Electrical automation quality detection method and system

Abstract

The application belongs to the technical field of quality detection, and discloses an electrical automation quality detection method and system; the method comprises the following steps: collecting original voltage and current signals of a PCC point of an electric network and performing FFT to obtain a background harmonic frequency set and a background harmonic amplitude set; generating an initial frequency grid based on a preset target frequency band, and combining the background harmonics to generate a sparse frequency grid; calculating the estimated signal-to-noise ratio of each candidate frequency and performing ascending order sorting to obtain an injection sequence, injecting a sinusoidal signal and collecting the voltage response and total current of the PCC point, and calculating the frequency impedance data pair; performing rational function fitting based on the frequency impedance data pair to obtain a wideband impedance curve; analyzing the curve to extract quality characteristic parameters, and performing quality determination based on the quality characteristic parameters; and the application effectively solves the technical problem that the prior art cannot quickly and automatically determine the grid-connected stability quality of a single new energy converter in an offline detection scene of a production line.

Description

Technical Field

[0001] This invention relates to the field of quality inspection technology, and more specifically, to an electrical automation quality inspection method and system. Background Technology

[0002] With the large-scale grid connection of new energy sources, the oscillation problems such as subsynchronous and supersynchronous oscillations caused by their interaction with the grid are becoming increasingly prominent. Wideband impedance characteristics, as a core key indicator for analyzing the grid-connected stability of new energy power plants and converters, are crucial for assessing the grid-connected system's disturbance rejection capability and mitigating oscillation risks. Currently, the traditional methods used in the industry to measure wideband impedance mainly include the frequency sweep method and the random signal injection method. The frequency sweep method requires sequentially injecting disturbance signals at different frequency points and measuring the response, which is not only time-consuming but also potentially excites system oscillations due to the long duration and large amplitude of the disturbance signals, posing a threat to the safe operation of the power grid. While the random signal injection method can achieve... While the frequency response acquisition is relatively fast, the dispersed energy of the injected signal leads to a low signal-to-noise ratio. Especially in the field environment of the common connection point (PCC) of new energy power plants, where background harmonics are abundant, the measurement accuracy is difficult to guarantee effectively. The core contradiction of the existing technology is that it cannot balance the speed of impedance measurement, the safety of disturbance injection, and the accuracy of measurement results. At the same time, in the scenario of testing the grid connection stability of individual new energy converters after they are off the production line, the traditional method still relies on manual analysis and judgment of impedance data. It lacks a standardized automated judgment process, which makes it difficult to meet the needs of the production line for rapid, accurate, and batch qualification judgment of individual converters. It cannot adapt to the high-efficiency scenario characteristics of off-line testing.

[0003] Chinese patent CN119315514A discloses a method and system for rapid evaluation of wind farm grid-connected stability based on impedance models: S1. Establishing a physical model of the wind farm grid-connected system: Collecting data from the wind farm grid-connected system, analyzing the overall structure and parameters of the system, and constructing a physical model of the system. The physical model shows the interaction relationships between the components, providing a framework for subsequent impedance modeling; S2. Establishing impedance models for the wind farm and the power grid: The impedance model of the wind farm includes the impedance of the wind turbine system, the impedance of the converter, and the total impedance of the wind farm. Impedance; The impedance model of the power grid includes the impedance of transmission lines, transformers, and the load impedance of the power grid. By calculating the impedances of the wind farm and the power grid, the complex impedance matrix of the wind farm grid-connected system is obtained; S3. Select appropriate evaluation indicators to determine the evaluation indicators to measure the stability of the wind farm grid-connected system. Evaluation indicators include voltage stability, frequency stability, and power factor; When selecting evaluation indicators, the characteristics of the wind farm and the requirements of the power grid should be considered, such as selecting indicators for voltage transient response, frequency response, and power fluctuation for analysis; S4. Perform impedance matching analysis by comparing the impedances of the wind farm and the power grid. S5. Evaluate dynamic response characteristics: Assess the wind farm's response to disturbances during grid connection by simulating the actual operation of the wind farm and the grid, and observing the system's response behavior under disturbances. S6. Perform steady-state analysis: Evaluate the power balance, power factor, voltage, and frequency stability of the wind farm grid-connected system by calculating the system's power flow, load distribution, and voltage and frequency stability, and determine the system's stability under steady-state conditions. S7. Evaluate the impact of grid connection on the grid: Assess the impact of wind farm grid connection on the grid, and determine whether it will negatively affect grid stability. Analyze grid voltage fluctuations, frequency changes, and system harmonics using power system simulation tools to simulate and analyze the grid after grid connection. S8. Design control strategies: Control strategies include wind turbine power control, converter current control, and system voltage and frequency regulation. Improve system stability and performance by designing control algorithms such as PID control and fuzzy control, and conduct multiple simulation tests to determine the optimal control parameters and strategies.

[0004] While the above methods can meet the needs of most scenarios, research and practical application of these methods and existing technologies have revealed at least the following shortcomings:

[0005] The above methods cannot quickly and automatically determine the grid connection stability quality of a single new energy converter in a production line off-line testing scenario.

[0006] In view of this, the present invention proposes an electrical automation quality inspection method and system to solve the above problems. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an electrical automation quality inspection method, comprising:

[0008] The raw voltage and current signals at the power grid's point of common coupling (PCC) are collected and subjected to fast Fourier transform to analyze and obtain the set of background harmonic frequencies and the set of background harmonic amplitudes.

[0009] Preset the target scanning frequency band and generate an initial frequency grid at fixed intervals; generate a sparse frequency grid based on the background harmonic frequency set and the initial frequency grid.

[0010] The estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid is calculated and sorted in ascending order to obtain the injection sequence. The injection sequence is input into the control power amplifier to obtain the sinusoidal injection signal. The sinusoidal injection signal is injected into the PCC point through the current coupler.

[0011] Simultaneously with the injection of a sinusoidal signal, the voltage response and total current at the PCC point are collected; the signal parameters corresponding to the sinusoidal injection signal are extracted, and the complex impedance is calculated based on the signal parameters to obtain frequency impedance data pairs.

[0012] Based on the frequency impedance data, a rational function is fitted, and the coefficients are solved using the least squares method to obtain the broadband impedance curve.

[0013] The broadband impedance curve is analyzed to extract quality characteristic parameters, and the quality is judged based on the quality characteristic parameters.

[0014] Furthermore, methods for obtaining background harmonic frequencies and background harmonic amplitudes include:

[0015] The original voltage and current signals are preprocessed to obtain preprocessed voltage and current signals.

[0016] Perform N-point Fast Fourier Transform on the preprocessed voltage and current signals respectively to obtain complex voltage and current frequency domain signals;

[0017] The voltage amplitude spectrum and current amplitude spectrum are calculated by combining the voltage frequency domain signal and the current frequency domain signal with the signal gain and energy dispersion coefficient;

[0018] The actual frequency corresponding to the m-th index is calculated based on the sampling rate and the number of sampling points N, and the frequency axis is obtained by counting all actual frequencies.

[0019] The frequency corresponding to the spectral peak with the largest amplitude in the frequency axis is statistically extracted as the fundamental frequency; the voltage amplitude spectrum and current amplitude spectrum corresponding to the fundamental frequency are obtained as the fundamental voltage amplitude and fundamental current amplitude.

[0020] Traverse the frequency axis to obtain all frequency points whose actual frequency is greater than twice the fundamental frequency as the frequencies to be sorted. Sort the frequencies to be sorted in ascending order that meet the screening conditions, and extract the corresponding frequencies and amplitudes to obtain the background harmonic frequency set and the corresponding harmonic amplitude set.

[0021] Furthermore, the screening conditions include condition 1 and condition 2, wherein condition 1: the voltage amplitude of the frequency to be sorted is not lower than the harmonic voltage identification threshold or the current amplitude of the candidate frequency is not lower than the harmonic current identification threshold.

[0022] Condition 2: Within the allowable deviation range, the frequency to be sorted is n times the fundamental frequency, where n is an integer not less than 2.

[0023] Furthermore, methods for generating sparse frequency grids include:

[0024] The target scanning frequency band is divided according to a fixed interval to obtain an initial grid frequency set. The number of grid points is calculated based on the upper and lower limits of the target scanning frequency band and the fixed interval.

[0025] Set an interference avoidance bandwidth. For each initial frequency in the initial grid frequency set, if the initial frequency point is within the interference avoidance bandwidth of the background harmonic frequency, then remove the corresponding initial frequency and construct a sparse frequency grid based on the remaining frequency points.

[0026] Furthermore, methods for obtaining the injection sequence include:

[0027] The background noise amplitude is obtained by performing cubic spline interpolation on the harmonic spectrum of each candidate frequency in the sparse frequency grid.

[0028] The estimated signal-to-noise ratio is calculated based on the preset effective value of the injected signal, the system calibration coefficient, and the background noise amplitude. The estimated signal-to-noise ratios are then sorted in ascending order to obtain the injection sequence.

[0029] Furthermore, methods for obtaining the background noise amplitude include:

[0030] Within each segmented interval, define a cubic spline interpolation function that satisfies the interpolation conditions;

[0031] Construct the cubic interpolation function expression for each segmented interval;

[0032] Establish a system of equations based on the interpolation conditions, solve for the coefficients of all segmented intervals, obtain the interpolation function for all segmented intervals, and obtain the continuous interpolation function.

[0033] Traverse the frequency points of the sparse frequency grid to determine the segment interval where the candidate frequency is located; extract the corresponding segment interval and the corresponding interpolation coefficient;

[0034] Substitute the candidate frequencies into the corresponding segmented cubic interpolation function to calculate the background noise amplitude.

[0035] Furthermore, the interpolation conditions include:

[0036] At the endpoints of the segmented interval, the interpolated value is equal to the original magnitude;

[0037] The first and second derivatives are continuous at the junctions of piecewise intervals;

[0038] Natural boundaries are used, meaning the second derivative values ​​at the two endpoints of the segmented interval are 0.

[0039] Furthermore, methods for obtaining frequency impedance data pairs include:

[0040] A sinusoidal injection signal is calculated based on the injection sequence. This signal is then injected into the PCC point via a current coupler. Simultaneously, the voltage response and total current at the PCC point are acquired. The signal parameters corresponding to the sinusoidal injection signal are extracted, including the effective value of the voltage response, voltage phase, effective value of the total current, and current phase. The impedance phase is calculated based on the voltage and current phases. The complex impedance is then calculated by combining the effective value of the voltage response, the effective value of the total current, and the impedance phase. Finally, the data are spliced ​​together with the frequency to obtain frequency-impedance data pairs.

[0041] Furthermore, the quality characteristic parameters include phase margin, gain margin, Nyquist distance, and maximum impedance at a specific frequency band;

[0042] Methods for obtaining phase margin include:

[0043] Traverse the wideband impedance curve to obtain the frequency corresponding to the impedance value of 1 as the amplitude crossover frequency, calculate the difference between the impedance phase at the amplitude crossover frequency and -180°, and obtain the phase margin.

[0044] Methods for obtaining gain margin include:

[0045] Traverse the wideband impedance curve to obtain the frequency corresponding to the impedance phase of -180° as the phase crossover frequency, calculate the negative logarithm of the impedance amplitude corresponding to the phase crossover frequency, and obtain the amplitude margin.

[0046] Methods for obtaining the maximum impedance in a specific frequency band include:

[0047] The impedance curve of the target frequency band is extracted from the continuous impedance curve. A discrete frequency sequence is obtained by sampling based on the sampling step size. Each frequency point in the discrete frequency sequence is substituted into the continuous impedance function to calculate the real and imaginary parts of the corresponding complex impedance. For each discrete frequency point, the corresponding amplitude is calculated based on the real and imaginary parts of the corresponding complex impedance. All NL discrete frequency points are traversed, where NL is the number of discrete frequency points, and the amplitude of each point is calculated to generate an impedance amplitude array for a specific frequency band. The maximum impedance amplitude is extracted from the impedance amplitude array as the maximum impedance value for the specific frequency band.

[0048] Furthermore, methods for obtaining the Nyquist distance include:

[0049] Calculate the rated reference impedance, and then calculate the per-unit complex impedance using the continuous impedance function.

[0050] The perimetric complex impedance is sampled within a preset frequency range with a frequency step size of [missing information]. Discrete sampling is performed to generate a discrete frequency sequence; each frequency point in the discrete frequency sequence is substituted into the calculation to obtain the corresponding per-unit complex impedance, and a discrete complex impedance array is obtained, which is then split to obtain the real part array and the imaginary part array.

[0051] Construct a complex plane coordinate system with the real axis as the horizontal axis and the imaginary axis as the vertical axis, and draw the Nyquist curve by combining the real part array and the imaginary part array;

[0052] For the g-th point in the discrete complex impedance array, calculate its Euclidean distance to the point (-1, j0). Iterate through all discrete complex impedance points, calculate the Euclidean distance for each point, generate a distance array, and take the minimum value in the distance array as the Nyquist distance.

[0053] Furthermore, methods for quality assessment include:

[0054] If all quality characteristic parameters meet the following conditions: phase margin not lower than phase threshold, amplitude margin not lower than amplitude threshold, Nyquist distance not lower than distance threshold, and maximum impedance in a specific frequency band not higher than impedance threshold, then the quality is deemed qualified; otherwise, it is deemed unqualified, and the corresponding unqualified parameters and deviation values ​​are output.

[0055] An electrical automation quality inspection system, implementing the aforementioned electrical automation quality inspection method, includes:

[0056] Passive monitoring module: Collects raw voltage and current signals from the power grid's point of common coupling (PCC) and performs fast Fourier transform to analyze and obtain the set of background harmonic frequencies and the set of background harmonic amplitudes;

[0057] Mesh generation module: Preset target scanning frequency band, generate initial frequency mesh at fixed intervals; generate sparse frequency mesh based on background harmonic frequency set and initial frequency mesh;

[0058] Intelligent injection module: Calculates the estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid and sorts them in ascending order to obtain the injection sequence. The injection sequence is input into the control power amplifier to obtain a sinusoidal injection signal. The sinusoidal injection signal is injected into the PCC point through a current coupler.

[0059] Data acquisition module: While injecting a sinusoidal injection signal, it acquires the voltage response and total current at the PCC point; extracts the signal parameters corresponding to the sinusoidal injection signal, calculates the complex impedance based on the signal parameters, and obtains frequency impedance data pairs;

[0060] Impedance calculation module: Fits rational functions to frequency impedance data and solves for coefficients using the least squares method to obtain wideband impedance curves;

[0061] Quality Inspection Module: Analyzes wideband impedance curves, extracts quality characteristic parameters, and makes quality judgments based on these parameters.

[0062] The technical effects and advantages of the electrical automation quality inspection method and system of the present invention are as follows:

[0063] This invention effectively solves the problems of low detection accuracy, excessive manual intervention, and low judgment efficiency in the current technology for detecting the grid-connected stability of individual new energy converters off the production line. This is achieved through passive monitoring and analysis of background harmonics at the PCC point, generating a sparse frequency grid to avoid interference, optimizing the sinusoidal signal injection strategy by sorting signals in ascending order of estimated signal-to-noise ratio, synchronously acquiring response signals to calculate discrete frequency impedance data pairs, reconstructing a wideband impedance curve based on rational function fitting and least squares method, and automatically extracting four core quality characteristic parameters: phase margin, gain margin, Nyquist distance, and maximum impedance value in a specific frequency band. Finally, it automatically compares all parameters with preset thresholds derived from international standards, power grid guidelines, and enterprise internal controls. The technical problem of poor consistency in results, making it difficult to adapt to the needs of rapid batch testing on production lines, is addressed by employing background harmonic avoidance, rational function fitting, and standardized parameter calculation to ensure that the errors in impedance measurement and feature parameter extraction are controlled at a low level, providing reliable data support for judgment. From frequency grid generation, signal injection, impedance calculation, parameter extraction to threshold comparison, the entire process requires no manual intervention, eliminating subjective errors. By reducing the number of invalid injections through sparse frequency grids, the automated process significantly shortens the testing time for a single converter. At the same time, the automatic output of unqualified parameters and deviation values ​​can guide rapid rectification. This perfectly adapts to the core requirements of production line scenarios for rapid, automated, and accurate qualification judgment of the grid connection stability of individual new energy converters, significantly improving the testing efficiency and judgment reliability of the production line. Attached Figure Description

[0064] Figure 1This is a schematic diagram of an electrical automation quality inspection method according to the present invention;

[0065] Figure 2 This is a schematic diagram of the data flow in this invention;

[0066] Figure 3 This is a schematic flowchart of the method for obtaining background noise amplitude according to the present invention;

[0067] Figure 4 This is a schematic diagram of an electrical automation quality inspection system according to the present invention. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] Example 1

[0070] Please see Figure 1 , Figure 2 As shown in this embodiment, an electrical automation quality inspection method includes:

[0071] The raw voltage and current signals of the power grid's point of common coupling (PCC) are collected and subjected to fast Fourier transform to analyze and obtain the background harmonic frequency and amplitude.

[0072] Methods for obtaining background harmonic frequencies and background harmonic amplitudes include:

[0073] The original voltage and current signals are preprocessed to obtain preprocessed voltage and current signals.

[0074] Perform N-point Fast Fourier Transform on the preprocessed voltage and current signals respectively to obtain complex form voltage frequency domain signals and current frequency domain signals; such as , ,in, m is the frequency index, and N is the number of sampling points. This is the Fast Fourier Transform; where the real part of the complex number corresponds to the in-phase component of the signal, and the imaginary part corresponds to the quadrature component.

[0075] The voltage amplitude spectrum and current amplitude spectrum are calculated by combining the voltage frequency domain signal and the current frequency domain signal with the signal gain and energy dispersion coefficient; for example, the voltage amplitude spectrum... Current amplitude spectrum ,in, Let be the voltage magnitude of the Fast Fourier Transform at the m-th frequency point; Let be the fast Fourier transform current magnitude at the m-th frequency point; For voltage signal gain; For current signal gain; The energy dispersion coefficient;

[0076] The actual frequency corresponding to the m-th index is calculated based on the sampling rate and the number of sampling points N. For example, the actual frequency corresponding to the m-th index... The frequency axis is obtained by statistically analyzing all actual frequencies. Sampling rate;

[0077] The frequency corresponding to the spectral peak with the largest amplitude in the frequency axis is statistically extracted as the fundamental frequency; the voltage amplitude spectrum and current amplitude spectrum corresponding to the fundamental frequency are obtained as the fundamental voltage amplitude and fundamental current amplitude.

[0078] Traverse the frequency axis to obtain all frequency points whose actual frequency is greater than twice the fundamental frequency as the frequencies to be sorted. Sort the frequencies to be sorted in ascending order that meet the screening conditions, and extract the corresponding frequencies and amplitudes to obtain the background harmonic frequency set and the corresponding harmonic amplitude set.

[0079] The screening criteria include criterion 1 and criterion 2. Criterion 1: The voltage amplitude of the frequency to be sorted is not lower than the harmonic voltage identification threshold or the current amplitude of the candidate frequency is not lower than the harmonic current identification threshold.

[0080] Condition 2: Within the allowable deviation range, the frequency to be sorted is n times the fundamental frequency, where n is an integer not less than 2.

[0081] By collecting the raw voltage and current signals from the grid common connection point (PCC) during the grid connection testing of new energy converters, the signals are first preprocessed to eliminate DC components and suppress interference. Then, N-point Fast Fourier Transforms are performed to obtain complex voltage and current frequency domain signals. The voltage and current amplitude spectra are calculated by combining the voltage and current signal gains and the energy dispersion coefficient. Based on the sampling rate and the number of sampling points N, the actual frequencies corresponding to each FFT index are determined, forming a frequency axis. The fundamental frequency and fundamental voltage and current amplitudes corresponding to the maximum amplitude peak in the frequency axis are further identified. Finally, frequencies greater than twice the fundamental frequency, whose voltage amplitude is not lower than the harmonic voltage identification threshold or whose current amplitude is not lower than the harmonic current identification threshold and are within the allowable deviation range of n times the fundamental frequency (where n is an integer not less than 2) are selected to obtain the background harmonic frequency set. The core function of this process, which combines the corresponding harmonic amplitude sets, is to accurately identify the interference frequency and intensity of grid background harmonics in the scenario of testing a single new energy converter off the production line. This provides a key basis for subsequent adaptive avoidance of these interference frequencies and optimization of test signal injection strategies, effectively avoiding interference from background harmonics on the converter's grid-connected impedance measurement, significantly reducing the number of invalid signal injections to improve impedance measurement speed, and ensuring impedance measurement accuracy. Accurate broadband impedance data is the basis for the subsequent automatic extraction of stability characteristic parameters such as phase margin and Nyquist distance, and automatic comparison with qualified thresholds. Ultimately, this achieves rapid and automated qualification judgment of the grid-connected stability quality of a single new energy converter, solving the technical problems of existing technologies that are unable to effectively eliminate background harmonic interference, resulting in long testing times, low accuracy, reliance on manual intervention, and difficulty in adapting to the high-efficiency testing needs of the production line.

[0082] Preset the target scanning frequency band and generate an initial frequency grid at fixed intervals; generate a sparse frequency grid based on the background harmonic frequency set and the initial frequency grid.

[0083] Methods for generating sparse frequency grids include:

[0084] The target scanning frequency band is divided according to a fixed interval to obtain an initial grid frequency set. The number of grid points is calculated based on the upper and lower limits of the target scanning frequency band and the fixed interval; for example, the number of grid points... ,in, and These are the upper and lower limits of the target scanning frequency band, respectively; For fixed intervals;

[0085] Set an interference avoidance bandwidth. For each initial frequency in the initial grid frequency set, if the initial frequency point is within the interference avoidance bandwidth of the background harmonic frequency, then remove the corresponding initial frequency and construct a sparse frequency grid based on the remaining frequency points.

[0086] By eliminating background harmonic interference frequencies, measurement errors caused by background interference are avoided when injecting test signals into the converter's PCC point, ensuring the accuracy of grid-connected impedance measurement from the source. This provides a reliable data foundation for the subsequent extraction of stability characteristic parameters such as phase margin and Nyquist distance. The sparse frequency grid significantly reduces the number of invalid frequency point measurements, significantly shortening the testing time for a single converter and adapting to the pace of efficient off-line testing on the production line. At the same time, the standardized frequency grid generation and interference elimination process requires no manual intervention and can be directly embedded into the automated testing system. This provides standardized frequency dimension support for the subsequent extraction of stability characteristic parameters based on impedance data, automatic comparison with qualified thresholds, and final qualification determination. It effectively solves the technical problems of existing technologies, such as low measurement accuracy and long testing time due to failure to avoid background interference, and the difficulty in achieving automated judgment due to reliance on manual adjustment of frequency points, which cannot meet the needs of off-line testing on the production line.

[0087] The estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid is calculated and sorted in ascending order to obtain the injection sequence. The injection sequence is input into the control power amplifier to obtain the sinusoidal injection signal. The sinusoidal injection signal is injected into the PCC point through the current coupler.

[0088] Methods for obtaining the injection sequence include:

[0089] The background noise amplitude is obtained by performing cubic spline interpolation on the harmonic spectrum of each candidate frequency in the sparse frequency grid.

[0090] Reference Figure 3 Methods for obtaining background noise amplitude include:

[0091] The sparse frequency grid is divided into K-1 segmented intervals. Within each segmented interval of the sparse frequency grid, a cubic spline interpolation function satisfying the interpolation conditions is defined. ;

[0092] Interpolation conditions include:

[0093] At the endpoints of the segmented intervals, the interpolated value equals the original amplitude; that is... , ,in, For segmented intervals The corresponding cubic spline interpolation function is the first Frequency points The function value at that location; where, For the first One frequency point; For the first The cubic spline interpolation function value corresponding to each frequency point; For the first The original amplitude corresponding to each frequency point; For the first One frequency point; For segmented intervals The corresponding cubic spline interpolation function is the first Frequency points The function value at that location; For the first The original amplitude corresponding to each frequency point;

[0094] The first and second derivatives are continuous at the junctions of the segmented intervals; that is... , ;in, For cubic spline interpolation functions No. Frequency points The first derivative at that point; For cubic spline interpolation functions No. Frequency points The first derivative at that point; For cubic spline interpolation functions No. Frequency points The second derivative at point; For cubic spline interpolation functions No. Frequency points The second derivative at point;

[0095] Natural boundaries are used, meaning the second derivative values ​​at the two endpoints of the segmented interval are 0; ; ; This is the value of the second derivative at the first frequency point; For the first The second derivative values ​​at each frequency point;

[0096] Construct the cubic interpolation function expression for each segmented interval; such as the cubic interpolation function. ,in, , , , For the first Interpolation coefficients for each segment; For local variables within a segment, For the cubic interpolation function variables;

[0097] Establish a system of equations based on the interpolation conditions, solve for the coefficients of all segmented intervals, obtain the interpolation function for all segmented intervals, and obtain the continuous interpolation function; for example:

[0098] By interpolation conditions Substituting into ;

[0099] right Find the first derivative : ,Depend on have to: ;in, The length of the segmented interval;

[0100] right Find the second derivative : ,Depend on have to: Organized ;

[0101] Substituting this into the first derivative equation and combining it with the condition of continuity of the first derivative... Establish about The three-moment equation system: ;

[0102] Substituting natural boundary conditions , Solving the above system of linear equations yields all... ;

[0103] Awaiting further explanation and : , .

[0104] By following the steps above, the interpolation coefficients for all segmented intervals can be determined, thus identifying the complete continuous interpolation function. ,exist inside, when , .

[0105] Traverse the frequency points of the sparse frequency grid to determine the segment interval where the candidate frequency is located; extract the corresponding segment interval and the corresponding interpolation coefficient;

[0106] Substitute the candidate frequencies into the corresponding piecewise cubic interpolation function to calculate the background noise amplitude; for example, the background noise amplitude... ,in, For candidate frequencies, , , and All of these are interpolation coefficients.

[0107] The estimated signal-to-noise ratio (SNR) is calculated based on the preset effective value of the injected signal, the system calibration coefficient, and the background noise amplitude. The estimated SNRs are then sorted in ascending order to obtain the injection sequence. (The estimated SNR is...) ,in, The preset effective value of the injected signal is set according to the station capacity, such as 100MW station. ; The system calibration coefficients are determined by the gain of the acquisition unit and the efficiency of the coupling device, and are typically calibrated beforehand using a 100Ω standard resistor. =1.05-1.2.

[0108] Cubic spline interpolation can accurately fit the background noise amplitude, ensuring the accuracy of the estimated signal-to-noise ratio (SNR). Prioritizing injection of low SNR frequencies in ascending order of SNR ensures high SNR across the entire impedance measurement band with minimal injection energy, avoiding insufficient measurement accuracy due to unreasonable signal injection strategies. This provides reliable data support for subsequent accurate calculation of wideband impedance and extraction of stability characteristic parameters such as phase margin. The entire process, from background noise amplitude calculation to injection sequence generation, is automatically executed through standardized algorithms, eliminating the need for manual adjustment of the injection frequency or judgment of interference intensity, significantly reducing manual intervention. Prioritizing injection of key frequencies avoids wasting time on invalid signal injection, significantly improving the detection efficiency of a single converter. This perfectly meets the core requirements of speed and automation for production line off-line testing, providing efficient and accurate pre-injection signal guarantee for subsequent extraction of stability characteristic parameters based on impedance data, automatic comparison with qualified thresholds, and final qualification determination. It effectively solves the technical problems of existing technologies, such as reliance on manual signal injection strategies, measurement accuracy being affected by background noise, and low detection efficiency, making it difficult to achieve rapid and automated qualification determination of grid-connected stability for a single converter in production line off-line scenarios.

[0109] A sinusoidal injection signal is generated by a controlled power amplifier according to the injection sequence and injected into the PCC point. At the same time as the sinusoidal injection signal is injected, the voltage response and total current of the PCC point are collected. The signal parameters corresponding to the sinusoidal injection signal are extracted, and the complex impedance is calculated based on the signal parameters to obtain frequency impedance data pairs.

[0110] Methods for obtaining frequency impedance data pairs include:

[0111] A sinusoidal injection signal is calculated based on the injection sequence. This signal is then injected into the PCC point via a current coupler. Simultaneously, the voltage response and total current at the PCC point are acquired. Signal parameters corresponding to the sinusoidal injection signal are extracted, including the RMS voltage response, voltage phase, RMS total current, and current phase. The impedance phase is calculated based on the voltage and current phases. The complex impedance is then calculated by combining the RMS voltage response, RMS total current, and impedance phase. These parameters are then concatenated with the frequency data to obtain a frequency-impedance pair. (This is how a sinusoidal injection signal is generated.) ,in, For the initial phase, For the first in the injection sequence One frequency point; For time; the injection sequence in the [time]th [time]. Complex impedance at each frequency point ,in, The impedance amplitude, This is the effective value of the voltage response. This is the effective value of the total current; For impedance phase, For voltage phase, For current phase; It is an imaginary number; It is a mathematical constant; It is the effective value of the current.

[0112] By employing a standardized signal injection and parameter extraction process, automated calculation of complex impedance is achieved, avoiding errors and inefficiencies caused by manual data reading and calculation. This ensures the accuracy and consistency of frequency impedance data pairs. The generated frequency impedance data pairs serve as the direct basis for subsequent rational function fitting to obtain wideband impedance curves and extract stability characteristic parameters such as phase margin. Their accuracy directly determines the reliability of stability determination. Simultaneously, combined with the previous sparse frequency grid and optimized injection sequence, this process only performs signal injection and impedance calculation on key effective frequency points, significantly reducing redundant operations and substantially improving the detection speed of a single converter. This perfectly adapts to the needs of production line off-line scenarios for speed and automation, providing core data support for the final automatic qualification determination of grid-connected stability quality of a single new energy converter by comparing stability characteristic parameters and qualification thresholds. This effectively solves the technical problems of existing technologies that rely on manual impedance measurement processes, have insufficient data accuracy, and low detection efficiency, making it difficult to achieve rapid and automated qualification determination in production line off-line scenarios.

[0113] Based on the frequency impedance data, a rational function is fitted, and the coefficients are solved using the least squares method to obtain a broadband impedance curve; if a fitted model is obtained... ,in, , The order of the polynomial is set according to the number of data points. , The coefficients are to be determined; the goal is to minimize the sum of squared errors between the fitted values ​​and the measured values, and the coefficients are solved using the least squares method. ,in, To fit the phase of the impedance, M is the number of frequency points in the injected sequence. For the first in the injection sequence Fitted model values ​​at each frequency point For the first in the injection sequence Measured impedance amplitude at each frequency point For the first in the injection sequence The measured impedance phase at each frequency point.

[0114] The rational function model aligns with the physical characteristics of the grid-connected impedance of new energy converters. Combined with optimal error control using the least squares method, it can keep the fitting error of discrete data at a low level, ensuring that the wideband impedance curve accurately reflects the full-band impedance characteristics of the converter. This provides a continuous and reliable data carrier for the subsequent extraction of key stability characteristic parameters such as phase margin and Nyquist distance. Discrete frequency impedance data pairs cannot directly support full-band stability analysis, while continuous curves can completely cover key oscillation frequency bands such as subsynchronous and supersynchronous, avoiding omissions or deviations in characteristic parameter extraction. The entire fitting process can be embedded into an automated detection system through standardized algorithms, eliminating the need for manual adjustment of model parameters or intervention in the calculation process. This fully adapts to the "automated" detection requirements of the production line. At the same time, the least squares method has high solution efficiency and does not add extra detection time. In conjunction with the previously optimized signal injection and impedance calculation process, it further ensures a "fast" detection rhythm. This effectively solves the technical problems in existing technologies, such as the difficulty in extracting stability characteristic parameters due to the lack of accurate and continuous impedance curves, reliance on manual experience to supplement data, poor consistency of detection results, and difficulty in achieving rapid and automated qualification of grid-connected stability of a single converter in production line scenarios.

[0115] The broadband impedance curve is analyzed to extract quality characteristic parameters, and the quality is judged based on the quality characteristic parameters.

[0116] Methods for obtaining quality characteristic parameters include:

[0117] Quality characteristic parameters include phase margin, gain margin, Nyquist distance, and maximum impedance at a specific frequency band;

[0118] Traverse the wideband impedance curve to obtain the frequency corresponding to an impedance value of 1 as the amplitude crossover frequency. Calculate the difference between the impedance phase at the amplitude crossover frequency and -180° to obtain the phase margin; if the phase margin... ,in, The amplitude crossover frequency; amplitude crossover frequency The impedance phase margin is an internationally standardized quantitative indicator of grid-connected stability. Its value directly reflects the converter's disturbance rejection capability, replacing the subjective judgment method of relying on manual experience to analyze the impedance curve in existing technologies. This eliminates the errors and inconsistencies of manual judgment and provides a standardized and quantifiable core basis for stability assessment. From traversing the broadband impedance curve to locate the amplitude crossover frequency to calculating the phase margin, the entire process can be automatically executed by the algorithm. It can be seamlessly integrated into the automated testing process of the production line, significantly shortening the time for indicator extraction and adapting to the rapid pace of offline testing. At the same time, accurate phase margin data is a key input for subsequent automated comparison with preset qualified thresholds to ultimately determine whether the converter's grid-connected stability is qualified. Without this indicator, quantitative qualification judgment cannot be completed. This effectively solves the technical problem of existing technologies lacking standardized and automatically extracted quantitative stability indicators and relying on manual experience, resulting in low judgment efficiency, inconsistent results, and difficulty in achieving rapid and automated qualification judgment of grid-connected stability of individual converters in offline production line scenarios.

[0119] Traverse the broadband impedance curve to obtain the frequency corresponding to an impedance phase of -180° as the phase crossover frequency. Calculate the negative logarithm of the impedance magnitude corresponding to the phase crossover frequency to obtain the magnitude margin; for example, the magnitude margin... ,in, The phase crossover frequency; Phase crossover frequency impedance, It is a logarithmic function. Gain margin is a key stability quantification indicator clearly defined in international standards and power grid guidelines. Its standardized calculation replaces the subjective mode of relying on manual observation of impedance curves and experience to judge stability in existing technologies, completely eliminating errors and inconsistencies in manual judgment, and providing an objective and unified quantitative basis for stability assessment. From automatically locating the phase crossover frequency in the broadband impedance curve to accurately calculating the gain margin according to the formula, the entire process can be completed through an algorithm embedded in the automated detection system, eliminating the need for manual frequency point searching or logarithm calculation, significantly shortening the indicator extraction time, and perfectly adapting to the fast and efficient pace of production line offline testing. Simultaneously, accurate gain margin data is the core input for subsequent automated comparison with preset pass thresholds. Without this indicator, quantitative pass determination of stability cannot be achieved. This effectively solves the technical problem of existing technologies lacking automatically extracted standardized gain margin indicators and relying on manual experience, resulting in low judgment efficiency and poor result reliability, making it difficult to quickly and automatically determine the grid-connected stability of a single converter in production line offline scenarios.

[0120] Methods for obtaining the Nyquist distance include:

[0121] Calculate the rated reference impedance, and then use the continuous impedance function to calculate the per-unit complex impedance; for example, the rated reference impedance. Per-unit complex impedance ,in, The rated voltage at the PCC point; Rated capacity for PCC point; for The real part, ; for The imaginary part, ; The impedance of the cubic interpolation function variable f;

[0122] The perimetric complex impedance is sampled within a preset frequency range with a frequency step size of [missing information]. Discrete sampling is performed to generate a discrete frequency sequence; each frequency point in the discrete frequency sequence is... Substitute the values ​​into the calculation to obtain the corresponding per-unit complex impedance, obtain the discrete complex impedance array, and split it to obtain the real part array and the imaginary part array;

[0123] Construct a complex plane coordinate system with the real axis as the horizontal axis and the imaginary axis as the vertical axis, and draw the Nyquist curve by combining the real part array and the imaginary part array;

[0124] For the g-th point in the discrete complex impedance array, calculate its relationship with... The Euclidean distance of a point is calculated by iterating through all discrete complex impedance points, calculating the Euclidean distance for each point, generating a distance array, and then taking the minimum value in the array as the Nyquist distance. (Euclidean distance example follows.) ,in, Let g be the real part corresponding to the g-th point; Let g be the imaginary part corresponding to the g-th point. This represents the g-th complex impedance element in the discrete complex impedance array. Perimeter normalization eliminates the impedance differences between converters with different rated parameters, providing a unified and comparable quantitative standard for the Nyquist distance, thus meeting the needs of batch testing converters of different specifications on production lines. The entire process, from perimeter calculation and discrete sampling to distance traversal solution, is automatically executed through standardized algorithms, eliminating the need for manual drawing of the Nyquist curve and subjective judgment of the distance to the (-1,j0) point, completely avoiding human error and significantly improving testing efficiency. Furthermore, as an internationally recognized core stability indicator, the precise value of the Nyquist distance can be directly and automatically compared with a preset pass threshold, serving as a key quantitative basis for determining whether the grid-connected stability of a converter is qualified. This effectively solves the technical problem of existing technologies relying on manual analysis of the Nyquist curve, resulting in low efficiency and poor consistency of results, making it difficult to achieve rapid and automated qualification determination of the grid-connected stability of a single converter in production line scenarios.

[0125] The impedance curve of the target frequency band is extracted from the continuous impedance curve. A discrete frequency sequence is obtained by sampling based on the sampling step size. Each frequency point in the discrete frequency sequence is substituted into the continuous impedance function to calculate the real and imaginary parts of the corresponding complex impedance. For each discrete frequency point, the corresponding amplitude is calculated based on the real and imaginary parts of the corresponding complex impedance. All NL discrete frequency points are traversed, where NL is the number of discrete frequency points, and the amplitude of each point is calculated to generate an impedance amplitude array for a specific frequency band. The maximum impedance amplitude is extracted from the impedance amplitude array as the maximum impedance value for the specific frequency band. The precise interception of the target frequency band focuses on the key risk range of grid-connected stability analysis, avoiding interference from irrelevant frequency band data. This ensures that the extracted maximum value directly reflects the impedance characteristics of the converter in the high-oscillation-risk frequency band, providing a targeted basis for the quantitative assessment of specific types of oscillation risks such as subsynchronous and supersynchronous. The entire process, from frequency band interception and discrete sampling to amplitude calculation and maximum value extraction, is automatically executed through standardized algorithms. There is no need for manual selection of key frequency bands, manual calculation of amplitude, or judgment of peak values, completely eliminating the errors and efficiency bottlenecks of manual operation. This significantly shortens the time required for extracting indicators for a single converter, perfectly meeting the rapid needs of batch testing on production lines. At the same time, the maximum impedance value of this specific frequency band, as an oscillation risk quantification indicator recognized by international standards and power grid guidelines, can be directly and automatically compared with the pre-stored qualified threshold. This is the core risk assessment input for subsequent completion of the grid-connected stability qualification determination of the converter. It effectively solves the technical problems of low detection efficiency and poor result consistency caused by the reliance on manual data processing, subjective judgment of oscillation risk, and non-automated processes in existing technologies, making it difficult to achieve rapid and automated qualification determination of grid-connected stability of a single converter in production line scenarios.

[0126] Methods for quality assessment include:

[0127] If all quality characteristic parameters meet the following conditions: phase margin not lower than phase threshold, amplitude margin not lower than amplitude threshold, Nyquist distance not lower than distance threshold, and maximum impedance in a specific frequency band not higher than impedance threshold, then the quality is deemed qualified; otherwise, it is deemed unqualified, and the corresponding unqualified parameters and deviation values ​​are output.

[0128] In the scenario of detecting the grid connection stability quality of a single new energy converter offline on the production line, by automatically comparing the four core quality characteristic parameters of phase margin, amplitude margin, Nyquist distance, and maximum impedance in a specific frequency band extracted in the early stage with the pre-stored phase threshold, amplitude threshold, distance threshold, and impedance threshold derived from international standards, grid guidelines, and enterprise internal control requirements respectively, the quality is only determined to be qualified when all parameters meet the conditions that the phase margin is not lower than the phase threshold, the amplitude margin is not lower than the amplitude threshold, the Nyquist distance is not lower than the distance threshold, and the maximum impedance in the specific frequency band is not higher than the impedance threshold. Otherwise, it is determined to be unqualified and the specific unqualified parameters and deviation values are automatically output. The core function of this process is to completely replace the mode in the existing technology that relies on manual experience to analyze impedance curves and subjectively judge stability. Through the standardized judgment logic of quantifying indicators, defining thresholds, and automatic comparison, it eliminates the errors and result inconsistencies of manual judgment. At the same time, the entire judgment process does not require manual intervention, and the process from parameter input to result output can be completed within seconds, greatly improving the detection and judgment efficiency of a single converter, perfectly adapting to the rapid demand of the production line for batch offline detection. In addition, automatically outputting unqualified parameters and deviation values can also provide accurate guidance for subsequent fault troubleshooting and parameter debugging on the production line, forming a closed-loop of detection, judgment, and rectification, effectively solving the technical problem in the existing technology that it is difficult to achieve rapid and automatic qualified judgment of the grid connection stability quality of a single new energy converter in the production line offline scenario due to the manual-dependent judgment process, large subjective factors, and low efficiency.

[0129] Embodiment 2

[0130] Please refer to Figure 4 As shown, an electrical automation quality detection system described in this embodiment includes:

[0131] Passive monitoring module: Collect the original voltage signal and original current signal at the point of common coupling (PCC) of the power grid, perform fast Fourier transform, and analyze to obtain the background harmonic frequency set and background harmonic amplitude set;

[0132] Grid generation module: Preset the target scanning frequency band, generate an initial frequency grid at a fixed interval; generate a sparse frequency grid according to the background harmonic frequency set and the initial frequency grid;

[0133] Intelligent injection module: Calculate the estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid and sort them in ascending order to obtain an injection sequence, input the injection sequence into the control power amplifier to obtain a sinusoidal injection signal, and inject the sinusoidal injection signal into the PCC point through a current coupler;

[0134] Data acquisition module: While injecting a sinusoidal injection signal, it acquires the voltage response and total current at the PCC point; extracts the signal parameters corresponding to the sinusoidal injection signal, calculates the complex impedance based on the signal parameters, and obtains frequency impedance data pairs;

[0135] Impedance calculation module: Fits rational functions to frequency impedance data and solves for coefficients using the least squares method to obtain wideband impedance curves;

[0136] Quality Inspection Module: Analyzes wideband impedance curves, extracts quality characteristic parameters, and makes quality judgments based on these parameters.

[0137] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0138] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An electrical automation quality detection method, characterized by, The method comprises the following steps: Collecting original voltage signals and original current signals of a power grid public connection point (PCC) and performing fast Fourier transform to obtain a background harmonic frequency set and a background harmonic amplitude set; Generating an initial frequency grid according to a preset target scanning frequency band at a fixed interval; Generating a sparse frequency grid according to the background harmonic frequency set and the initial frequency grid; Calculating an estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid and sorting in ascending order to obtain an injection sequence, inputting the injection sequence into a control power amplifier to obtain a sinusoidal injection signal, and injecting the sinusoidal injection signal into the PCC through a current coupler; While injecting the sinusoidal injection signal, collecting voltage response and total current of the PCC; extracting signal parameters corresponding to the sinusoidal injection signal, calculating complex impedance according to the signal parameters, and obtaining a frequency impedance data pair; Performing rational function fitting according to the frequency impedance data pair, solving the coefficients based on the least square method, and obtaining a wideband impedance curve; Analyzing the wideband impedance curve to obtain quality characteristic parameters, and performing quality determination according to the quality characteristic parameters; The quality characteristic parameters include phase margin, amplitude margin, Nyquist distance and specific frequency band impedance maximum value; The method for obtaining the phase margin comprises the following steps: Traversing the wideband impedance curve to obtain a frequency corresponding to an impedance value of 1 as an amplitude crossover frequency, calculating a difference between an impedance phase of the amplitude crossover frequency and -180° to obtain the phase margin; The method for obtaining the amplitude margin comprises the following steps: Traversing the wideband impedance curve to obtain a frequency corresponding to an impedance phase of -180° as a phase crossover frequency, calculating a negative logarithm of an impedance amplitude corresponding to the phase crossover frequency to obtain the amplitude margin; The method for obtaining the specific frequency band impedance maximum value comprises the following steps: Extracting an impedance curve of a target frequency band in a continuous impedance curve, sampling to obtain a discrete frequency sequence based on a sampling step, substituting each frequency point in the discrete frequency sequence into a continuous impedance function to calculate corresponding complex impedance real part and imaginary part, calculating a corresponding amplitude for each discrete frequency point according to the corresponding complex impedance real part and imaginary part, traversing all NL discrete frequency points, NL being the number of discrete frequency points, calculating the amplitude of each point to generate an impedance amplitude array of the specific frequency band; extracting an impedance amplitude maximum value from the impedance amplitude array as the specific frequency band impedance maximum value.

2. The electrical automation quality detection method of claim 1, wherein, The method for obtaining the background harmonic frequency and the background harmonic amplitude comprises the following steps: Preprocessing the original voltage signal and the original current signal to obtain a preprocessed voltage signal and a preprocessed current signal; Performing N-point fast Fourier transform on the preprocessed voltage signal and the preprocessed current signal to obtain complex voltage frequency domain signals and current frequency domain signals; Calculating voltage amplitude spectrum and current amplitude spectrum according to the voltage frequency domain signals and the current frequency domain signals combined with signal gain and energy dispersion coefficient; Calculating an actual frequency corresponding to an mth index according to a sampling rate and a sampling point number N, and counting all actual frequencies to obtain a frequency axis; Counting and extracting a frequency corresponding to a spectral peak with the largest amplitude in the frequency axis as a fundamental frequency; obtaining a voltage amplitude spectrum value and a current amplitude spectrum value corresponding to the fundamental frequency as a fundamental voltage amplitude and a fundamental current amplitude; All frequency points greater than twice the fundamental frequency are obtained as the to-be-sorted frequencies, the to-be-sorted frequencies meeting the screening condition are sorted in ascending order, and the corresponding frequencies and amplitudes are extracted to obtain a background harmonic frequency set and a corresponding harmonic amplitude set.

3. The method of claim 2, wherein, The screening condition includes condition 1 and condition 2, wherein condition 1: the voltage amplitude of the to-be-sorted frequency is not lower than a harmonic voltage identification threshold or the current amplitude of the candidate frequency is not lower than a harmonic current identification threshold; Condition 2: within an allowable deviation range, the to-be-sorted frequency is n times of the fundamental frequency, and n is an integer not lower than 2.

4. The method of claim 1, wherein, The method for generating a sparse frequency grid comprises: Dividing the target scanning frequency band according to a fixed interval to obtain an initial grid frequency set, and calculating the number of grid points according to the upper limit, the lower limit of the target scanning frequency band and the fixed interval; Setting an interference avoidance bandwidth, for each initial frequency in the initial grid frequency set, if the initial frequency point is within the interference avoidance bandwidth of the background harmonic frequency, the corresponding initial frequency is removed, and a sparse frequency grid is formed based on the remaining frequency points.

5. The method of claim 1, wherein, The method for obtaining an injection sequence comprises: Performing three times spline interpolation on the harmonic spectrum of each candidate frequency in the sparse frequency grid to obtain a background noise amplitude; Calculating an estimated signal-to-noise ratio according to a preset injection signal effective value, a system calibration coefficient and the background noise amplitude, sorting the estimated signal-to-noise ratio in ascending order to obtain an injection sequence.

6. The method of claim 1, wherein, The method for obtaining a background noise amplitude comprises: Defining a three times spline interpolation function meeting an interpolation condition in each segmented interval; Constructing a three times interpolation function expression in each segmented interval; Solving the coefficients of all segmented intervals by establishing an equation group according to the interpolation condition to obtain the interpolation function of all segmented intervals and a continuous interpolation function; Iterating the frequency points of the sparse frequency grid to determine the segmented interval where the candidate frequency is located, extract the corresponding segmented interval and the corresponding interpolation coefficient; Substituting the candidate frequency into the three times interpolation function of the corresponding segmented interval to calculate the background noise amplitude.

7. The method of claim 6, wherein the method further comprises: The interpolation condition comprises: At the endpoints of the segmented interval, the interpolated value is equal to the original amplitude; The first and second derivatives at the connection of the segmented interval are continuous; The natural boundary is adopted, that is, the second derivative value of the two endpoints of the segmented interval is 0.

8. The method of claim 1, wherein, The method for obtaining a frequency impedance data pair comprises: Calculating a sinusoidal injection signal according to the injection sequence, injecting the sinusoidal injection signal into the PCC point through a current coupler, collecting the voltage response and the total current of the PCC point while injecting the sinusoidal injection signal, extracting signal parameters corresponding to the sinusoidal injection signal, the signal parameters including the voltage response effective value, the voltage phase, the total current effective value and the current phase, calculating the impedance phase according to the voltage phase and the current phase, combining the voltage response effective value, the total current effective value and the impedance phase to calculate the complex impedance, and combining the frequency to obtain the frequency impedance data pair.

9. The method of claim 1, wherein, The method for obtaining a Nyquist distance comprises: Calculating a rated reference impedance, combining the continuous impedance function to calculate a normalized complex impedance; The reference-impedance-normalized complex impedance is sampled in a preset frequency range with a discrete sampling frequency step of to generate a discrete frequency sequence; each frequency point in the discrete frequency sequence is substituted into a calculation to obtain a corresponding reference-impedance-normalized complex impedance, a discrete complex impedance array is obtained, and a real part array and an imaginary part array are obtained by splitting Building a complex plane coordinate system with the real axis as the horizontal axis and the imaginary axis as the vertical axis, and drawing a Nyquist curve by combining the real part array and the imaginary part array. For the gth point in the discrete complex impedance array, calculate its Euclidean distance with the (-1, j0) point, traverse all discrete complex impedance points, calculate the Euclidean distance of each point, generate a distance array, and obtain the minimum value in the distance array as the Nyquist distance.

10. The method of claim 1, wherein, The method for quality determination comprises the following steps: If all quality characteristic parameters satisfy that the phase margin is not lower than the phase threshold value, the amplitude margin is not lower than the amplitude threshold value, the Nyquist distance is not lower than the distance threshold value, and the specific frequency band impedance maximum value is not higher than the impedance threshold value, the quality is determined to be qualified, otherwise, the quality is determined to be unqualified, and the corresponding unqualified parameters and deviation values are output.

11. An electrical automation quality detection system, implementing the electrical automation quality detection method of any one of claims 1-10, characterized in that, Comprise: The passive monitoring module: collects original voltage signals and original current signals of a power grid point of common coupling (PCC) and performs fast Fourier transform to obtain a background harmonic frequency set and a background harmonic amplitude set; The grid generation module: presets a target scanning frequency band and generates an initial frequency grid according to a fixed interval; A sparse frequency grid is generated according to the background harmonic frequency set and the initial frequency grid; The intelligent injection module: calculates the estimated signal-to-noise ratio of each candidate frequency in the sparse frequency grid and sorts them in ascending order to obtain an injection sequence, inputs the injection sequence into a control power amplifier to obtain a sinusoidal injection signal, and injects the sinusoidal injection signal into the PCC point through a current coupler; The data acquisition module: acquires voltage responses and total currents of the PCC point while the sinusoidal injection signal is injected; Signal parameters corresponding to the sinusoidal injection signal are extracted, complex impedance is calculated according to the signal parameters, and a frequency impedance data pair is obtained; The impedance calculation module: performs rational function fitting according to the frequency impedance data pair, and solves the coefficients based on the least square method to obtain a wideband impedance curve; The quality detection module: analyzes the wideband impedance curve, extracts quality characteristic parameters, and determines the quality according to the quality characteristic parameters.

Images