Converter transformer ultra-high frequency partial discharge high-speed acquisition and defect positioning system

Abstract

This invention discloses a high-speed acquisition and defect location system for ultra-high frequency partial discharge (UHF) in converter transformers, relating to the field of UHF detection. It includes: an acquisition module for real-time acquisition of raw UHF UHF signals and temperature-related signals from key parts of the winding; and a conditioning module for performing dielectric-adaptive amplitude normalization on the acquired signals and simultaneously suppressing non-partial discharge electromagnetic interference signals. This invention efficiently extracts multi-dimensional characteristic parameters related to insulation defects by real-time capture of relevant raw signals and temperature-related data, followed by adaptive amplitude normalization and precise interference suppression. The system then combines these parameters with a typical defect map library to accurately determine the defect type.

Description

Technical Field

[0001] This invention relates to the field of partial discharge detection technology, specifically a high-speed acquisition and defect location system for ultra-high frequency partial discharge in converter transformers. Background Technology

[0002] The converter transformer is the core equipment connecting the AC system and the DC system. Its main task is to provide the converter with a suitable AC voltage and to achieve electrical isolation between the two systems.

[0003] The invention patent application with application number 202311632471.X discloses a GIS partial discharge diagnosis system, which aims to solve the problem that "in the current process of GIS partial discharge diagnosis, there is a lot of time for human intervention, the diagnosis time is long, and the efficiency of GIS partial discharge diagnosis is low".

[0004] However, existing technologies for detecting UHF partial discharge in converter transformers often face problems such as signal acquisition being susceptible to electromagnetic interference, signal distortion caused by differences in dielectric properties, incomplete defect feature extraction, insufficient accuracy in type identification, and defect location being affected by propagation attenuation.

[0005] To address this, we propose a high-speed acquisition and defect location system for UHF partial discharge in converter transformers. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a high-speed acquisition and defect location system for ultra-high frequency partial discharge of converter transformer, which can effectively solve the problems of the existing technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions;

[0008] This invention discloses a high-speed acquisition and defect location system for ultra-high frequency partial discharge of a converter transformer, comprising:

[0009] The system comprises the following modules: an acquisition module for real-time acquisition of UHF partial discharge signals and temperature-related signals from key winding components; a conditioning module for medium-adaptive amplitude normalization of the acquired signals and simultaneous suppression of non-partial discharge electromagnetic interference signals; an extraction module for receiving the processed signals and extracting pulse distortion characteristics, spectral distribution characteristics, and amplitude variation trend parameters related to converter transformer insulation defects; an identification module for determining the specific type of partial discharge defect based on a pre-defined database of typical converter transformer defect discharge characteristic maps, using feature parameter correlation matching and discharge mode comparison; a positioning module for calculating the spatial coordinates of the defect based on a three-dimensional layout model of the converter transformer windings and core, and the differences in propagation attenuation of partial discharge signals in different media; and a calibration module for generating a load-adaptive standard calibration signal based on the real-time operating load parameters of the converter transformer to dynamically correct the deviation between signal acquisition and positioning calculation.

[0010] The acquisition module is interconnected with the conditioning module and the extraction module via a wireless network. The extraction module is interconnected with the identification module via a wireless network. The identification module is interconnected with the positioning module via a wireless network. The positioning module is interconnected with the calibration module via a wireless network.

[0011] In the acquisition module, the key parts of the winding are customized by the system user.

[0012] Furthermore, the acquisition module includes an ultra-high frequency sensor array and a temperature sensing unit;

[0013] The sampling frequency of the ultra-high frequency sensor array is dynamically adapted based on the rated voltage level of the converter transformer; the temperature sensing unit adopts a combination of contact and non-contact sensing structures to simultaneously collect real-time temperature values, ambient reference temperature values, and temperature change rates of key parts of the winding.

[0014] The response time of the temperature sensing unit is limited to no more than 1 / 5 of the duration of the partial discharge signal.

[0015] Furthermore, when the system-side user customizes key parts of the winding, the system provides a visual 3D modeling interface. The system-side user determines the location of the key parts by selecting parameters in three dimensions: winding layer, number of winding segments, and distance from the iron core. At the same time, the system performs a rationality check on the user-defined key parts based on the converter transformer structural design specifications. If the customized part exceeds the range of the converter transformer's physical structure or there is a signal acquisition blind zone, the system issues a prompt to inform the system-side user to change the customized part.

[0016] Furthermore, the amplitude normalization processing for media adaptation in the conditioning module is as follows:

[0017] ;

[0018] In the formula: The normalized signal amplitude; The original signal amplitude input to the acquisition module; The dielectric constant of the preset reference medium; The dielectric constant of the actual dielectric medium inside the converter transformer; This is the correction factor for dielectric loss; This is the actual frequency of the partial discharge signal; The reference frequency for the partial discharge signal is 10 times the rated frequency of the transformer.

[0019] When suppressing the non-partial discharge electromagnetic interference signal, two suppression thresholds are set. The first threshold is the upper limit of the interference signal amplitude set based on historical defect-free operation data, and the second threshold is the signal rising edge slope threshold. Only signals that simultaneously satisfy the condition that the amplitude exceeds the first threshold and the rising edge slope is greater than the second threshold are retained.

[0020] Furthermore, the pulse distortion features extracted by the extraction module include the pulse leading edge distortion degree, the pulse trailing edge attenuation coefficient, and the pulse top flatness.

[0021] The pulse leading edge distortion ;

[0022] In the formula: The rise time of the pulse leading edge is selected as the time from 10% to 90% of the peak value; This represents the instantaneous amplitude of the actual pulse signal; The instantaneous amplitude of the preset ideal pulse signal; This represents the peak value of the actual pulse signal.

[0023] The spectral distribution characteristics include the characteristic frequency amplitude ratio, the spectral centroid frequency, and the spectral bandwidth. The amplitude change trend parameters include the variance of the amplitude peak sequence, the amplitude change period, and the amplitude decay index.

[0024] Furthermore, the feature parameter association and matching operation in the recognition module follows the following rules:

[0025] ;

[0026] In the formula: The similarity value is associated with the feature parameters; This represents the total number of feature parameters involved in the matching. The weight coefficient for the i-th feature parameter; This represents the value of the i-th feature parameter extracted so far. This represents the i-th standard feature parameter value corresponding to the defect type in the typical defect discharge feature map library;

[0027] when When the similarity is less than the preset threshold, the discharge mode comparison process is initiated. By comparing the discharge phase distribution and pulse sequence interval characteristics of the current signal with the fit of the standard mode in the spectrum library, the type of partial discharge defect is determined.

[0028] Furthermore, the formula for calculating the degree of fit is:

[0029] ;

[0030] In the formula: Discharge mode compatibility; , These are the weighting coefficients; Let be the discharge phase distribution function of the current signal; This represents the discharge phase distribution function of the standard modes in the spectrum library; This is the interval difference correction factor; The average interval of the pulse sequence of the current signal; The average interval of the pulse sequence in the standard mode;

[0031] when When the degree of partial discharge exceeds the preset fit threshold, the partial discharge defect is ultimately determined to be the type corresponding to the standard mode.

[0032] Furthermore, in the defect spatial coordinate calculation stage of the positioning module, the coordinates are calculated based on a three-dimensional Cartesian coordinate system, with the geometric center of the converter transformer core as the origin.

[0033] ;

[0034] In the formula: For the defect space coordinates; These represent the propagation speeds of partial discharge signals in the two different media of the converter transformer; This is the preset base time; These represent the times when the UHF sensors at three different locations received the partial discharge signal; The vertical distance from the plane containing the three sensors to the origin of the coordinate system; The horizontal angle between the line connecting the first sensor and the origin; These are the attenuation correction coefficients for the signal propagation paths corresponding to the three sensors.

[0035] Furthermore, the standard calibration signal for load adaptation in the calibration module is generated based on a dynamic calibration model constructed from the real-time operating load parameters of the converter transformer. The real-time operating load parameters include load current, load power factor, and winding hot spot temperature. The amplitude and frequency of the standard calibration signal are output through the dynamic calibration model.

[0036] ;

[0037] In the formula: The amplitude of the standard calibration signal; The amplitude is calibrated as a reference. This is the load current; This is the rated current of the converter transformer; The load power factor; This refers to the hot spot temperature of the winding. This refers to the rated hot spot temperature of the converter transformer. The frequency of the standard calibration signal; The reference calibration frequency; This is the load adaptability factor.

[0038] Compared with the known prior art, the technical solution provided by this invention has the following beneficial effects:

[0039] This invention captures relevant raw signals and temperature-related data in real time, performs adaptive amplitude normalization and precise interference suppression, efficiently extracts multi-dimensional feature parameters related to insulation defects, accurately determines the defect type by combining a typical defect map library, and accurately calculates the spatial coordinates of the defect based on the three-dimensional layout model and propagation attenuation differences. It also dynamically calibrates and corrects deviations using real-time load parameters. Furthermore, the system supports user-defined key monitoring locations and allows for rationality verification, adapting to different voltage levels and dielectric environments. The feature extraction and matching algorithms balance comprehensiveness and specificity, and the positioning calculation incorporates multi-dimensional correction factors, effectively improving the completeness of partial discharge signal acquisition, the accuracy of defect identification, and the precision of positioning. This reduces errors caused by environmental interference and load changes, providing support for insulation status monitoring and fault early warning in converter transformer equipment, and ensuring the safe and stable operation of the equipment. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0041] Figure 1 This is a schematic diagram of a converter transformer high-frequency partial discharge high-speed acquisition and defect location system. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0043] The present invention will be further described below with reference to embodiments.

[0044] Example 1:

[0045] This embodiment provides a high-speed acquisition and defect location system for ultra-high frequency partial discharge in a converter transformer, such as... Figure 1 As shown, it includes:

[0046] The acquisition module is used to acquire the raw signal of UHF partial discharge and the temperature-related signal of key parts of the winding in real time.

[0047] The acquisition module includes an ultra-high frequency sensor array and a temperature sensing unit;

[0048] The sampling frequency of the ultra-high frequency sensor array is dynamically adapted based on the rated voltage level of the converter transformer, enabling distortion-free capture of partial discharge pulse signals; the temperature sensing unit adopts a combination of contact and non-contact sensing structures to simultaneously collect real-time temperature values, ambient reference temperature values, and temperature change rates of key parts of the winding.

[0049] The response time of the temperature sensing unit is limited to no more than 1 / 5 of the duration of the partial discharge signal;

[0050] When system users define key parts of the winding, the system provides a visual 3D modeling interface. System users can determine the location of key parts by selecting parameters in three dimensions: winding layer, number of winding segments, and distance from the iron core. At the same time, the system verifies the rationality of user-defined key parts based on the converter transformer structural design specifications. If the defined part exceeds the range of the converter transformer physical structure or there is a signal acquisition blind zone, the system will issue a prompt to inform the system user to change the defined part.

[0051] The conditioning module is used to perform medium-adaptive amplitude normalization processing on the acquired signal and simultaneously suppress non-partial discharge electromagnetic interference signals.

[0052] The amplitude normalization process for media adaptation in the conditioning module is as follows:

[0053] ;

[0054] In the formula: The normalized signal amplitude; The original signal amplitude input to the acquisition module; The dielectric constant of the preset reference medium; The dielectric constant of the actual dielectric medium inside the converter transformer; This is the correction factor for dielectric loss; This is the actual frequency of the partial discharge signal; The reference frequency for the partial discharge signal is 10 times the rated frequency of the transformer.

[0055] The above formula fully considers the difference between the actual dielectric constant of the converter transformer and the preset reference value, and also takes into account the influence of the change in the actual frequency of the partial discharge signal relative to the reference frequency. The signal amplitude is precisely adjusted by the dielectric loss correction coefficient to ensure that the amplitude of the acquired signal under different dielectric environments and frequency conditions is comparable. The value of the dielectric loss correction coefficient is in line with the actual state of the dielectric itself, such as aging, humidity, and impurities, so that the amplitude regularity is more in line with the actual operating conditions of the equipment.

[0056] When suppressing non-partial discharge electromagnetic interference signals, two suppression thresholds are set. The first threshold is the upper limit of the interference signal amplitude set based on historical defect-free operation data, and the second threshold is the signal rising edge slope threshold. Only signals that simultaneously meet the conditions of having an amplitude exceeding the first threshold and a rising edge slope greater than the second threshold are retained.

[0057] in, The preset value range is [0.01, 0.08]. Its value increases with the aging degree, humidity content and impurity concentration of the medium inside the converter transformer, and decreases with the improvement of medium purity, stability and cleanliness of the operating environment.

[0058] The extraction module is used to receive the signal processed by the conditioning module and extract pulse distortion characteristics, spectral distribution characteristics and amplitude variation trend parameters related to the insulation defects of the converter transformer.

[0059] The pulse distortion features extracted by the extraction module include the pulse leading edge distortion degree, the pulse trailing edge attenuation coefficient, and the pulse top flatness.

[0060] Pulse leading edge distortion ;

[0061] In the formula: The rise time of the pulse leading edge is selected as the time from 10% to 90% of the peak value; This represents the instantaneous amplitude of the actual pulse signal; The instantaneous amplitude of the preset ideal pulse signal; This represents the peak value of the actual pulse signal.

[0062] The above formula calculates the integral of the deviation between the actual pulse signal and the ideal pulse signal during the rise time of the leading edge, and then normalizes it by combining the actual pulse peak value and the rise time of the leading edge. This can accurately quantify the distortion degree of the leading edge of the pulse, provide intuitive and accurate feature parameters for subsequent insulation defect identification, and effectively reflect the waveform distortion characteristics of the partial discharge signal caused by the presence of defects.

[0063] The spectral distribution characteristics include the proportion of characteristic frequency amplitude, the spectral centroid frequency, and the spectral bandwidth. The amplitude change trend parameters include the variance of the amplitude peak sequence, the amplitude change period, and the amplitude decay index. Each characteristic parameter is obtained synchronously through the time window sliding extraction method, and the time window length is adapted to the typical duration of the partial discharge signal.

[0064] The identification module is used to determine the specific type of partial discharge defect based on a preset library of discharge feature maps of typical defects in converter transformers, by matching feature parameters and comparing discharge modes.

[0065] The feature parameter association and matching operation in the recognition module follows the following rules:

[0066] ;

[0067] In the formula: The similarity value is associated with the feature parameters; This represents the total number of feature parameters involved in the matching. The weight coefficient for the i-th feature parameter; This represents the value of the i-th feature parameter extracted so far. This represents the i-th standard feature parameter value corresponding to the defect type in the typical defect discharge feature map library;

[0068] The above formula assigns a weight coefficient to each feature parameter involved in the matching. The weight is determined based on its contribution to distinguishing different defect types. Then, by calculating the relative deviation between the current feature parameter and the standard feature parameter and summing them by weight, the degree of matching of the feature parameters is scientifically quantified. The setting of the weight coefficient makes the similarity calculation more in line with the core needs of defect identification, thereby improving the accuracy of the matching results.

[0069] when When the similarity is less than the preset threshold, the discharge mode comparison process is initiated. By comparing the discharge phase distribution and pulse sequence interval characteristics of the current signal with the fit of the standard mode in the spectrum library, the type of partial discharge defect is determined.

[0070] in, The value range is preset to (0,1), and the sum of all weight coefficients is 1. The higher the contribution rate to the differentiation of different defect types, the larger the value; the lower the contribution rate, the smaller the value.

[0071] The formula for calculating compatibility is:

[0072] ;

[0073] In the formula: Discharge mode compatibility; , These are the weighting coefficients; Let be the discharge phase distribution function of the current signal; This represents the discharge phase distribution function of the standard modes in the spectrum library; This is the interval difference correction factor; The average interval of the pulse sequence of the current signal; The average interval of the pulse sequence in the standard mode;

[0074] The above formula comprehensively considers the correlation of discharge phase distribution and the difference of pulse sequence interval. It balances the proportion of the two indicators in the fit evaluation by weighting coefficients. The correlation of phase distribution is calculated in integral form, and the pulse sequence interval is adjusted by introducing a correction coefficient to adjust the degree of influence of the difference. It not only fully covers the key characteristics of discharge mode, but also reasonably adapts the sensitivity difference of the two indicators according to different defect types, thereby improving the reliability of defect type determination.

[0075] when When the degree of partial discharge exceeds the preset fit threshold, the final determination is that the partial discharge defect is of the type corresponding to the standard mode;

[0076] in, , All are positive numbers, and their sum is 1. The preset value range is [0.1, 1.5]. The value is larger when the pulse sequence interval is more sensitive to the distinction of defect types, and smaller when the pulse sequence interval accounts for a smaller proportion of defect features.

[0077] The positioning module is used to calculate the spatial coordinates of defects based on the three-dimensional layout model of the converter transformer winding and the iron core, as well as the difference in propagation attenuation of partial discharge signals in different media.

[0078] In the defect spatial coordinate calculation stage of the positioning module, the coordinates are obtained based on a three-dimensional rectangular coordinate system, with the geometric center of the converter transformer core as the origin.

[0079] ;

[0080] In the formula: For the defect space coordinates; These represent the propagation speeds of partial discharge signals in the two different media of the converter transformer; This is the preset base time; These represent the times when the UHF sensors at three different locations received the partial discharge signal; The vertical distance from the plane containing the three sensors to the origin of the coordinate system; The horizontal angle between the line connecting the first sensor and the origin; These are the attenuation correction coefficients for the signal propagation paths corresponding to the three sensors;

[0081] The above formula constructs a three-dimensional coordinate system with the geometric center of the converter core as the origin. Combining the difference in the propagation speed of partial discharge signals in different media, a propagation attenuation correction coefficient is introduced to correct the signal propagation time. The attenuation correction coefficient comprehensively considers factors such as the dielectric constant of the medium, temperature, and propagation path length. By using the signal reception time of sensors at three different positions, the three-dimensional spatial coordinates of the defect are accurately solved.

[0082] This is a sign function used to determine the z-axis direction;

[0083] ;

[0084] In the formula: This is the propagation attenuation correction factor; The reference coefficient for dielectric attenuation and the coefficient for temperature influence; This is the propagation path length of the partial discharge signal from the defect point to the corresponding sensor; The dielectric constant of the actual medium along the propagation path; The dielectric constant of the preset reference medium; The average temperature along the propagation path; This is the preset reference temperature;

[0085] The above formula takes into account the dielectric characteristics, temperature conditions and path length on the partial discharge signal propagation path. It constructs a functional relationship through the dielectric attenuation reference coefficient and temperature influence coefficient, compares the actual dielectric constant and average temperature of the dielectric with the preset reference value, quantifies the influence of these factors on signal propagation attenuation, and enables the correction coefficient to dynamically adapt to different propagation environments, so as to improve the accuracy of signal propagation time calculation in defect location.

[0086] The value of 'a' is preset to [0.001, 0.05], and its value decreases as the insulation strength of the medium inside the converter transformer increases, and increases as the impurity content, humidity and aging degree of the medium increase; the value of 'b' is preset to [0.002, 0.08], and its value increases as the temperature sensitivity coefficient of the medium inside the converter transformer increases, and decreases as the thermal stability of the medium increases.

[0087] The calibration module is used to generate a standard calibration signal for load adaptation based on the real-time operating load parameters of the converter transformer, so as to dynamically correct the deviation of signal acquisition and positioning calculation.

[0088] The standard calibration signal for load adaptation in the calibration module is generated based on a dynamic calibration model constructed from the real-time operating load parameters of the converter transformer. The real-time operating load parameters include load current, load power factor, and winding hot spot temperature. The amplitude and frequency of the standard calibration signal are output through the dynamic calibration model.

[0089] ;

[0090] In the formula: The amplitude of the standard calibration signal; The amplitude is calibrated as a reference. This is the load current; This is the rated current of the converter transformer; The load power factor; This refers to the hot spot temperature of the winding. This refers to the rated hot spot temperature of the converter transformer. The frequency of the standard calibration signal; The reference calibration frequency; This refers to the load adaptability factor.

[0091] The above formula is based on key parameters such as load current, power factor, and winding hot spot temperature of the converter transformer in real time. These parameters are associated with the reference calibration value through the load adaptation coefficient. The load adaptation coefficient is determined by fitting the equipment model and historical operating data, and follows the preset value range and constraints to ensure that the generated standard calibration signal can dynamically match different operating load states of the equipment, thereby effectively correcting the deviation in the signal acquisition and positioning calculation process.

[0092] in, The converter transformer model is determined by fitting historical operating data, and is pre-defined to follow the following:

[0093] a′∈[0.3,0.6], b′∈ [0.2,0.4], c′∈[0.1,0.3], d′∈[0.15,0.45], e′∈[0.25,0.55], and a′+b′+c′≤1.2, d′+e′≤0.9;

[0094] Among them, the key parts of the winding in the acquisition module are defined by the system user;

[0095] The acquisition module is interconnected with the conditioning module and the extraction module via a wireless network. The extraction module is interconnected with the identification module via a wireless network. The identification module is interconnected with the positioning module via a wireless network. The positioning module is interconnected with the calibration module via a wireless network.

[0096] In this embodiment, the acquisition module acquires the original UHF partial discharge signal and the temperature-related signal of key parts of the winding in real time. The conditioning module performs medium-adaptive amplitude normalization processing on the acquired signal and simultaneously suppresses non-partial discharge electromagnetic interference signals. The extraction module further receives the signal processed by the conditioning module and extracts pulse distortion characteristics, spectral distribution characteristics, and amplitude change trend parameters related to the converter transformer insulation defects. Then, the identification module determines the specific type of partial discharge defect based on a preset typical defect discharge feature spectrum library of converter transformers by matching feature parameters and comparing discharge modes. The positioning module calculates the spatial coordinates of the defect based on the three-dimensional layout model of the converter transformer winding and core and the difference in propagation attenuation of the partial discharge signal in different media. Finally, the calibration module generates a standard calibration signal for load adaptation based on the real-time operating load parameters of the converter transformer to dynamically correct the deviation between signal acquisition and positioning calculation.

[0097] In the above embodiments, the system can accurately capture converter transformer-related signals, effectively eliminate interference, accurately identify insulation defect types and locate spatial positions, and dynamically correct deviations based on operating load, thereby improving the accuracy of signal acquisition and positioning. This helps to promptly identify potential hazards, reduce fault risks, and extend equipment lifespan.

[0098] Referring to the system in the above embodiments, the following is an example of the application of this system:

[0099] In daily operation, a 500kV converter transformer at the XX South District substation uses this system to monitor its insulation status.

[0100] After the system starts up, the UHF sensor array in the acquisition module dynamically adapts the sampling frequency according to the rated voltage level of the converter transformer, and captures the original UHF partial discharge signal without distortion; the temperature sensing unit adopts a combination structure of contact and non-contact, and synchronously acquires the real-time temperature value, ambient reference temperature value and temperature change rate of the key parts of the winding customized by the user through the visual 3D modeling interface, and the customized parts are verified to be normal by the system.

[0101] The conditioning module first performs amplitude normalization processing on the acquired raw signal to adapt to the medium. Combined with parameters such as the dielectric constant of the actual medium inside the converter transformer, the normalized signal amplitude is finally 1.03 times that of the original signal amplitude. Then, a dual suppression threshold is set to retain signals whose amplitude exceeds the upper limit of the interference signal amplitude set by the historical defect-free operation data and whose rising edge slope is greater than the set threshold, effectively suppressing non-partial discharge electromagnetic interference.

[0102] The extraction module uses a time window sliding extraction method to simultaneously acquire pulse distortion features such as pulse leading edge distortion degree and pulse trailing edge attenuation coefficient, spectral distribution features such as characteristic frequency amplitude ratio and spectral centroid frequency, and amplitude change trend parameters such as variance of amplitude peak sequence and amplitude change period.

[0103] The identification module correlates and matches the extracted feature parameters with a preset library of typical converter transformer defect discharge feature maps. The calculated feature parameter correlation similarity value is 0.12, which is less than the preset similarity threshold, thus initiating the discharge mode comparison process. By comparing the discharge phase distribution and pulse sequence interval characteristics of the current signal with those of standard modes in the map library, the discharge mode fit is found to be 0.85, which is greater than the preset fit threshold. Therefore, it is ultimately determined that the converter transformer has a partial discharge defect, classified as an insulation aging defect.

[0104] The positioning module is based on the three-dimensional layout model of the converter winding and the iron core. A three-dimensional rectangular coordinate system is established with the geometric center of the iron core as the origin. Combining the propagation speed of the partial discharge signal in different media and the time when the UHF sensor at three different positions receives the signal, and considering the attenuation correction coefficient of the propagation path, the spatial coordinates of the defect are calculated to be (125mm, 86mm, 210mm).

[0105] The calibration module generates a standard calibration signal for load adaptation based on the real-time load current, load power factor, and winding hot spot temperature of the converter transformer through a dynamic calibration model. The amplitude of the standard calibration signal is 1.1 times that of the reference calibration signal, and the frequency is 1.08 times that of the reference calibration frequency. This dynamically corrects the deviation between signal acquisition and positioning calculation, ensuring accurate and reliable monitoring results.

[0106] Based on the monitoring results of the system, the operation and maintenance personnel promptly addressed the insulation aging defect, effectively preventing the converter transformer from failing and ensuring the safe and stable operation of the substation.

[0107] In summary, the system in the above embodiments efficiently extracts multi-dimensional feature parameters related to insulation defects by capturing relevant raw signals and temperature-related data in real time, performing adaptive amplitude normalization and precise interference suppression, accurately determining the defect type by combining a typical defect map library, and accurately calculating the spatial coordinates of defects based on the three-dimensional layout model and propagation attenuation differences. It also dynamically calibrates and corrects deviations using real-time load parameters. Furthermore, the system supports user-defined key monitoring locations and allows for rationality verification, adapting to different voltage levels and dielectric environments. The feature extraction and matching algorithms balance comprehensiveness and specificity, and the positioning calculation incorporates multi-dimensional correction factors, effectively improving the completeness of partial discharge signal acquisition, the accuracy of defect identification, and the precision of positioning. This reduces errors caused by environmental interference and load changes, providing support for insulation status monitoring and fault early warning of converter transformer equipment, and ensuring the safe and stable operation of the equipment.

[0108] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A high-speed acquisition and defect location system for ultra-high frequency partial discharge of a converter transformer, characterized in that, include: The acquisition module is used to acquire the raw signal of UHF partial discharge and the temperature-related signal of key parts of the winding in real time. The conditioning module is used to perform medium-adaptive amplitude normalization processing on the acquired signal and simultaneously suppress non-partial discharge electromagnetic interference signals. The extraction module is used to receive the signal processed by the conditioning module and extract pulse distortion characteristics, spectral distribution characteristics and amplitude variation trend parameters related to the insulation defects of the converter transformer. The identification module is used to determine the specific type of partial discharge defect based on a preset library of discharge feature maps of typical defects in converter transformers, by matching feature parameters and comparing discharge modes. The positioning module is used to calculate the spatial coordinates of defects based on the three-dimensional layout model of the converter transformer winding and the iron core, as well as the difference in propagation attenuation of partial discharge signals in different media. The calibration module is used to generate a standard calibration signal for load adaptation based on the real-time operating load parameters of the converter transformer, so as to dynamically correct the deviation of signal acquisition and positioning calculation. In the acquisition module, the key parts of the winding are customized by the system user.

2. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The acquisition module includes an ultra-high frequency sensor array and a temperature sensing unit; The sampling frequency of the ultra-high frequency sensor array is dynamically adapted based on the rated voltage level of the converter transformer; the temperature sensing unit adopts a combination of contact and non-contact sensing structures to simultaneously collect real-time temperature values, ambient reference temperature values, and temperature change rates of key parts of the winding. The response time of the temperature sensing unit is limited to no more than 1 / 5 of the duration of the partial discharge signal.

3. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, When the system user defines key parts of the winding, the system provides a visual 3D modeling interface. The system user determines the location of the key parts by selecting parameters in three dimensions: winding layer, number of winding segments, and distance from the iron core. At the same time, the system verifies the rationality of the user-defined key parts based on the converter transformer structural design specifications. If the defined part exceeds the range of the converter transformer physical structure or there is a signal acquisition blind zone, the system issues a prompt to inform the system user to change the defined part.

4. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The amplitude normalization process for media adaptation in the conditioning module is as follows: ； In the formula: The normalized signal amplitude; The original signal amplitude input to the acquisition module; The dielectric constant of the preset reference medium; The dielectric constant of the actual dielectric medium inside the converter transformer; This is the correction factor for dielectric loss; This is the actual frequency of the partial discharge signal; The reference frequency for the partial discharge signal is 10 times the rated frequency of the transformer. When suppressing the non-partial discharge electromagnetic interference signal, two suppression thresholds are set. The first threshold is the upper limit of the interference signal amplitude set based on historical defect-free operation data, and the second threshold is the signal rising edge slope threshold. Only signals that simultaneously satisfy the condition that the amplitude exceeds the first threshold and the rising edge slope is greater than the second threshold are retained.

5. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The pulse distortion features extracted by the extraction module include pulse leading edge distortion, pulse trailing edge attenuation coefficient, and pulse top flatness. The pulse leading edge distortion ; In the formula: The rise time of the pulse leading edge is selected as the time from 10% to 90% of the peak value; This represents the instantaneous amplitude of the actual pulse signal; The instantaneous amplitude of the preset ideal pulse signal; This represents the peak value of the actual pulse signal. The spectral distribution characteristics include the characteristic frequency amplitude ratio, the spectral centroid frequency, and the spectral bandwidth. The amplitude change trend parameters include the variance of the amplitude peak sequence, the amplitude change period, and the amplitude decay index.

6. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The feature parameter association and matching operation in the recognition module follows the following: ； In the formula: The similarity value is associated with the feature parameters; This represents the total number of feature parameters involved in the matching. The weight coefficient for the i-th feature parameter; This represents the value of the i-th feature parameter extracted so far. This represents the i-th standard feature parameter value corresponding to the defect type in the typical defect discharge feature map library; when When the similarity is less than the preset threshold, the discharge mode comparison process is initiated. By comparing the discharge phase distribution and pulse sequence interval characteristics of the current signal with the standard patterns in the spectrum library, the type of partial discharge defect is determined.

7. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 6, characterized in that, The formula for calculating the degree of fit is: ； In the formula: Discharge mode compatibility; , These are the weighting coefficients; Let be the discharge phase distribution function of the current signal; This represents the discharge phase distribution function of the standard modes in the spectrum library; This is the interval difference correction factor; The average interval of the pulse sequence of the current signal; The average interval of the pulse sequence in the standard mode; when When the degree of partial discharge exceeds the preset fit threshold, the partial discharge defect is ultimately determined to be the type corresponding to the standard mode.

8. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, In the defect spatial coordinate calculation stage of the positioning module, the coordinates are calculated based on a three-dimensional rectangular coordinate system, with the geometric center of the converter core as the origin. ； In the formula: For the defect space coordinates; These represent the propagation speeds of partial discharge signals in the two different media of the converter transformer; This is the preset base time; These represent the times when the UHF sensors at three different locations received the partial discharge signal; The vertical distance from the plane containing the three sensors to the origin of the coordinate system; The horizontal angle between the line connecting the first sensor and the origin; These are the attenuation correction coefficients for the signal propagation paths corresponding to the three sensors.

9. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The standard calibration signal for load adaptation in the calibration module is generated based on a dynamic calibration model constructed from the real-time operating load parameters of the converter transformer. The real-time operating load parameters include load current, load power factor, and winding hot spot temperature. The amplitude and frequency of the standard calibration signal are output through the dynamic calibration model. ； In the formula: The amplitude of the standard calibration signal; The amplitude is calibrated as a reference. This is the load current; This is the rated current of the converter transformer; The load power factor; This refers to the hot spot temperature of the winding. This refers to the rated hot spot temperature of the converter transformer. The frequency of the standard calibration signal; The reference calibration frequency; This is the load adaptability factor.

10. The high-frequency partial discharge high-speed acquisition and defect location system for converter transformers according to claim 1, characterized in that, The acquisition module is interconnected with the conditioning module and the extraction module via a wireless network. The extraction module is interconnected with the identification module via a wireless network. The identification module is interconnected with the positioning module via a wireless network. The positioning module is interconnected with the calibration module via a wireless network.

Images