Harmonic source dynamic monitoring and responsibility distribution platform and method

Abstract

The invention relates to the technical field of power systems, in particular to a harmonic source dynamic monitoring and responsibility distribution platform and method, and the platform comprises the following steps: a real-time monitoring module, a harmonic source positioning module, a responsibility distribution module, an alarm and event recording module and a data management and analysis module. According to the system, the time domain sampling density of high-frequency harmonics is remarkably improved, the sampling rate of 20 kHz enables the number of sampling points in a single period to be increased to 6.67 points from 1.67 points in the sampling process of 5 kHz, the problem of waveform distortion caused by time domain signal discretization is effectively solved, the actually measured harmonic amplitude error is converged to be within + / -5% from + / -30%, the phase error is reduced to be negligible, and the method has the advantages that the method is simple and easy to implement. The method can automatically detect and correct error frequency components in a frequency spectrum, and when 18kHz harmonic waves are aliasing to 2kHz due to undersampling, the algorithm matches with a harmonic wave prior model through amplitude correlation analysis, and redistributes energy to real frequency points, so that the identification accuracy of harmonic wave times is improved, and a real and reliable data basis is provided for subsequent analysis.

Description

[0001] The present application relates to the technical field of power systems, in particular to a harmonic source dynamic monitoring and responsibility allocation platform and method. BACKGROUND

[0002] The harmonic source dynamic monitoring and responsibility allocation platform is an important tool for monitoring and managing harmonic problems in power systems. It deploys monitoring devices at key nodes in the power grid, such as installing monitoring instruments on busbars and lines, to collect real-time voltage, current, and other data. It uses Fourier transform and other techniques to analyze and process the collected data, accurately obtaining harmonic characteristics such as frequency and amplitude. Meanwhile, it uses advanced algorithms such as piecewise aggregation approximation and shape dynamic time warping to handle data non-synchronization and noise issues, enabling accurate positioning of harmonic sources. Based on correlation analysis and other methods, the responsibility of different harmonic sources is reasonably divided, providing data support for the "reward and punishment scheme" of harmonic control, effectively controlling harmonic pollution, and improving power quality.

[0003] However, in the field of signal processing, the Nyquist criterion requires that the sampling frequency be at least twice the highest frequency component in the signal, so that the original signal can be restored without distortion through sampled data. High-frequency harmonics have high frequencies, and if the sampling frequency does not meet the criterion, the spectrum of high-frequency signals will fold into the low-frequency band, overlapping with the spectrum of low-frequency signals, resulting in spectral aliasing. In this case, the frequency information of high-frequency harmonics in the sampled data will be incorrectly mapped as low-frequency components, making it impossible to accurately distinguish the true harmonic frequency, amplitude, and other characteristics during analysis. For example, if a high-frequency harmonic has a frequency of 500 Hz and the sampling frequency is only 800 Hz (lower than the Nyquist requirement of 1000 Hz), its spectrum will alias to 300 Hz, causing the monitoring system to misidentify the high-frequency harmonic as a low-frequency component, affecting accurate identification of harmonic sources and responsibility allocation, and in severe cases, leading to ineffective harmonic control solutions. SUMMARY

[0004] To address the deficiencies in the prior art, the present application provides a harmonic source dynamic monitoring and responsibility allocation platform and method, thereby solving the technical problems mentioned in the background art.

[0005] To achieve the above objectives, the present application is implemented through the following technical solutions:

[0006] The harmonic source dynamic monitoring and responsibility allocation platform comprises:

[0007] Real-time monitoring module: Real-time acquisition of voltage, current, and other electrical quantity data at each key node of the power system and transmission to the master station system, obtaining harmonic amplitude, frequency, and phase information through data processing, and realizing real-time perception of power system harmonics.

[0008] Harmonic source positioning module: based on the electrical quantity data collected by the real-time monitoring module, taking the harmonic analysis method of Fourier transform as the core, combining the power direction method and the instantaneous reactive power method as auxiliary positioning technology, determining the source position of the harmonic through signal processing and analysis algorithm, providing technical support for harmonic tracing;

[0009] Responsibility allocation module: according to the harmonic source positioning module output harmonic source position results, considering the data asynchronization, power grid operation scene division, data correlation factors, adopting big data analysis idea, constructing harmonic responsibility index and total harmonic responsibility index in each scene through correlation analysis method, quantitatively evaluating and allocating the responsibility of different harmonic sources in harmonic pollution, providing implementation basis for harmonic control "reward and punishment scheme";

[0010] Alarm and event recording module: compare the harmonic responsibility quantitative results output by the responsibility allocation module with the preset harmonic content standard limit value, when the harmonic content exceeds the standard limit value, immediately send alarm signal to the relevant personnel in the form of graphics and sound, and automatically record the detailed information of harmonic abnormal event such as occurrence time, position, harmonic parameter and over standard degree, which is used for subsequent fault analysis and processing;

[0011] Data management and analysis module: store and manage the collected data of real-time monitoring module and the abnormal event information of alarm and event recording module, generate harmonic data report and graphics through data statistics and trend analysis, mine potential harmonic problems and development trend, and provide decision support for planning, operation and maintenance of power system.

[0012] Preferably, for the low sampling rate data point sequence collected by the real-time monitoring module , interpolation processing is adopted to improve the data density, that is:

[0013] N-1 new points are inserted between each two adjacent points, and N is the interpolation multiple;

[0014] The value of the new point is calculated linearly from the adjacent two points:

[0015] The original data points are and When N=4, the inserted points are 1.5, 2 and 2.5, which makes the data density increase by 4 times.

[0016] Preferably, in the harmonic source positioning module, fast Fourier transform is performed on the interpolated time domain signal to obtain the frequency domain spectrum;

[0017] An aliasing frequency mapping model is established:

[0018] The sampling rate is , and the high frequency is Aliased frequency ,

[0019] When , by traversing the frequency points in the spectrum, the frequency satisfies There are components with similar amplitudes at , which are judged as aliasing products.

[0020] Preferably, for the aliasing components identified by the harmonic source positioning module, according to the inverse relationship of the aliasing frequency mapping model:

[0021] , it is stripped from the spectrum and replaced by the real high-frequency component.

[0022] Preferably, in the real-time monitoring module, the sampling rate and the interpolation multiple satisfy a dynamic matching relationship: ,

[0023] wherein, is the interpolation multiple, is the target monitoring frequency, is the original sampling rate, and through the matching relationship, it is ensured that the sampling data can completely retain the time domain details of the target harmonic.

[0024] Preferably, for the oversampling data collected by the real-time monitoring module, FIR low-pass filtering processing is first performed to filter out frequency components higher than , and then interpolation is performed to avoid amplifying high-frequency noise in the interpolation process.

[0025] Preferably, in the harmonic source positioning module, harmonic characteristics are used to assist in identifying aliasing components: the algorithm first determines whether the harmonic order corresponding to each frequency point in the frequency domain spectrum conforms to the rule that "odd orders are dominant" in the power grid, if a certain frequency point corresponds to an even order harmonic and there is no other evidence to support its authenticity, the weight of the frequency component is reduced, and it is determined as a false component generated by aliasing; at the same time, according to the type of equipment in the power grid, the harmonic order that it can generate is preset, if a frequency component that does not conform to the inherent harmonic characteristics of the equipment appears in the spectrum, it is also determined as an aliasing component.

[0026] A method of a harmonic source dynamic monitoring and responsibility allocation platform, comprising the following steps:

[0027] S1: Obtain the original electrical quantity data of the key nodes of the power system - set the monitoring device sampling rate to 4-8 times the target harmonic frequency, and deploy a low-pass filter with a cutoff frequency of 1 / 3 of the sampling rate in the sampling front end to filter out noise and stray signals higher than the cutoff frequency, to obtain interference-free oversampling electrical quantity data;

[0028] S2: Based on the oversampling electrical quantity data obtained in step S1, an interpolation algorithm is selected according to the signal change characteristics: linear interpolation is adopted in the signal gentle area, and cubic spline interpolation is adopted at the harmonic mutation, to obtain a high-density interpolated signal continuous in time domain;

[0029] S3: Based on the high-density interpolated signal obtained in step S2, fast Fourier transform is performed to convert to the frequency domain to generate a frequency spectrum, and the "alias frequency mapping model" is used to traverse the frequency spectrum to find the frequency points with symmetric distribution and similar amplitude, to obtain the frequency domain spectrum of the identified aliasing component;

[0030] S4: Based on the frequency domain spectrum containing aliasing components obtained in step S3, according to the mapping relationship of , the aliasing component is moved from the wrong frequency point to the real high frequency point to complete the spectrum correction, and then inverse Fourier transform is performed on the corrected spectrum to obtain a real high-frequency harmonic time domain signal without aliasing interference;

[0031] S5: Based on the real high-frequency harmonic time domain signal obtained in step S4, the amplitude, frequency and phase information of the harmonic are extracted and input into a harmonic source positioning module, the harmonic source position is determined through Fourier transform analysis and power direction method and instantaneous reactive power method auxiliary positioning, the positioning result is input into a responsibility allocation module, and the responsibility index is constructed by combining data asynchrony, scene division and data correlation factors to quantify the responsibility of each harmonic source. Finally, a harmonic responsibility evaluation report, abnormal alarm information and harmonic trend analysis result are output, which provides data support for harmonic control scheme formulation and power system planning, operation and maintenance decision.

[0032] Compared with the prior art, the beneficial effects are:

[0033] 1. The sampling frequency is increased to 4-8 times the target harmonic frequency, the system significantly improves the time domain sampling density of high-frequency harmonics, avoids the spectrum aliasing phenomenon, and for example, 20kHz sampling rate increases the single cycle sampling point number from 1.67 points at 5kHz sampling to 6.67 points, effectively solves the waveform distortion problem caused by time domain signal discretization, the actual measured harmonic amplitude error converges to within ±5%, and the phase error is reduced to a negligible level;

[0034] 2. The aliasing identification algorithm constructed based on the harmonic frequency mapping rule can automatically detect and correct the error frequency component in the frequency spectrum, when the 18kHz harmonic is aliasing to 2kHz due to undersampling, the algorithm matches the harmonic prior model through amplitude correlation analysis, and reassigns the energy to the real frequency point, so that the harmonic order identification accuracy is improved, and a real and reliable data basis is provided for subsequent analysis;

[0035] 3. This invention, through dynamically adjustable sampling rates (e.g., setting the sampling rate to 4-8 times the target frequency according to the equipment type), can simultaneously monitor low-frequency 50Hz fundamental wave and high-frequency 100kHz switching harmonic components. For example, in new energy power plants, it can accurately analyze the second harmonic at 100Hz and also identify the 10kHz switching harmonic generated by the inverter. Combined with prior knowledge of harmonics, such as fixed harmonic orders in industrial equipment and anti-aliasing filtering, it can reduce aliasing misjudgments caused by transient signals during sudden load changes, such as motor start-up and shutdown. Attached Figure Description

[0036] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0037] Figure 1 This is a flowchart of the harmonic source dynamic monitoring and responsibility allocation platform of the present invention. Detailed Implementation

[0038] Preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present invention can also be implemented in various different forms, and therefore the present invention is not limited to the embodiments described below. In addition, for the purpose of more clearly describing the present invention, parts not connected to the invention will be omitted from the drawings.

[0039] The technical solution in this application embodiment is to solve the problems mentioned in the background art, and the overall idea is as follows:

[0040] Please refer to Figure 1 As shown in the figure, this embodiment introduces a harmonic source dynamic monitoring and responsibility allocation platform, including:

[0041] Real-time monitoring module: By collecting electrical quantity data such as voltage and current in real time at various key nodes of the power system and transmitting them to the main station system, the module obtains the amplitude, frequency and phase information of harmonics through data processing, thereby realizing real-time perception of harmonics in the power system.

[0042] Harmonic source location module: Based on the electrical quantity data collected by the real-time monitoring module, it takes the harmonic analysis method of Fourier transform as the core, and combines the power direction method and instantaneous reactive power method as auxiliary location technologies. Through signal processing and analysis algorithms, it determines the source location of harmonics and provides technical support for harmonic source tracing.

[0043] Responsibility Allocation Module: Based on the harmonic source location results output by the harmonic source location module, and taking into account factors such as data asynchrony, power grid operation scenario division, and data correlation, this module adopts big data analysis ideas and uses correlation analysis methods to construct harmonic responsibility indicators and total harmonic responsibility indicators for each scenario. This allows for the quantitative assessment and allocation of the responsibilities of different harmonic sources in harmonic pollution, providing a basis for the implementation of a "reward and punishment scheme" for harmonic governance.

[0044] Alarm and event logging module: Compares the harmonic responsibility quantification results output by the responsibility allocation module with the preset harmonic content standard limit. When the harmonic content exceeds the standard limit, it immediately issues an alarm signal in the form of graphics and sound to notify relevant personnel. At the same time, it automatically records the time, location, harmonic parameters, degree of exceedance and other detailed information of the harmonic abnormal event for subsequent fault analysis and processing.

[0045] Data Management and Analysis Module: Stores and manages the data collected by the real-time monitoring module and the abnormal event information from the alarm and event recording module. It generates harmonic data reports and graphs through data statistics and trend analysis, and explores potential harmonic problems and development trends to provide decision support for the planning, operation and maintenance of the power system.

[0046] In the real-time monitoring module, the sampling rate is increased to 4-8 times the target harmonic frequency. Oversampling allows more details of the high-frequency signal to be retained in the time domain. A low-pass filter is deployed before sampling, with the cutoff frequency set to 1 / 3 of the sampling rate to filter out noise above the cutoff frequency and prevent interference from out-of-band high-frequency signals. For the already acquired low-sampling rate data point sequence... Insert N-1 new points between every two adjacent points, where N is the interpolation factor;

[0047] The value of the new point is obtained by linear calculation from the two adjacent points: For example, when the original data points are x(0)=1 and x(1)=3, the inserted points are 1.5, 2, and 2.5 when N=4, which increases the data density by 4 times. When there is a sudden change in the signal, cubic spline interpolation is used to avoid waveform distortion caused by linear interpolation by fitting the continuity of the second derivative of three adjacent points.

[0048] Perform a Fast Fourier Transform (FFT) on the interpolated time-domain signal to obtain the frequency-domain spectrum. If aliasing components exist, such as a 3kHz signal aliased at the original 5kHz sampling rate appearing as 2kHz, the 3kHz component will actually exist in the spectrum at a 20kHz sampling rate, while the amplitude of the 2kHz component will be significantly reduced. Establish an aliasing frequency mapping model: Sampling rate... High frequency Frequency after aliasing ,when At that time, by traversing the frequency points in the spectrum, the frequency... satisfy if there are components with similar amplitudes at certain frequencies, it is determined that the 2kHz component is the aliasing component of the 18kHz component, for the identified aliasing component, according to its mapping relationship: , it is stripped from the spectrum and replaced with the real high-frequency component, the amplitude of the 2kHz component is shifted to 18kHz, and then it is restored to the time domain signal through inverse Fourier transform, realizing the real reconstruction of the harmonic, by increasing the sampling frequency to 4-8 times of the target harmonic frequency, the system significantly improves the time domain sampling density of high-frequency harmonics, avoids spectral aliasing, takes 3kHz harmonic monitoring as an example, 20kHz sampling rate increases the number of single-cycle sampling points from 1.67 points at 5kHz sampling to 6.67 points, effectively solving the waveform distortion problem caused by time domain signal discretization, the actual harmonic amplitude error converges from ±30% to within ±5%, and the phase error is reduced to a negligible level.

[0049] Dynamic matching of oversampling rate and interpolation multiple:

[0050] , target monitoring frequency, original sampling rate, assuming the target monitoring frequency is 10kHz and the original sampling rate is 4kHz, then N=[10*2 / 4]=5, i.e. the interpolation multiple is set to 5, so that the equivalent sampling rate reaches 20kHz, first perform FIR low-pass filtering on the oversampled data to filter out frequency components higher than , then perform interpolation to avoid amplifying high-frequency noise. For example, under 20kHz sampling rate, the filter cutoff frequency is set to 8kHz (with a 20% margin), ensuring that harmonics above 10kHz are filtered out and only valid signals are retained, when analyzing the aliased spectrum, the algorithm first determines whether the harmonic order corresponding to each frequency point conforms to the rule that odd orders are dominant, if the harmonic corresponding to the frequency point is even and there is no other evidence to support it, the algorithm reduces the weight of this frequency component and considers it as a pseudo-component generated by aliasing, the algorithm presets the possible harmonic orders according to the device type, if a frequency component appears in the spectrum that does not conform to the characteristics of the device's inherent harmonics, it will also be considered as an aliasing component.

[0051] If the frequency converter of an enterprise usually generates 3, 5, 7 harmonics, when the even harmonic aliasing component appears in the frequency spectrum, it can be judged as interference through the historical data model, the reconstruction result of the odd harmonic is preferentially reserved, the aliasing identification algorithm constructed based on the harmonic frequency mapping rule can automatically detect and correct the error frequency component in the frequency spectrum, when the 18kHz harmonic is aliasing to 2kHz due to under-sampling, the algorithm matches through amplitude correlation analysis and harmonic prior model, re-distributes the energy to the real frequency point, improves the harmonic frequency identification accuracy, provides a real and reliable data basis for subsequent analysis, the sampling rate is dynamically adjustable, such as setting the sampling rate to 4-8 times of the target frequency according to the type of equipment, the low-frequency 50Hz fundamental wave and high-frequency 100kHz switching harmonic component can be monitored at the same time, for example, in a new energy power station, both 2nd harmonic 100Hz and 10kHz switching harmonic generated by the inverter can be accurately analyzed, combined with harmonic prior knowledge such as fixed harmonic number of industrial equipment and anti-aliasing filter, the aliasing misjudgment caused by transient signal can be reduced when the load suddenly changes, such as motor start-stop.

[0052] The method of the harmonic source dynamic monitoring and responsibility allocation platform comprises the following steps:

[0053] S1: set the sampling rate of the monitoring equipment to 4-8 times of the target harmonic frequency, so as to obtain more data points, completely reserve the high-frequency harmonic time domain waveform details, and avoid aliasing, deploy a low-pass filter with a cut-off frequency of 1 / 3 of the sampling rate at the sampling front end, and filter out noise and stray signals higher than the cut-off frequency;

[0054] S2: for the oversampling data, select linear or cubic spline interpolation algorithm according to the signal change characteristics, linear interpolation inserts new points in proportion in the flat signal area, improves the data density and signal time domain continuity, and cubic spline interpolation fits the curve trend of the adjacent three points for the harmonic mutation, so as to ensure the continuity of the second derivative of the inserted point;

[0055] S3: convert the interpolated time domain signal to frequency domain by means of fast Fourier transform, and generate a frequency spectrum, high-frequency harmonics will be displayed in the form of independent peaks, and aliasing components will be displayed as "pseudo-peak values" symmetrically distributed with the real frequency, according to the principle of "aliasing frequency mapping", the sampling rate is , the high-frequency frequency ( ) is aliasing to , and the frequency points symmetrically distributed and with similar amplitudes are searched in the frequency spectrum;

[0056] S4: move the aliasing component amplitude from the error frequency point to the real frequency point, correct the frequency spectrum, and convert it back to the time domain by inverse Fourier transform, so as to obtain the high-frequency harmonic signal without aliasing interference, and provide data support for harmonic source positioning and responsibility allocation.

[0057] Working principle:

[0058] First, set the monitoring device sampling rate to 4-8 times the target harmonic frequency, for example, when monitoring 10 kHz harmonics, set the sampling rate to 40 kHz. Higher sampling rates can obtain more data points in a unit of time, fully retain the waveform details of high-frequency harmonics in the time domain, avoid aliasing of signal waveforms due to sparse sampling points, and at the same time, deploy a low-pass filter in the sampling front end, with a cutoff frequency set to 1 / 3 of the sampling rate, such as 13.3 kHz cutoff frequency for 40 kHz sampling rate. The low-pass filter can filter out noise and stray signals above the cutoff frequency, preventing these signals from entering the sampling link and interfering with the effective frequency band, ensuring the purity of the collected data.

[0059] Second, process the oversampled data according to the signal change characteristics using linear interpolation or cubic spline interpolation algorithm. In the linear interpolation process, for flat signal areas, new data points are inserted between adjacent data points at a fixed ratio, effectively improving data density and enhancing signal continuity in the time domain. For positions where harmonics have mutations, such as peak signals, use the cubic spline interpolation algorithm, which ensures the continuity of the second derivative of the inserted point by fitting the curve trend of the adjacent three points, avoiding waveform distortion problems that may be caused by linear interpolation, accurately restoring the true form of the signal, and fully retaining the characteristic details of high-frequency harmonics.

[0060] Third, convert the interpolated time-domain signal to the frequency domain using fast Fourier transform and generate a frequency spectrum. On this spectrum, high-frequency harmonics will be clearly displayed in the form of independent peaks, while aliasing components will be represented as "pseudo-peak" symmetrically distributed with the true frequency. Using the "aliasing frequency mapping" principle, if the sampling rate is , the high-frequency frequency ( ) aliasing frequency is , traverse the spectrum to find frequency points that are symmetrically distributed and have similar amplitudes to identify aliasing components.

[0061] Fourth, after determining the aliasing components, remove their amplitudes from the wrong frequency points and transfer them to the true frequency points according to the mapping relationship to complete the spectrum correction. Finally, convert the corrected frequency-domain signal back to the time domain through inverse Fourier transform to obtain a signal that removes aliasing interference and truly reflects the characteristics of high-frequency harmonics. This signal can be used for subsequent harmonic source positioning, responsibility allocation, and power system operation state evaluation, providing reliable data support for stable operation and management of the power system.

[0062] It should be pointed out finally that the above embodiments are merely examples for clearly illustrating the present application and are not intended to limit the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. It is unnecessary and impossible to enumerate all the embodiments. The changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A harmonic source dynamic monitoring and responsibility allocation platform, characterized in that, include: Real-time monitoring module: By collecting electrical quantity data such as voltage and current in real time at various key nodes of the power system and transmitting them to the main station system, the module obtains the amplitude, frequency and phase information of harmonics through data processing, thereby realizing real-time perception of harmonics in the power system. Harmonic source location module: Based on the electrical quantity data collected by the real-time monitoring module, it takes the harmonic analysis method of Fourier transform as the core, and combines the power direction method and instantaneous reactive power method as auxiliary location technologies. Through signal processing and analysis algorithms, it determines the source location of harmonics and provides technical support for harmonic source tracing. Responsibility Allocation Module: Based on the harmonic source location results output by the harmonic source location module, and taking into account factors such as data asynchrony, power grid operation scenario division, and data correlation, this module employs big data analysis to construct harmonic responsibility indicators and total harmonic responsibility indicators for each scenario through correlation analysis. This allows for the quantitative assessment and allocation of the responsibilities of different harmonic sources in harmonic pollution, providing a basis for the implementation of a "reward and punishment scheme" for harmonic governance. Alarm and event logging module: Compares the harmonic responsibility quantification results output by the responsibility allocation module with the preset harmonic content standard limit. When the harmonic content exceeds the standard limit, it immediately issues an alarm signal in the form of graphics and sound to notify relevant personnel. At the same time, it automatically records the time, location, harmonic parameters, degree of exceedance and other detailed information of the harmonic abnormal event for subsequent fault analysis and processing. Data Management and Analysis Module: Stores and manages the data collected by the real-time monitoring module and the abnormal event information from the alarm and event recording module. It generates harmonic data reports and graphs through data statistics and trend analysis, and explores potential harmonic problems and development trends to provide decision support for the planning, operation and maintenance of the power system.

2. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 1, characterized in that, It also includes low-sampling-rate data point sequences collected by the real-time monitoring module. To increase data density, interpolation is used, which is: Insert N-1 new points between every two adjacent points, where N is the interpolation factor; The value of the new point is obtained by linear calculation from the two adjacent points: The original data points are and When N=4, the inserted points are 1.5, 2, 2.5, which increases the data density by 4 times.

3. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 1, characterized in that, The harmonic source localization module includes: performing a fast Fourier transform on the interpolated time-domain signal to obtain the frequency domain spectrum; Establish an aliasing frequency mapping model: Sampling rate High frequency Frequency after aliasing , when At that time, by traversing the frequency points in the spectrum, the frequency... satisfy There are components with similar amplitudes at that point, so we can determine... It is a product of folding.

4. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 3, characterized in that, The harmonic source localization module also includes, for the aliasing components identified by the harmonic source localization module, applying the inverse relationship of the aliasing frequency mapping model: This is done by stripping it from the spectrum and replacing it with real high-frequency components.

5. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 1, characterized in that, The real-time monitoring module includes a dynamic matching relationship between the sampling rate and the interpolation factor. ； in, The interpolation factor. For target monitoring frequency, The original sampling rate is used to ensure that the sampling data can fully preserve the time-domain details of the target harmonics.

6. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 1, characterized in that, This also includes: performing FIR low-pass filtering on the oversampled data collected by the real-time monitoring module to filter out data higher than the specified values. The frequency components are then interpolated to avoid amplifying high-frequency noise during the interpolation process.

7. The harmonic source dynamic monitoring and responsibility allocation platform as described in claim 1, characterized in that, The harmonic source localization module also includes using harmonic features to assist in identifying aliasing components: the algorithm first determines whether the harmonic order corresponding to each frequency point in the frequency domain spectrum conforms to the rule of "odd order is dominant" of power grid harmonics. If a frequency point corresponds to an even order harmonic and there is no other evidence to support its authenticity, the weight of that frequency component is reduced and it is determined to be a pseudo component generated by aliasing. At the same time, the harmonic order that the equipment in the power grid may generate is preset according to the type of equipment. If a frequency component that does not conform to the inherent harmonic characteristics of the equipment appears in the spectrum, it is also determined to be an aliasing component.

8. A method for a harmonic source dynamic monitoring and responsibility allocation platform, characterized in that, Includes the following steps: S1: Obtain raw electrical quantity data of key nodes in the power system - by setting the sampling rate of the monitoring device to 4-8 times the target harmonic frequency, and deploying a low-pass filter with a cutoff frequency of 1 / 3 of the sampling rate at the sampling front end, noise and spurious signals higher than the cutoff frequency are filtered out to obtain interference-free oversampled electrical quantity data. S2: Based on the oversampled electrical quantity data obtained in step S1, select the interpolation algorithm according to the signal change characteristics: use linear interpolation in the signal smooth region and cubic spline interpolation at the harmonic abrupt point to obtain a time-domain continuous high-density interpolated signal. S3: Based on the high-density interpolated signal obtained in step S2, perform a fast Fourier transform to convert it to the frequency domain to generate a spectrum. Use the aliasing frequency mapping model to traverse the spectrum, find frequency points with symmetrical distribution and similar amplitudes, and obtain the frequency domain spectrum of the identified aliasing components. S4: Based on the frequency domain spectrum containing aliasing components obtained in step S3, according to... The mapping relationship is used to move the aliased components from the erroneous frequency point to the true high frequency point to complete the spectrum correction. Then, the inverse Fourier transform is performed on the corrected spectrum to obtain the true high frequency harmonic time domain signal without aliasing interference. S5: Based on the real high-frequency harmonic time-domain signal obtained in step S4, the amplitude, frequency, and phase information of the harmonics are extracted and input into the harmonic source location module. The location of the harmonic source is determined by Fourier transform analysis and the power direction method and instantaneous reactive power method. The location results are then input into the responsibility allocation module. Responsibility indicators are constructed by combining data asynchrony, scenario division, and data correlation factors to quantify the responsibility of each harmonic source. Finally, a harmonic responsibility assessment report, abnormal alarm information, and harmonic trend analysis results are output to provide data support for the formulation of harmonic mitigation schemes and the planning, operation, and maintenance decisions of the power system.

Images