CN-121978401-A - Power grid voltage frequency detection method for inhibiting harmonic interference

CN121978401ACN 121978401 ACN121978401 ACN 121978401ACN-121978401-A

Abstract

The invention relates to the technical field of power quality detection of a power grid and discloses a power grid voltage frequency detection method for inhibiting harmonic interference, which has the technical scheme that the method is characterized by comprising the steps of topology coding type signal acquisition, heterogeneous topology cooperative immunity and fundamental wave extraction, light-weight intelligent self-adaptive calibration and intelligent adaptive synchronous interactive output; the invention adopts a full-link cooperative technical scheme of 'topology coding immunity-heterogeneous cooperative extraction-intelligent self-adaptive calibration-intelligent adaptive synchronous interactive output', and achieves the purposes of high precision, high stability, strong self-adaptation and reliable output detection of voltage frequency in a complex power grid environment.

Inventors

YIN ZHENJIE
LI JINBO
WANG JINMEI
YIN JIANMING
LI HAIJIE
ZHU MING

Assignees

宁夏大学

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (10)

1. A method for detecting a voltage frequency of a power grid to suppress harmonic interference, the method comprising the steps of: S1, collecting topology coding type signals, namely amplifying and bandpass filtering conditioning an original voltage signal of a power grid to obtain a conditioned analog signal, carrying out discrete sampling on the conditioned analog signal under the triggering of a synchronous clock to obtain a sampling sequence, mapping the sampling sequence into node attributes of a topological graph, constructing an adjacent matrix, and calculating correlation strength of elements of the adjacent matrix based on amplitude differences and phase differences of corresponding sampling points; S2, heterogeneous topology cooperative immunity and fundamental wave extraction, namely clustering the topological graph obtained in the step S1 based on a density peak clustering algorithm, dividing nodes into a core cluster leading fundamental wave signals and an edge cluster leading harmonic interference, windowing and fast Fourier transforming sampling signals in the core cluster to obtain a frequency spectrum, positioning fundamental wave spectral line peaks through spectral analysis, calculating fundamental wave frequency initial values by adopting an interpolation algorithm, judging the effectiveness of the fundamental wave frequency initial values based on spectral line amplitude threshold values, spectral line purity and spectral flatness, and triggering resampling if the judgment is invalid; S3, constructing a dynamic topological graph reflecting harmonic wave propagation characteristics of a power grid, utilizing a trained lightweight graph neural network model, outputting correction quantity of the fundamental wave frequency initial value according to node characteristics of the dynamic topological graph to obtain a preliminary calibration frequency, adopting a reinforcement learning algorithm to dynamically optimize parameters of the lightweight graph neural network, and carrying out weighted fusion on the preliminary calibration frequency and the historical calibration frequency based on a reward value of the reinforcement learning algorithm to obtain a final calibration frequency; S4, intelligent adaptive synchronous interactive output, namely converting the final calibration frequency into an analog signal, adjusting output impedance through a self-adaptive impedance matching network to match with receiving equipment, generating final calibration frequency data with a time stamp, wherein the synchronous precision of the time stamp is dynamically adjusted according to the distortion degree of the working condition of a power grid, automatically adapting a communication protocol through an intelligent protocol identification module, packaging a data frame containing a basic field and an optional extension field for output, receiving a feedback signal from the receiving equipment, dynamically adjusting a transmission strategy according to the feedback signal, and triggering step S3 to perform secondary calibration if continuous feedback is abnormal.
2. The method for detecting voltage and frequency of power grid capable of suppressing harmonic interference as recited in claim 1, wherein in S1, said constructing an adjacency matrix comprises the elements of The calculation formula of (2) is as follows: ; Wherein, the And For the magnitudes of the i and j sample points, And For the phase of which it is a phase, Is the amplitude attenuation coefficient.
3. The method for detecting the voltage frequency of the power grid for suppressing harmonic interference according to claim 1, wherein in S2, the clustering based on the density peak clustering algorithm comprises: calculating the matching degree of the relevance vector of each sampling point in the current topological graph and the standard fundamental wave signal relevance vector Based on the matching degree Calculating the local density of each sampling point Sum distance According to Judging a clustering center, and dividing sampling points into core clusters according to the difference value of the matching degree between the sampling points and the clustering center Or edge clusters 。
4. The method for detecting voltage and frequency of power grid for suppressing harmonic interference as recited in claim 1, wherein in S2, said interpolation algorithm is adopted to calculate initial value of fundamental frequency The formula of (2) is: ; Wherein, the For the fundamental spectral line peak number, In order to achieve a frequency resolution of the device, For the spectral amplitude values, The direct current component for the FFT operation corresponds to frequency.
5. The method for detecting a voltage frequency of a power grid for suppressing harmonic interference according to claim 1, wherein in S3, the constructing a dynamic topological graph includes: Based on real-time measurement data of a synchronous phasor measurement unit, dynamically updating the edge weight matrix, wherein an updating formula is as follows: ; Wherein, the For the edge weight to be the weight of the edge, In order to update the coefficients of the coefficients, In order to update the period of time, For a real-time equivalent impedance, Is the reference impedance.
6. The method for detecting voltage and frequency of power grid capable of suppressing harmonic interference as recited in claim 1, wherein in S3, the reinforcement learning algorithm is a soft Actor-Critic algorithm, and the reward function is a function of the soft Actor-Critic algorithm The method comprises the following steps: ; Wherein, the In order to calibrate the error of the calibration, In order to calibrate the error threshold value, The penalty coefficients are adjusted for the parameters, The network parameter adjustment amount is used; the weight of the weighted fusion According to the rewards value And (5) dynamically adjusting.
7. The method for detecting a grid voltage frequency while suppressing harmonic interference as recited in claim 1, wherein S3 further comprises calibrating stability determination by calculating variance of the final calibrated frequency within a sliding window When (1) Exceeding a preset stability threshold And automatically shortening the updating period of the dynamic topological graph and the parameter optimization period of the reinforcement learning algorithm.
8. The method for detecting a voltage frequency of a power grid for suppressing harmonic interference according to claim 1, wherein in S4, the adaptive impedance matching network detects the input impedance of the receiving device in real time And adjusting the matching coefficient To achieve impedance matching, wherein Is the reference impedance.
9. The method for detecting a power grid voltage frequency capable of suppressing harmonic interference according to claim 1, wherein in S4, the dynamic adjustment of the synchronization accuracy level is based on a working condition distortion degree coefficient The calculation formula is as follows: Wherein, the method comprises the steps of, For the total rate of distortion of the harmonics, As a factor of the voltage sag level, And Is a weight coefficient; According to Determining a synchronization precision level at different threshold intervals 。
10. The method for detecting a power grid voltage frequency for suppressing harmonic interference according to claim 1, wherein in S4, dynamically adjusting the transmission strategy according to the feedback signal comprises: transmission anomaly identification based on feedback Dynamically adjusting current communication rate And the number of data retransmissions ; When in continuous And when the secondary feedback is abnormal, the step S3 is automatically triggered to perform secondary calibration on the frequency.

Description

Power grid voltage frequency detection method for inhibiting harmonic interference Technical Field The invention relates to the technical field of power grid power quality detection, in particular to a power grid voltage frequency detection method for inhibiting harmonic interference. Background With the large-scale grid connection of new energy and the wide access of power electronic equipment, the running environment of a power grid is increasingly complex, the interference problems of higher harmonics, inter-harmonics, voltage sag and the like are prominent, and strict requirements are put on the accurate and stable detection of the voltage frequency of the power grid. The existing detection method based on fixed parameter filtering, self-adaptive filtering or phase-locked loop improvement generally has the technical bottlenecks that the anti-interference range is limited, the precision and stability are insufficient under dynamic working conditions, the parameters are difficult to self-adaptively adjust, the output mode stiffness depends on manual configuration and the like, and the reliable measurement requirements of complex scenes such as high-harmonic distortion, rapid load fluctuation and the like are difficult to meet. Therefore, the invention provides a power grid voltage frequency detection method for inhibiting harmonic interference, and the technical problems are improved. Disclosure of Invention The embodiment of the disclosure aims at overcoming the defects of the prior art, providing a power grid voltage frequency detection method for restraining harmonic interference, the invention realizes the detection of high precision, high stability, strong self-adaption and high reliability of frequency in complex power grid environment through the full-link innovative design of fusion topology coding, heterogeneous cooperative immunity, intelligent self-adaption calibration and intelligent adaption synchronous interactive output. The technical aim of the invention is realized by the following technical scheme that the power grid voltage frequency detection method for inhibiting harmonic interference comprises the following steps: S1, collecting topology coding type signals, namely amplifying and bandpass filtering conditioning an original voltage signal of a power grid to obtain a conditioned analog signal, carrying out discrete sampling on the conditioned analog signal under the triggering of a synchronous clock to obtain a sampling sequence, mapping the sampling sequence into node attributes of a topological graph, constructing an adjacent matrix, and calculating correlation strength of elements of the adjacent matrix based on amplitude differences and phase differences of corresponding sampling points; S2, heterogeneous topology cooperative immunity and fundamental wave extraction, namely clustering the topological graph obtained in the step S1 based on a density peak clustering algorithm, dividing nodes into a core cluster leading fundamental wave signals and an edge cluster leading harmonic interference, windowing and fast Fourier transforming sampling signals in the core cluster to obtain a frequency spectrum, positioning fundamental wave spectral line peaks through spectral analysis, calculating fundamental wave frequency initial values by adopting an interpolation algorithm, judging the effectiveness of the fundamental wave frequency initial values based on spectral line amplitude threshold values, spectral line purity and spectral flatness, and triggering resampling if the judgment is invalid; S3, constructing a dynamic topological graph reflecting harmonic wave propagation characteristics of a power grid, utilizing a trained lightweight graph neural network model, outputting correction quantity of the fundamental wave frequency initial value according to node characteristics of the dynamic topological graph to obtain a preliminary calibration frequency, adopting a reinforcement learning algorithm to dynamically optimize parameters of the lightweight graph neural network, and carrying out weighted fusion on the preliminary calibration frequency and the historical calibration frequency based on a reward value of the reinforcement learning algorithm to obtain a final calibration frequency; S4, intelligent adaptive synchronous interactive output, namely converting the final calibration frequency into an analog signal, adjusting output impedance through a self-adaptive impedance matching network to match with receiving equipment, generating final calibration frequency data with a time stamp, wherein the synchronous precision of the time stamp is dynamically adjusted according to the distortion degree of the working condition of a power grid, automatically adapting a communication protocol through an intelligent protocol identification module, packaging a data frame containing a basic field and an optional extension field for output, receiving a feedback signal from the receiving equipment, dyna