the vibration feature value extraction unit.

7. The transformer winding deformation determination system based on the fusion of intrusion wave and vibration signal diagnosis according to any one of claims 1-6, characterized in that, The Wiener filtering unit is connected to the intrusion wave signal acquisition unit and the intrusion wave feature value extraction unit, respectively. The vibration denoising unit is connected to the vibration signal acquisition unit and the vibration feature value extraction unit, respectively. The intrusion wave signal acquisition unit acquires the intrusion wave signal and transmits it to the Wiener filtering unit. The Wiener filtering unit filters the intrusion wave signal and transmits it to the intrusion wave feature value extraction unit. The vibration signal acquisition unit acquires the transformer vibration signal and transmits it to the vibration denoising unit. The vibration denoising unit denoises the transformer vibration signal and transmits it to the vibration feature value extraction unit.

Images