Alternating current voltage withstanding characteristic detection method of 110KV grounding transformer winding

Abstract

The invention provides an alternating current withstand voltage characteristic detection method of a 110KV grounding transformer winding, and belongs to the technical field of grounding transformer windings, and the method comprises the steps: building a scaling type grounding transformer winding physical model, carrying out the power frequency alternating current voltage test under different voltage levels, employing the orthogonal decomposition to analyze the voltage and current waveforms, building a multi-dimensional characteristic correlation matrix, and carrying out the detection of the alternating current withstand voltage characteristic of the 110KV grounding transformer winding. And calculating power frequency and harmonic impedance coefficients. According to the method, the alternating-current voltage withstanding capability of the actual grounding transformer winding is accurately evaluated on the basis of the proportional relation between the model and the actual 110KV grounding transformer winding, and an innovative research method is provided for diagnosis of the insulating performance of the transformer. According to the detection method provided by the invention, the defects in the prior art are overcome, comprehensive and deep analysis of the AC voltage withstanding characteristic of the grounding transformer winding is realized, and the technical problem that the AC voltage withstanding characteristic of the grounding transformer winding is difficult to comprehensively evaluate by an existing single detection means is solved.

Description

Technical Field

[0001] The invention belongs to the technical field of grounding transformer windings, and in particular relates to a method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding. Background Art

[0002] As an important high-voltage transmission equipment in the power system, the AC withstand voltage performance of the 110KV grounding transformer winding is directly related to the safe and stable operation of the power grid. Traditional grounding transformer windings mainly adopt oil-immersed winding structure. Working under high voltage, high current and complex electromagnetic environment, its insulation will experience severe electrical and thermal stress, and it is very easy to have insulation failure problems such as partial discharge and breakdown. In order to ensure the reliability and safety of the grounding transformer, it is necessary to conduct a comprehensive evaluation and reliability analysis of the AC withstand voltage characteristics of its windings.

[0003] At present, the commonly used detection methods for AC withstand voltage characteristics of grounding transformer windings in the industry mainly include: power frequency withstand voltage test, partial discharge detection and harmonic analysis. The power frequency withstand voltage test can directly obtain the breakdown voltage of the winding, but it cannot provide insight into factors that affect the insulation strength, such as partial discharge and temperature rise inside the winding. Partial discharge detection can reveal tiny defects in the winding insulator, but it cannot accurately predict its insulation margin for long-term operation. The harmonic analysis method can reflect the frequency dependence of the insulation characteristics of the winding, but it is difficult to establish a quantitative relationship between the insulation performance and the operating status. In general, the existing single detection method is difficult to comprehensively evaluate the AC withstand voltage characteristics of the grounding transformer winding, and it is difficult to provide a sufficient basis for insulation design and operation management. Summary of the invention

[0004] In view of this, the present invention provides a method for detecting the AC withstand voltage characteristics of a 110KV grounding transformer winding, which can solve the technical problem that the existing single detection means is difficult to comprehensively evaluate the AC withstand voltage characteristics of the grounding transformer winding.

[0005] The present invention is achieved in that:

[0006] The present invention provides a method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding, which comprises the following steps:

[0007] S10, constructing a scaled-down grounding transformer winding physical model, wherein the scaled-down grounding transformer winding physical model comprises a main transformer winding, a neutral point lead, a grounding lead, and a measurement system, wherein the measurement system comprises a high-precision voltage sensor, a high-precision current sensor, a partial discharge detection device, and an optical fiber temperature sensing system;

[0008] S20, applying an AC power voltage to the physical model of the scaled-down grounding transformer winding, wherein the amplitude of the AC power voltage increases step by step from 0.5 times the rated voltage to 2.5 times the rated voltage, the increment step of the AC power voltage is 0.25 times the rated voltage, and the duration of each voltage level is 10 minutes, and the effective value data of the AC power voltage, the waveform data of the AC power voltage, the effective value data of the AC power current, the waveform data of the AC power current, the partial discharge data and the temperature distribution data are collected in real time through the measurement system;

[0009] S30, performing orthogonal decomposition analysis on the power frequency AC voltage waveform data, decomposing the power frequency AC voltage waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC voltage waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental voltage components and harmonic voltage components according to the projection components;

[0010] S40, performing orthogonal decomposition analysis on the power frequency AC current waveform data, decomposing the power frequency AC current waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC current waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental current components and harmonic current components according to the projection components;

[0011] S50, establishing a multi-dimensional feature correlation matrix of the power frequency fundamental voltage component, the power frequency fundamental current component, the harmonic voltage component, the harmonic current component, the partial discharge data and the temperature distribution data;

[0012] S60, based on the multi-dimensional characteristic correlation matrix, calculating the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, and establishing a relationship matrix between the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix and the AC withstand voltage characteristics of the winding;

[0013] S70, based on the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, calculating the insulation margin coefficient and the breakdown risk coefficient at different voltage levels, wherein the insulation margin coefficient is calculated by the ratio of the breakdown voltage to the rated voltage, and the breakdown risk coefficient is calculated by Weibull distribution;

[0014] S80, establishing a voltage withstand characteristic evaluation index system for a physical model of a scaled-down grounding transformer winding according to the insulation margin coefficient and the breakdown risk coefficient, wherein the voltage withstand characteristic evaluation index system includes an insulation margin index, a breakdown risk index, a partial discharge intensity index, and a temperature distribution index;

[0015] S90. Determine the AC withstand voltage capability of the actual 110KV grounding transformer winding based on the withstand voltage characteristic evaluation index system and the proportional relationship between the scaled-down grounding transformer winding physical model and the actual 110KV grounding transformer winding.

[0016] The physical model of the scaled-down grounding transformer winding has a ratio of 1:10 or 1:15.

[0017] Furthermore, the main transformer winding is wound with enameled wire, and the number of winding turns is reduced proportionally according to the actual number of transformer turns, the number of turns is 200 to 300 turns, and the wire diameter is selected from 0.8 mm to 1.2 mm; the neutral point lead is made of multi-strand twisted wire with a cross-sectional area of ​​10 square millimeters to 16 square millimeters; the grounding lead is made of copper busbar with a width of 30 mm to 50 mm and a thickness of 2 mm to 3 mm.

[0018] Furthermore, it also includes the steps of comprehensively considering various indicators, determining the maximum allowable operating voltage and overvoltage tolerance capacity of the actual equipment, and giving safety margin recommendations.

[0019] In the above steps, the calculation process involved is described in detail as follows:

[0020] 1. The orthogonal decomposition analysis of the power frequency AC voltage waveform in step S30 is specifically expressed as follows:

[0021]

[0022] Where V(t) is the instantaneous value of the voltage waveform (V); V0 is the DC bias component (V); n is the harmonic order; N is the highest harmonic order, which is 10; ω is the fundamental angular frequency (rad / s); a n ,b n is the Fourier coefficient; ∈ v is the measurement error (V); t is the time (s).

[0023] Fourier coefficient calculation formula:

[0024]

[0025] Where T is the signal period (s).

[0026] Fundamental voltage component:

[0027] Harmonic voltage components:

[0028] 2. The orthogonal decomposition analysis of the power frequency AC current waveform in step S40 is specifically expressed as follows:

[0029]

[0030] Where I(t) is the instantaneous value of the current waveform (A); I0 is the DC bias component (A); c n ,d n is the Fourier coefficient; ∈i is the measurement error (A).

[0031] Fourier coefficient calculation formula:

[0032]

[0033] Fundamental current component:

[0034] Harmonic current components:

[0035] 3. The multidimensional feature correlation matrix in step S50 is specifically expressed as follows:

[0036]

[0037] In the formula, r ij is the Pearson correlation coefficient between the ith feature and the jth feature; m is the feature dimension.

[0038] Pearson correlation coefficient calculation formula:

[0039]

[0040] In the formula, x ik is the kth observation value of the ith feature; is the average value of the i-th feature; n is the number of observed samples; ∈ r is the calculation error.

[0041] 4. The impedance coefficient matrix in step S60 is specifically expressed as follows: Power frequency impedance coefficient matrix:

[0042]

[0043] In the formula, Z fij It is the complex impedance ratio of the i-th component of the power frequency fundamental voltage to the j-th component of the current (Ω).

[0044] Harmonic impedance coefficient matrix:

[0045]

[0046] In the formula, Z hij is the complex impedance ratio of the i-th harmonic voltage component to the j-th current component (Ω).

[0047] Complex impedance calculation formula:

[0048]

[0049] In the formula, |V i | is the amplitude of the i-th voltage component; |I j| is the amplitude of the jth component of the current; θ vi is the phase angle of the i-th voltage component; θ ij is the phase angle of the jth component of the current; ∈ z is the measurement error.

[0050] 5. The calculation of the insulation margin coefficient and the breakdown risk coefficient in step S70 is specifically expressed as follows:

[0051] Insulation margin factor:

[0052]

[0053] In the formula, K i is the insulation margin factor; V b is the breakdown voltage (V); V r is the rated voltage (V); α is the temperature correction factor; β is the humidity correction factor; γ is the pressure correction factor (Pa -1 );ΔT is the temperature deviation (K); ΔH is the humidity deviation (%); ΔP is the pressure deviation (Pa);∈ k is the calculation error.

[0054] Breakdown risk factor:

[0055] P f =1-exp[-(V / η) β ]+∈ p ;

[0056] In the formula, p f is the breakdown probability; V is the operating voltage (V); η is the scale parameter; β is the shape parameter; ∈ p is the calculation error.

[0057] 6. The withstand voltage characteristic evaluation index system in step S80 is specifically expressed as follows: Comprehensive evaluation score:

[0058] S=w1K i +w2(1-P f )+w3exp(-Q / W0)+w4exp(-ΔT max / ΔT0)+∈ s ;

[0059] Where S is the comprehensive evaluation score; w1, w2, w3, w4 are weight coefficients, which are 0.3, 0.3, 0.2, 0.2 respectively; Q is the maximum partial discharge (pC); Q0 is the partial discharge reference value, which is 50pC; ΔT max is the maximum temperature rise (K); ΔT0 is the temperature rise reference value, which is 65K; ∈ s To evaluate the error.

[0060] The construction principle and significance of the above equation are explained in detail:

[0061] 1. Orthogonal decomposition analysis equation of voltage waveform: The reasons for using Fourier series expansion are: (1) The power frequency voltage waveform has periodic characteristics, and the Fourier series can accurately describe the periodic signal; (2) The orthogonal basis functions are independent of each other, which is conducive to separating the fundamental and harmonic components; (3) The DC bias term V0 is introduced to eliminate the zero drift that may be introduced by the sampling system; (4) The error term ∈ v Used to characterize the random error and quantization error of the measurement system; (5) The 10th harmonic is chosen as the cutoff because the amplitude of higher-order harmonics is usually smaller and can be ignored.

[0062] The Fourier coefficient calculation formula is based on the orthogonality principle: (1) the use of definite integral form can effectively suppress the influence of random noise; (2) the coefficient It is used for normalization processing so that the decomposition results have clear physical meanings; (3) The integration interval is selected as a complete period to avoid spectrum leakage.

[0063] The expressions of fundamental and harmonic components are in the form of trigonometric functions: (1) It is convenient to analyze the phase relationship and power characteristics; (2) The amplitude is in the form of square sum and square root to ensure the stability of the calculation results; (3) The phase uses the inverse tangent function, which can handle the phase angle of any quadrant.

[0064] 2. Orthogonal decomposition analysis equation of current waveform: The same mathematical form is used as that of voltage waveform. The main considerations are: (1) to maintain the consistency of voltage and current analysis methods, which is convenient for subsequent impedance calculation; (2) the current waveform may contain more harmonic components, and the measurement error ∈ needs to be considered i ; (3) Fourier coefficient c n ,d n The coefficient is expressed separately from the voltage coefficient to avoid confusion.

[0065] 3. Construction of multi-dimensional feature correlation matrix: The reasons for using Pearson correlation coefficient are: (1) It can describe the degree of linear correlation, and the value range is normalized between [-1, 1]; (2) It is insensitive to data scale and suitable for correlation analysis between parameters of different dimensions; (3) The matrix form is convenient for expressing the relationship between multiple features; (4) The introduction of calculation error ∈ r Rounding errors of numerical calculations are taken into account.

[0066] 4. Impedance coefficient matrix expression: The reasons for using complex form are: (1) it can express the amplitude ratio and phase difference information at the same time; (2) the matrix form can describe the coupling relationship between different frequency components; (3) the measurement error ∈ zIt includes the comprehensive errors of amplitude and phase; (4) The impedance variation characteristics with frequency can reflect the frequency response of the insulation system.

[0067] 5. Insulation margin coefficient calculation formula: The reasons for using the multi-factor correction form are: (1) The linear correction term reflects the impact of environmental factors on insulation strength; (2) The correction coefficients α, β, and γ can be calibrated through experiments; (3) The error term ∈ k The cumulative effect of experimental error and calculation error is taken into account; (4) The numerator and denominator form facilitates relative comparison.

[0068] The reasons for using Weibull distribution for the breakdown risk coefficient are: (1) it is consistent with the physical mechanism and statistical characteristics of insulation breakdown; (2) the shape parameter β reflects the dominant factor of the breakdown mechanism; (3) the scale parameter η characterizes the insulation strength level; (4) the error term ∈ p The uncertainty of the statistical model is taken into account.

[0069] 6. Calculation formula for comprehensive evaluation score: The reasons for using weighted summation are: (1) Each evaluation index is normalized to the interval [0,1] through an exponential function; (2) The weight coefficient reflects the relative importance of each index; (3) The partial discharge and temperature rise are in the form of exponential decay, reflecting the penalty effect of exceeding the standard; (4) The error term ∈ s Characterizes the uncertainty of the assessment process.

[0070] The above equation system constructs a complete mathematical model from multiple perspectives such as waveform analysis, feature extraction, impedance calculation, and reliability assessment, which can quantitatively describe and evaluate the AC withstand voltage characteristics of the winding. Compared with the existing technology, this method has the following advantages: (1) It takes into account the influencing factors of multiple aspects such as electrical, thermal, and reliability; (2) It introduces error terms to improve the adaptability and robustness of the model; (3) It adopts standardization and normalization processing to facilitate horizontal comparison between different devices; (4) It establishes a quantitative evaluation system to provide a basis for equipment operation and maintenance decisions.

[0071] Compared with the prior art, the present invention provides a method for detecting the AC withstand voltage characteristics of a 110KV grounding transformer winding, which includes first building a set of scaled-down grounding transformer winding physical models, and using high-precision sensors to comprehensively measure its parameters such as voltage, current, partial discharge and temperature at different voltage levels. Then, an orthogonal decomposition algorithm is used to conduct an in-depth analysis of the measurement data, including the extraction of power frequency fundamental and harmonic characteristics, the calculation of impedance characteristics, and the establishment of insulation state indicators. Finally, based on the impedance characteristics and insulation state indicators, combined with similarity theory, the AC withstand voltage capability of the actual 110KV grounding transformer winding is comprehensively evaluated, and safety margins and operation recommendations are given.

[0072] Compared with the traditional single detection method, the detection method proposed by the present invention has the following advantages:

[0073] 1. A scaled-down physical model was established, which can fully simulate the working environment and stress factors of the actual grounding transformer under experimental conditions, providing a reliable basis for insulation performance evaluation.

[0074] 2. The use of multi-parameter measurement and orthogonal decomposition analysis technology can deeply reveal the electrical and thermal characteristics inside the winding, providing a basis for establishing a quantitative relationship between insulation performance and operating status.

[0075] 3. Taking into account multiple indicators such as insulation margin, breakdown risk, partial discharge and temperature distribution, a comprehensive and reliable voltage withstand characteristic evaluation index system is constructed, which provides an effective basis for winding insulation design and life prediction.

[0076] 4. The similarity theory is used to scale the results of the scaled model, which can accurately predict the AC withstand voltage capability of the actual 110KV grounding transformer.

[0077] In summary, the detection method proposed in the present invention makes up for the shortcomings of the existing technology, realizes a comprehensive and in-depth analysis of the AC withstand voltage characteristics of the grounding transformer winding, and solves the technical problem that the existing single detection method is difficult to comprehensively evaluate the AC withstand voltage characteristics of the grounding transformer winding. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A flow chart of the method provided by the present invention. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0080] like Figure 1 As shown in FIG. 1 , it is a flow chart of a method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding provided by the present invention. The method comprises the following steps:

[0081] S10. Build a scaled-down grounding transformer winding physical model, which includes a main transformer winding, a neutral point lead, a grounding lead, and a measurement system. The measurement system includes a high-precision voltage sensor, a high-precision current sensor, a partial discharge detection device, and an optical fiber temperature sensing system.

[0082] S20, applying an AC power voltage to the physical model of the scaled-down grounding transformer winding, wherein the amplitude of the AC power voltage increases step by step from 0.5 times the rated voltage to 2.5 times the rated voltage, the increment step of the AC power voltage is 0.25 times the rated voltage, and the duration of each voltage level is 10 minutes, and real-time AC power voltage effective value data, AC power voltage waveform data, AC power current effective value data, AC power current waveform data, partial discharge data, and temperature distribution data are collected through a measurement system;

[0083] S30, performing orthogonal decomposition analysis on the power frequency AC voltage waveform data, decomposing the power frequency AC voltage waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC voltage waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental voltage components and harmonic voltage components according to the projection components;

[0084] S40, performing orthogonal decomposition analysis on the power frequency AC current waveform data, decomposing the power frequency AC current waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC current waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental current components and harmonic current components according to the projection components;

[0085] S50, establishing a multi-dimensional feature correlation matrix of power frequency fundamental wave voltage component, power frequency fundamental wave current component, harmonic voltage component, harmonic current component, partial discharge data and temperature distribution data;

[0086] S60, based on the multi-dimensional characteristic correlation matrix, calculating the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, and establishing a relationship matrix between the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix and the AC withstand voltage characteristics of the winding;

[0087] S70. Based on the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, the insulation margin coefficient and the breakdown risk coefficient at different voltage levels are calculated, the insulation margin coefficient is calculated by the ratio of the breakdown voltage to the rated voltage, and the breakdown risk coefficient is calculated by Weibull distribution;

[0088] S80, establishing a voltage withstand characteristic evaluation index system for a physical model of a scaled-down grounding transformer winding according to an insulation margin coefficient and a breakdown risk coefficient, wherein the voltage withstand characteristic evaluation index system includes an insulation margin index, a breakdown risk index, a partial discharge intensity index, and a temperature distribution index;

[0089] S90. Based on the voltage withstand characteristic evaluation index system and the proportional relationship between the physical model of the scaled-down grounding transformer winding and the actual 110KV grounding transformer winding, determine the AC voltage withstand capability of the actual 110KV grounding transformer winding.

[0090] The specific implementation methods of the above steps are described in detail below:

[0091] The specific implementation method of step S10 is to build a reliable experimental platform. First, the geometric size ratio of the scaled model and the actual 110KV grounding transformer winding is determined based on the electromagnetic field similarity theory. Usually, a ratio of 1:10 or 1:15 is selected to ensure both the feasibility of the experiment and the accuracy of the measurement results. The main transformer winding of the scaled grounding transformer winding physical model is wound with enameled wire. The number of winding turns is proportionally reduced according to the number of turns of the actual transformer, generally 200 to 300 turns, and the wire diameter is selected from 0.8mm to 1.2mm. The neutral point lead wire is a multi-strand twisted wire with a cross-sectional area of ​​10mm 2 Up to 16mm 2 , the grounding lead adopts copper bar with a width of 30mm to 50mm and a thickness of 2mm to 3mm. The high-precision voltage sensor in the measurement system adopts a capacitive voltage divider with a rated voltage of 20KV, an accuracy level of 0.2, and a sampling frequency of not less than 100KHz. The high-precision current sensor adopts a Rogowski coil with a range of 0 to 5A, an accuracy level of 0.2, and a sampling frequency of not less than 100KHz. The detection sensitivity of the partial discharge detection device is better than 5pC, and the bandwidth is 40KHz to 400KHz. The fiber optic temperature sensing system adopts grating sensing technology, with a temperature measurement range of 0℃ to 200℃, a temperature measurement accuracy better than 0.1℃, and a spatial resolution better than 1m. During the construction process, special attention should be paid to the electromagnetic compatibility of each measuring equipment, and shielded cables should be used to connect each measuring equipment to ensure good grounding.

[0092] The specific implementation method of step S20 is to perform a standardized withstand voltage test on the physical model of the scaled-down grounding transformer winding. First, a power frequency test transformer is used as a voltage source, and its rated capacity is not less than 5KVA, and the output voltage can be adjusted in the range of 0 to 30KV. Before the test, it is necessary to measure the environmental parameters of the test site to ensure that the ambient temperature is between 15°C and 35°C, the relative humidity is not higher than 85%, and the atmospheric pressure is between 90% and 110% of the standard atmospheric pressure. The voltage application process adopts a step-by-step increase method, starting from 0.5 times the rated voltage, that is, if the rated voltage is 10KV, the starting voltage is 5KV. Each increase of 0.25 times the rated voltage is an increase of 2.5KV, until it reaches 2.5 times the rated voltage, that is, 25KV. The duration of each voltage level is 10 minutes, during which various parameters are recorded in real time by the data acquisition system. The sampling frequency of the data acquisition system is set to 100KHz to ensure that the high-frequency component can be accurately captured. The collected data includes voltage effective value, voltage waveform, current effective value, current waveform, partial discharge and temperature distribution. To ensure the reliability of the data, each measurement point was collected at least 3 times, and the average value was taken as the final measurement result.

[0093] The specific implementation method of step S30 is to conduct an in-depth analysis of the collected power frequency AC voltage waveform data, mainly using an orthogonal decomposition algorithm for processing. First, the sampled voltage waveform data is preprocessed, including removing DC bias, eliminating power frequency interference and filtering high-frequency noise. The preprocessed data is decomposed using a Fourier orthogonal basis function system, and the basis functions include sine functions and cosine functions. By calculating the inner product of the voltage waveform data and each orthogonal basis function, the projection coefficient of the voltage waveform on each frequency component is obtained. Generally, the first 10 harmonics are selected for analysis, that is, the fundamental frequency 50Hz and its harmonic components. For the fundamental component, its amplitude and phase information are extracted, the amplitude is the effective value of the fundamental voltage, and the phase adopts the relative phase based on the zero crossing point of the power supply voltage. For the harmonic component, the amplitude and phase of each harmonic are calculated respectively, and the total harmonic distortion rate is calculated, and the total harmonic distortion rate should generally not exceed 5%. In addition, it is also necessary to analyze the time domain distribution characteristics of each harmonic component and calculate the peak factor and waveform factor of each harmonic.

[0094] The specific implementation method of step S40 is to perform orthogonal decomposition analysis on the power frequency AC current waveform data, using the same analysis method as the voltage waveform. First, the current waveform data is preprocessed, including eliminating sampling errors, filtering out power frequency interference and suppressing high-frequency noise. Then the preprocessed data is projected into the orthogonal basis function space, and the projection coefficient of each frequency component is calculated. The fundamental current component and the first 9 harmonic components, i.e., 50Hz, 100Hz, 150Hz to 500Hz, are analyzed in detail. For the fundamental current component, its effective value and phase angle are extracted, and compared with the fundamental voltage to calculate the power factor. For the harmonic current component, the amplitude ratio of each harmonic is calculated, i.e., the ratio of each harmonic current to the fundamental current. It is usually required that the amplitude ratio of each harmonic current does not exceed 10% of the fundamental current. At the same time, the phase sequence characteristics of the harmonic current are analyzed, and the distribution of the positive sequence component, the negative sequence component and the zero sequence component are calculated. In addition, the distortion factor of the current waveform needs to be calculated, and the distortion factor is generally required to be no more than 8%.

[0095] The specific implementation method of step S50 is to construct a multidimensional feature correlation matrix using data mining and statistical analysis methods. First, all measurement data are aligned in time series, and the sliding time window method is used for data synchronization. The time window length is 20ms and the sliding step length is 1ms. Then the Pearson correlation coefficient between the parameters is calculated to construct the initial correlation matrix. In order to improve the reliability of the correlation analysis, the partial correlation analysis method is used to eliminate the influence of other variables. On this basis, the principal component analysis method is used to extract the main eigenvectors, and generally the eigenvectors with a cumulative contribution rate of 85% are selected. For partial discharge data, it is necessary to analyze its amplitude distribution and phase distribution characteristics, and calculate the repetition rate and average discharge amount of partial discharge. For temperature distribution data, it is necessary to analyze its spatial distribution characteristics, calculate the temperature gradient and hot spot distribution. Finally, all characteristic parameters are standardized to construct a standardized feature matrix.

[0096] The specific implementation method of step S60 is to calculate the impedance coefficient matrix based on the multidimensional feature correlation matrix, and adopt the complex impedance analysis method. First, the complex ratio of the power frequency fundamental voltage and current is calculated to obtain the power frequency fundamental impedance. Then the complex ratio of each harmonic voltage and current is calculated respectively to obtain the harmonic impedance. For the power frequency impedance, it is necessary to analyze its characteristics that change with voltage and establish a functional relationship between the power frequency impedance and the voltage. For the harmonic impedance, it is necessary to analyze its frequency characteristics and establish a functional relationship between the harmonic impedance and the frequency. On this basis, the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix are constructed, and the matrix dimension is determined by the harmonic order. In general, the amplitude of the power frequency impedance should be in the range of 90% to 110% of the nominal value, and the phase angle should be in the range of -5° to +5°. The amplitude of the harmonic impedance increases with increasing frequency, but the growth rate should gradually decrease.

[0097] The specific implementation method of step S70 is to calculate the insulation margin coefficient and the breakdown risk coefficient, and adopt the reliability analysis method. First, based on the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, the insulation strength under different voltage levels is calculated. Then, the relationship curve between insulation strength and voltage is fitted by experimental data, and the least square method is used for curve fitting. The insulation margin coefficient is calculated by the ratio of breakdown voltage to rated voltage, and the ratio is generally required to be not less than 2.5. For the calculation of the breakdown risk coefficient, a two-parameter Weibull distribution model is adopted, in which the shape parameter reflects the breakdown mechanism and the scale parameter reflects the insulation strength level. The parameters of the Weibull distribution are determined by the maximum likelihood estimation method, and then the breakdown probability under rated voltage is calculated, requiring the breakdown probability not to exceed 0.001%. At the same time, it is necessary to consider the influence of environmental factors such as temperature and humidity, and establish an environmental factor correction coefficient.

[0098] The specific implementation method of step S80 is to construct a withstand voltage characteristic evaluation index system, and use the hierarchical analysis method to assign index weights. First, determine the hierarchical structure of the evaluation index system, including the target layer, the criterion layer and the index layer. The target layer is a comprehensive evaluation of the withstand voltage characteristics, the criterion layer includes three aspects: electrical performance, thermal stability and reliability, and the index layer includes specific evaluation indicators. The weight of the insulation margin index is 0.3, requiring the insulation margin coefficient to be greater than 2.5. The weight of the breakdown risk index is 0.3, requiring the breakdown probability to be less than 0.001%. The weight of the partial discharge intensity index is 0.2, requiring the maximum discharge amount not to exceed 50pC. The weight of the temperature distribution index is 0.2, requiring the maximum temperature rise not to exceed 65K. The fuzzy comprehensive evaluation method is used to determine the membership function of each index and establish a judgment matrix. The reliability of the judgment result is ensured by consistency test, and the consistency ratio should be less than 0.1.

[0099] The specific implementation method of step S90 is to determine the AC withstand voltage capability of the actual 110KV grounding transformer winding, and use similarity theory for scale conversion. First, according to the geometric size ratio of the scaled model, the conversion relationship of parameters such as electric field strength and current density is established. Then, based on the test results of the evaluation index system, the various indicators of the actual equipment are calculated in combination with the conversion relationship. For the insulation margin, it is necessary to consider the influence of the size effect, and the insulation margin coefficient of the actual equipment should not be less than 90% of the scaled model. For the breakdown risk, it is necessary to consider the influence of the volume effect, and correct the breakdown probability through a statistical model. The local discharge characteristics need to consider the similarity of the electric field distribution, and the discharge starting voltage of the actual equipment should conform to the proportional relationship.

[0100] Optionally, the temperature distribution characteristics need to consider the differences in heat conduction and convection conditions and establish a conversion model for temperature rise.

[0101] Optionally, it also includes comprehensive consideration of various indicators, determining the maximum allowable operating voltage and overvoltage withstand capacity of the actual equipment, and giving safety margin recommendations. In general, it is recommended that the operating voltage should not exceed 90% of the rated voltage, and the overvoltage withstand capacity should not be less than 2.2 times the rated voltage.

[0102] Specifically, the principle of the present invention is:

[0103] 1. Construction of scaled physical model. According to the electromagnetic field similarity theory, the geometric size ratio between the scaled model and the actual 110KV grounding transformer is first determined, and a scale ratio of 1:10 or 1:15 is usually selected. This can ensure the feasibility of the experiment and the accuracy of the measurement results. The scaled model mainly includes core components such as the main transformer winding, neutral point lead and grounding lead. It is made of materials such as enameled wire, multi-strand stranded wire and copper busbar. The number of winding turns and the cross-sectional area of ​​the conductor are reduced according to the geometric ratio. At the same time, it is equipped with high-precision voltage, current, partial discharge and temperature measurement systems to meet the requirements of comprehensive monitoring.

[0104] 2. Multi-parameter measurement and orthogonal decomposition analysis. By applying power frequency AC voltage to the scaled model, high-precision sensors are used to measure parameters such as voltage, current, partial discharge and temperature. In order to fully reveal the electrical and thermal characteristics inside the winding, the measured data is subjected to orthogonal decomposition analysis technology. The voltage and current waveform data are respectively expanded by Fourier series to extract the fundamental and harmonic components. The fundamental component reflects the power frequency characteristics of the winding, and the harmonic component reveals the frequency dependence of the winding insulation. At the same time, the impedance characteristics of the fundamental and harmonic waves are calculated, which provides an important basis for the insulation status assessment. In addition, the partial discharge and temperature distribution characteristics need to be analyzed to reveal the influence of insulation defects and thermal stress.

[0105] 3. Comprehensive evaluation of withstand voltage characteristics. Based on the power frequency and harmonic impedance characteristics calculated from the measured data, combined with the partial discharge and temperature distribution indicators, a complete set of withstand voltage characteristic evaluation index system is established. This index system includes: insulation margin index, breakdown risk index, partial discharge intensity index and temperature distribution index. Among them, the insulation margin index reflects the long-term withstand voltage capability of the winding insulator, the breakdown risk index reveals the safety margin of transient overvoltage, the partial discharge intensity index characterizes the severity of insulation defects, and the temperature distribution index reflects the thermal stability of the winding. Through the comprehensive analysis of these indicators, the AC withstand voltage characteristics of the scaled model can be comprehensively evaluated.

[0106] 4. Scale conversion and actual equipment evaluation. Based on the test results of the scaled model, the similarity theory is used to scale the various indicators to obtain the AC withstand voltage capability of the actual 110KV grounding transformer. This conversion process needs to consider the geometric scale effects of factors such as electric field strength, current density, and heat conduction. Finally, the maximum allowable operating voltage and overvoltage withstand capacity of the actual equipment are comprehensively evaluated to provide a scientific basis for on-site operation management.

[0107] The following is an example of a specific application scenario of the present invention: A power company recently conducted a routine inspection of its 110kV grounding transformer and found that the winding insulator had a serious partial discharge problem and uneven temperature distribution. In order to comprehensively evaluate the AC withstand voltage characteristics of the grounding transformer and effectively prevent the occurrence of insulation accidents, the company decided to use the detection method proposed by the present invention for in-depth analysis.

[0108] First, the company designed and manufactured a 1:10 scaled-down grounding transformer winding physical model based on the electromagnetic field similarity theory. The main transformer winding is wound with 0.8mm enameled wire, with 220 turns; the neutral point lead wire is 14mm 2 The grounding lead is a 40mm wide and 2mm thick copper bar. In the measurement system, the voltage sensor uses a 20kV / 100V capacitive voltage divider, the current sensor uses a 0-5A range Rogowski coil, the sensitivity of the partial discharge detection device is better than 5pC, and the spatial resolution of the optical fiber temperature sensing system is 0.5m. To ensure the electromagnetic compatibility of each measuring device, shielded cables are used for connection and good grounding measures are taken. The ambient temperature of the test site is controlled at around 25℃, and the relative humidity does not exceed 80%.

[0109] When conducting the power frequency withstand voltage test, first use a 5kVA power frequency test transformer to apply an AC voltage that increases step by step from 5kV (0.5 times the rated voltage) to the scaled model, increasing by 2.5kV (0.25 times the rated voltage) each time, and applying it to 25kV (2.5 times the rated voltage). Each voltage level lasts for 10 minutes, and the data acquisition system records the voltage RMS value, voltage waveform, current RMS value, current waveform, partial discharge, temperature distribution and other parameters in real time at a sampling frequency of 100kHz. In order to ensure the reliability of the data, each measuring point is measured three times, and the average value is taken as the final result.

[0110] By analyzing the collected data, the following results can be obtained:

[0111] 1. Voltage and current waveform analysis The voltage waveform data is analyzed by Fourier orthogonal decomposition to extract the fundamental and harmonic components. The effective value of the fundamental voltage increases linearly with the increase of the applied voltage, which is 5.05kV at 5kV and 25.02kV at 25kV, which is basically consistent with the theoretical value. The total harmonic distortion rate of the first five harmonics (100Hz, 150Hz, 200Hz, 250Hz, 300Hz) has been controlled below 4.7%, not exceeding the limit of 5%. The current waveform data also uses the same orthogonal decomposition analysis method. The effective value of the fundamental current increases with the increase of voltage, but the amplitude change is not strictly proportional, reflecting the nonlinear characteristics of the winding impedance with voltage. The proportion of each harmonic current to the fundamental current does not exceed 8%, which meets the requirements. Through calculation, the power factor varies between 0.92 and 0.97, which is in line with the normal range.

[0112] 2. Impedance characteristic analysis According to the fundamental and harmonic components of voltage and current, the power frequency impedance and harmonic impedance are calculated respectively. The power frequency impedance increases slowly with the increase of voltage, which is 14.2Ω at 5kV and 15.4Ω at 25kV. The amplitude fluctuation is within ±10% of the nominal value, and the phase angle is between -4° and +3°, which meets the requirements. The impedance of each harmonic increases with the increase of frequency, which is about 16Ω at 100Hz and about 21Ω at 300Hz, and the growth rate gradually decreases.

[0113] 3. Insulation status index The partial discharge test results show that the partial discharge is about 12pC at 5kV, which increases slowly with the increase of voltage, reaching 42pC at 25kV, which does not exceed the limit of 50pC. The temperature sensing system monitors that the maximum temperature rise of the main transformer winding of the scaled model is 48K, which is evenly distributed and has no obvious hot spots. According to the above test data and the relevant empirical formula, the insulation margin coefficient of the scaled model is calculated to be 3.1, which is within the safety range of more than 2.5; the breakdown probability is 0.08%, which is less than the requirement of 0.1%. The comprehensive evaluation index score is 88 points, which is a good level.

[0114] Finally, according to the geometric size ratio of the scaled model, the similarity theory was used to scale the above test results and predict the AC withstand voltage characteristics of the actual 110kV grounding transformer:

[0115] 1) The insulation margin factor is ≥2.8 and meets the requirement of ≥2.5.

[0116] 2) The breakdown probability at rated voltage is ≤ 0.12%, which is lower than the limit of 0.1%.

[0117] 3) The maximum partial discharge is ≤55pC and does not exceed the standard of 50pC.

[0118] 4) The maximum temperature rise is ≤60K, which is lower than the allowable value of 65K.

[0119] The comprehensive evaluation results show that the AC withstand voltage of the 110kV grounding transformer basically meets the requirements, but the partial discharge and temperature rise indicators are slightly exceeded, and the winding insulation structure and heat dissipation design need to be further optimized. It is recommended that the operating voltage of the transformer should not exceed 100kV (0.9 times the rated voltage), and the overvoltage withstand capacity should not be less than 242kV (2.2 times the rated voltage). At the same time, regular inspections and online monitoring should be strengthened to promptly detect and eliminate partial discharge problems and ensure long-term safe and reliable operation.

[0120] It should be noted that the variables mentioned in the present invention are explained as shown in the following Table 1:

[0121] Table 1 Variable explanation table

[0122]

[0123]

[0124] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding, characterized in that: The following steps are involved: S10, constructing a scaled-down grounding transformer winding physical model, wherein the scaled-down grounding transformer winding physical model comprises a main transformer winding, a neutral point lead, a grounding lead, and a measurement system, wherein the measurement system comprises a high-precision voltage sensor, a high-precision current sensor, a partial discharge detection device, and an optical fiber temperature sensing system; S20, applying an AC power voltage to the physical model of the scaled-down grounding transformer winding, wherein the amplitude of the AC power voltage increases step by step from 0.5 times the rated voltage to 2.5 times the rated voltage, the increment step of the AC power voltage is 0.25 times the rated voltage, and the duration of each voltage level is 10 minutes, and the effective value data of the AC power voltage, the waveform data of the AC power voltage, the effective value data of the AC power current, the waveform data of the AC power current, the partial discharge data and the temperature distribution data are collected in real time through the measurement system; S30, performing orthogonal decomposition analysis on the power frequency AC voltage waveform data, decomposing the power frequency AC voltage waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC voltage waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental voltage components and harmonic voltage components according to the projection components; S40, performing orthogonal decomposition analysis on the power frequency AC current waveform data, decomposing the power frequency AC current waveform data into power frequency orthogonal reference vectors, obtaining projection components of the power frequency AC current waveform data on the power frequency orthogonal reference vectors, and extracting power frequency fundamental current components and harmonic current components according to the projection components; S50, establishing a multi-dimensional feature correlation matrix of the power frequency fundamental voltage component, the power frequency fundamental current component, the harmonic voltage component, the harmonic current component, the partial discharge data and the temperature distribution data; S60, based on the multi-dimensional characteristic correlation matrix, calculating the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, and establishing a relationship matrix between the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix and the AC withstand voltage characteristics of the winding; S70, based on the power frequency impedance coefficient matrix and the harmonic impedance coefficient matrix, calculating the insulation margin coefficient and the breakdown risk coefficient at different voltage levels, wherein the insulation margin coefficient is calculated by the ratio of the breakdown voltage to the rated voltage, and the breakdown risk coefficient is calculated by Weibull distribution; S80, establishing a voltage withstand characteristic evaluation index system for a physical model of a scaled-down grounding transformer winding according to the insulation margin coefficient and the breakdown risk coefficient, wherein the voltage withstand characteristic evaluation index system includes an insulation margin index, a breakdown risk index, a partial discharge intensity index, and a temperature distribution index; S90. Determine the AC withstand voltage capability of the actual 110KV grounding transformer winding based on the withstand voltage characteristic evaluation index system and the proportional relationship between the scaled-down grounding transformer winding physical model and the actual 110KV grounding transformer winding.

2. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 1 is characterized in that: The orthogonal decomposition analysis of the power frequency AC voltage waveform in step S30 is specifically expressed as follows: Where V(t) is the instantaneous value of the voltage waveform; V0 is the DC bias component; n is the harmonic order; N is the highest harmonic order; ω is the fundamental angular frequency; a n ,b n is the Fourier coefficient; ∈ v is the voltage measurement error; t is the time.

3. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 2 is characterized in that: The orthogonal decomposition analysis of the power frequency AC current waveform in step S40 is specifically expressed as follows: Where I(t) is the instantaneous value of the current waveform; I0 is the DC bias component; c n ,d n is the Fourier coefficient; ∈ i is the current measurement error.

4. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 3 is characterized in that: The multidimensional feature correlation matrix in step S50 is specifically expressed as follows: In the formula, r ij is the Pearson correlation coefficient between the ith feature and the jth feature; m is the feature dimension.

5. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 4 is characterized in that: The impedance coefficient matrix in step S60 is specifically expressed as follows: Power frequency impedance coefficient matrix: In the formula, Z fij is the complex impedance ratio of the i-th component of the power frequency fundamental voltage to the j-th component of the current; Harmonic impedance coefficient matrix: In the formula, Z hij It is the complex impedance ratio of the i-th harmonic voltage component to the j-th current component.

6. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 5, characterized in that: The calculation of the insulation margin coefficient and the breakdown risk coefficient in step S70 is specifically expressed as follows: Insulation margin factor In the formula, K i is the insulation margin factor; V b is the breakdown voltage; V r is the rated voltage; α is the temperature correction coefficient; β is the humidity correction coefficient; γ is the pressure correction coefficient; ΔT is the temperature deviation; ΔH is the humidity deviation; ΔP is the pressure deviation; ∈ k Calculation error for insulation margin factor; Breakdown risk factor: p f =1-exp[-(V / η) β ]+∈ p ; Where, P f is the breakdown probability; V is the operating voltage; η is the scale parameter; β is the shape parameter; ∈ p Calculate the error for the breakdown risk factor.

7. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 6, characterized in that: The withstand voltage characteristic evaluation index system in step S80 is specifically expressed as follows: S=w1K i +w2(1-P f )+w3exp(-Q / Q0)+w4exp(-ΔT max / ΔT0)+∈ s ; Where S is the comprehensive evaluation score; w1, w2, w3, w4 are weight coefficients; Q is the maximum partial discharge; Q0 is the partial discharge reference value; ΔT max is the maximum temperature rise; ΔT0 is the temperature rise reference value; ∈ s To evaluate the error.

8. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 1, characterized in that: The physical model of the scaled-down grounding transformer winding has a ratio of 1:10 or 1:

15.

9. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 8, characterized in that: The main transformer winding is made of enameled wire, and the number of winding turns is reduced proportionally according to the actual number of transformer turns, the number of turns is 200 to 300 turns, and the wire diameter is selected from 0.8 mm to 1.2 mm; the neutral point lead is made of multi-strand twisted wire with a cross-sectional area of ​​10 square millimeters to 16 square millimeters; the grounding lead is made of copper busbar with a width of 30 mm to 50 mm and a thickness of 2 mm to 3 mm.

10. The method for detecting AC withstand voltage characteristics of a 110KV grounding transformer winding according to claim 9, characterized in that: It also includes the final step of comprehensively considering various indicators, determining the maximum allowable operating voltage and overvoltage withstand capacity of the actual equipment, and giving safety margin recommendations.

Images