Intelligent collaborative harmonic suppression method and related equipment

Abstract

The invention discloses an intelligent cooperative harmonic suppression method and related equipment, and the method comprises the steps: obtaining a voltage signal and a current signal, and carrying out the preprocessing, and obtaining the preprocessing data; performing spectrum analysis on the preprocessed data to obtain spectrum data of a plurality of harmonic waves; based on the frequency spectrum data, determining governance weights corresponding to the passive filtering unit, the active filtering unit and the reactor unit respectively; and based on the governance weight, respectively adjusting a capacitance parameter of the passive filter unit, a compensation current parameter of the active filter unit and an impedance parameter of the reactor unit. According to the invention, harmonic wave coverage in a wide frequency range is realized through an intelligent adaptive filtering mode, load fluctuation and power grid frequency deviation are adapted, the harmonic wave treatment effect is more stable, the defects of a traditional harmonic wave treatment technology in coverage range, scene adaptability and volume cost are overcome, and harmonic wave treatment requirements of different scenes can be met.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power supply, in particular to a method for intelligent collaborative harmonic suppression and related equipment. BACKGROUND

[0002] In the fields of industrial production, new energy grid connection, and traffic charging, the wide application of power electronic devices (such as frequency converters, photovoltaic inverters, and charging piles) has promoted energy efficient utilization and industrial upgrading, but also has generated a large number of harmonics (2 to 100 times) due to the nonlinear characteristics of the devices, causing deterioration of power quality. Current harmonic control technologies mainly include passive filtering, active filtering, and reactor technology, but often use single devices or traditional fixed parameter combination schemes for harmonic suppression, which can only effectively suppress a fixed number of frequencies and are difficult to adapt to wide frequency harmonic scenarios. SUMMARY

[0003] To solve the above problems, the embodiments of the present application provide a method for intelligent collaborative harmonic suppression and related equipment, which can intelligently allocate the control weights of passive filter units, active filter units, and reactor units based on current frequency spectrum data, achieving the comprehensive goals of wide frequency coverage, dynamic adaptation, and improved harmonic suppression rate.

[0004] According to a first aspect of the embodiments of the present application, a method for intelligent collaborative harmonic suppression is provided, which includes: obtaining a voltage signal and a current signal, the voltage signal and the current signal being electrical signals including a plurality of harmonics; preprocessing the voltage signal and the current signal to obtain preprocessing data; performing frequency spectrum analysis on the preprocessing data to obtain frequency spectrum data of the plurality of harmonics; determining control weights corresponding to passive filter units, active filter units, and reactor units respectively based on the frequency spectrum data; adjusting the capacitance parameters of the passive filter units, the compensation current parameters of the active filter units, and the impedance parameters of the reactor units respectively based on the control weights corresponding to the passive filter units, the active filter units, and the reactor units respectively.

[0005] Further, the above determination of control weights corresponding to passive filter units, active filter units, and reactor units respectively based on a plurality of frequency spectrum data includes: obtaining a load current change rate based on the preprocessing data; inputting the spectrum data and the load current change rate into a load prediction model to obtain predicted spectrum data of the plurality of harmonics, wherein the load prediction model is a model trained by load current change rates and a plurality of historical spectrum data obtained under different load fluctuation scenarios, and the predicted spectrum data is spectrum data in a preset future time period; Based on the predicted spectrum data, the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit are determined respectively.

[0006] Further, the above-mentioned predicted spectrum data includes a predicted amplitude, and after the spectrum data and the load current change rate are input into the load prediction model to obtain the predicted spectrum data of the plurality of harmonics, the method further comprises: If there is a target harmonic with a predicted amplitude greater than a first preset threshold value among the plurality of harmonics, the predicted amplitude of the target harmonic is replaced with the first preset threshold value, and the first preset threshold value is the maximum value of the historical amplitude of the target harmonic.

[0007] Further, the above-mentioned determination of the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively based on the spectrum data comprises: Obtaining the fundamental frequency of the voltage signal and / or the fundamental frequency of the current signal; According to the fundamental frequency of the voltage signal and / or the fundamental frequency of the current signal, the frequencies of the plurality of harmonics are calculated; According to the frequencies and the spectrum data, the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit are determined respectively.

[0008] Further, after the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit are respectively adjusted based on the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively, the method further comprises: When the fundamental frequency is offset, the capacitance parameter of the passive filter unit is corrected to ensure that the resonance point of the passive filter unit matches the offset fundamental frequency.

[0009] Further, the above-mentioned spectrum data includes distortion rates of the plurality of harmonics, and the determination of the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively based on the spectrum data comprises: The distortion rate of each of the harmonics is calculated to obtain a ratio of the distortion rate of each of the harmonics to the total harmonic distortion rate; According to the ratio of the distortion rate of each of the harmonics to the total harmonic distortion rate, the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit are determined respectively.

[0010] Further, after determining the governance weights of the passive filter unit, the active filter unit and the reactor unit respectively corresponding to the ratio of the distortion rate of each harmonic to the total harmonic distortion rate, the method further comprises: If there is an abnormal harmonic with a distortion rate greater than a second preset threshold among the plurality of harmonics, the governance weight of the passive filter unit is reduced, and the governance weight of the active filter unit is increased until the distortion rate of the abnormal harmonic is less than the second preset threshold.

[0011] According to a second aspect of the embodiments of the present application, an intelligent collaborative harmonic suppression device is provided, comprising: An acquisition module is configured to acquire a voltage signal and a current signal, the voltage signal and the current signal being electrical signals including a plurality of harmonics; A preprocessing module is configured to preprocess the voltage signal and the current signal to obtain preprocessing data; An analysis module is configured to perform frequency spectrum analysis on the preprocessing data to obtain frequency spectrum data of the plurality of harmonics; A processing module is configured to determine governance weights of a passive filter unit, an active filter unit and a reactor unit respectively corresponding to the frequency spectrum data; An execution module is configured to adjust a capacitance parameter of the passive filter unit, a compensation current parameter of the active filter unit and an impedance parameter of the reactor unit respectively based on the governance weights of the passive filter unit, the active filter unit and the reactor unit respectively.

[0012] According to a third aspect of the embodiments of the present application, a computer device is provided, comprising: A computer device comprises a memory and a processor, the memory stores computer readable instructions, and the processor executes the computer readable instructions to implement the steps of the intelligent collaborative harmonic suppression method according to any one of the above embodiments or implementation manners.

[0013] According to a fourth aspect of the embodiments of the present application, a computer readable storage medium is provided, comprising: The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by a processor to implement the steps of the intelligent collaborative harmonic suppression method according to any one of the above embodiments or implementation manners.

[0014] The application has the beneficial effects that: the application obtains the voltage and current signals with harmonics in the power grid, pre-processes the signals, analyzes to obtain multi-frequency spectrum data, then distributes the treatment weights to the passive filter, active filter and reactor unit according to the preset rules, and finally adjusts the related parameters of each unit. This way can realize harmonic coverage in a wide frequency range, adapt to load fluctuation and power grid frequency offset, make the harmonic treatment effect more stable, and also optimize the system size and cost, solve the problems of traditional harmonic treatment technology in coverage range, scene adaptability, size and cost. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings are included to provide a further understanding of the application, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the application and, together with the description, serve to explain the principles of the application. In the drawings: Figure 1 is a step flow diagram of a smart collaborative harmonic suppression method provided by the application; Figure 2 is a structural schematic diagram of a smart collaborative harmonic suppression device provided by the application; Figure 3 is a basic structure block diagram of a computer provided by the application. DETAILED DESCRIPTION

[0016] In order for those skilled in the art to better understand the technical solutions of the application, the application will be described in detail below in combination with the drawings and specific embodiments. Although the drawings and specific embodiments describe exemplary embodiments of the application, it should be understood that the application can be implemented in various forms and should not be limited by the embodiments described herein.

[0017] The "first", "second" and similar words used in the application do not represent any order, quantity or importance, but are only used to distinguish. The "including" and similar words used in the application mean that the elements before the word cover the elements listed after the word, and do not exclude the possibility of also covering other elements. The technical solutions of the application are not limited to the execution order described in the embodiments, and each step in the execution order can be combined, decomposed, and exchanged in order, as long as the logical relationship of the execution content is not affected.

[0018] All terms used herein, including technical terms or scientific terms, have the same meaning as understood by one of ordinary skill in the art to which this application pertains, unless otherwise specifically defined. It should also be understood that terms defined in commonly used dictionaries should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and should not be interpreted in an idealized or overly formal sense unless expressly so defined herein. Technical and scientific terms can not be discussed in detail because they are known to those skilled in the art, but they should be construed as part of the specification where appropriate.

[0019] First, the related terms involved in the application embodiments are explained and described.

[0020] The passive filter unit is a basic harmonic suppression device composed of passive elements such as capacitors, inductors, and resistors. The core provides a low-impedance path for specific frequency harmonics through the "resonance principle", achieving harmonic diversion or consumption. Common forms include single-tuned filters for single frequencies, double-tuned filters covering double frequencies, high-pass filters adapting to high-frequency bands, and composite filters combining multiple forms. It mainly undertakes the governance task of low-frequency, high-amplitude harmonics (such as 5th, 7th), with the advantages of low cost, small loss, and large capacity, becoming the "foundation force" in collaborative governance, which can significantly reduce the economic cost of the overall scheme, and lay the foundation for subsequent accurate compensation of active filtering.

[0021] The active filter unit is a dynamic compensation device relying on power electronic devices (such as IGBT), controllers, and sensors. It works through real-time detection of harmonics, generation of reverse compensation signals, and active intervention to offset the power grid. It needs external power supply for driving. Common forms include parallel type (offset current harmonics, most widely used), series type (offset voltage harmonics), hybrid type combined with passive, and modular type with flexible expansion. It focuses on making up for the limitations of passive filtering, focusing on governing high-frequency, complex frequency harmonics (such as 13 times or more, interharmonics), especially suitable for dynamic scenarios with load fluctuations and changing harmonic frequencies, with a response speed of microseconds, and ensuring that the governance effect meets the standards through accurate compensation.

[0022] The reactor unit is an auxiliary device based on electromagnetic induction principle, which generates inductance through coils to adjust the impedance of the power grid. Its core function is to change the harmonic flow path and amplitude to ensure system stability. Common forms include air-core reactors with good linearity and high-frequency adaptation, core reactors with high inductance and low-frequency adaptation, adjustable reactors with dynamic inductance, and split reactors suitable for multi-branch power grids. It is mainly used to suppress harmonic amplification (to avoid passive filtering and power grid inductance resonance), stabilize filtering effect (to maintain filter unit operating conditions when power grid impedance changes), and prevent high-frequency harmonic propagation by high-frequency high-impedance characteristics, providing protection for the reliable operation of passive and active filter units.

[0023] In the fields of industrial production, new energy grid connection, and traffic charging, the wide application of power electronic devices (such as frequency converters, photovoltaic inverters, and charging piles) has promoted the efficient use of energy and industrial upgrading, but also has generated a large number of harmonics (2 to 100 times) due to the nonlinear characteristics of the devices, causing the deterioration of power quality. Current harmonic control technologies mainly include passive filtering, active filtering, and reactor technology, but often use single devices or traditional fixed parameter combination schemes for harmonic suppression, which can only effectively suppress a few fixed frequencies of harmonics and cannot adapt to wide frequency harmonic scenarios. Passive filters are prone to tuning point deviation, active filters are high in cost and difficult to apply on a large scale, and reactors are large in size and high in iron loss. The overall system has poor adaptability to load fluctuations and grid frequency deviations, making it difficult to maintain stable control performance in complex scenarios. In addition, traditional schemes have the problems of large size and high cost, which cannot meet the collaborative control needs of distributed scenarios (such as new energy and charging piles).

[0024] To solve the above problems, the embodiments of the present application provide an intelligent collaborative harmonic suppression method, which can intelligently allocate the control weights of passive filter units, active filter units, and reactor units based on current frequency spectrum data, achieving the comprehensive goals of wide frequency coverage, dynamic adaptation, and improved harmonic suppression rate. Figure 1 As shown in the figure, Figure 1 is a schematic diagram of the overall steps of the method. The method can include the following S11-S15.

[0025] S11: Obtain a voltage signal and a current signal, the voltage signal and the current signal being electric signals including multiple harmonics.

[0026] It should be noted that the electric signal with harmonics is not limited to a certain frequency band or a certain intensity of harmonics, but covers all types of harmonics that may occur in power grid operation (such as low-frequency and high-frequency harmonics, steady-state and transient harmonics, etc.), as long as the harmonic-containing signal actually exists in the power grid.

[0027] In the selection of the power grid, key positions that can reflect harmonic characteristics (such as power grid inlet end, load access end, bus node, branch node, etc.) are deployed for collection points. Signal collection elements (such as various voltage sensors, current sensors, integrated signal collection modules, etc., such as Hall sensors, Rogowski coil sensors, voltage transformers, and current transformers) that meet the voltage level and current range of the power grid are used. Real-time collection of power grid signals with harmonics ensures that the collected signals can reflect the dynamic changes of power grid harmonics (such as harmonic fluctuations when the load suddenly changes), and the collection frequency and duration are adapted to the operation characteristics of the power grid (such as continuous collection and periodic collection, as long as the harmonic change period is covered).

[0028] S12: preprocessing the voltage signal and the current signal to obtain preprocessing data.

[0029] The collected original signal must contain interference components, such as electromagnetic interference in the signal transmission process, errors of the collection device itself, and transient interference in the power grid. If the original signal containing interference is directly used for frequency spectrum analysis, it will lead to incorrect identification of harmonic frequencies, calculation deviation of harmonic amplitudes, and subsequent weight distribution and parameter adjustment deviating from the target, resulting in a significant decrease in treatment effect. Therefore, the interference components must be removed and the effective harmonic signal must be retained through this step.

[0030] Specifically, interference removal means suitable for power grid signals (such as filtering algorithms, signal denoising techniques, wavelet transform algorithms, Kalman filtering, mean filtering, median filtering, and adaptive filtering) can be used to filter out high-frequency noise, pulse interference, random interference, and other invalid components in the original signal. System errors caused by the collection device (such as sensor accuracy deviation and signal transmission attenuation) are corrected (such as adjusting the signal amplitude through a calibration coefficient and correcting the signal phase shift through phase compensation), ensuring that the signal can accurately reflect the actual size and phase of the power grid harmonics. According to the input requirements of the subsequent frequency spectrum analysis device, the processed signal is converted into an adaptive format (such as converting an analog signal into a digital signal through a signal conversion means or standardizing the format of a digital signal to obtain preprocessing data.

[0031] S13: performing frequency spectrum analysis on the preprocessing data to obtain frequency spectrum data of a plurality of harmonics.

[0032] The key to harmonic treatment is targeted suppression. Different frequencies of harmonics have different hazards to the power grid (such as low-frequency harmonics easily causing transformer overheating and high-frequency harmonics easily interfering with communication signals), and different treatment units (passive, active, and reactor) have different suppression effects on different frequencies of harmonics (such as passive filtering being suitable for low-frequency harmonics and active filtering being suitable for high-frequency harmonics). If it is not possible to identify which frequencies of harmonics and how strong each frequency of harmonics is through frequency spectrum analysis, subsequent treatment weight distribution will become blind distribution. Therefore, frequency spectrum analysis needs to be performed on the cleaned data after preprocessing to extract key parameters existing in the power grid, such as the specific frequency of harmonics, the amplitude, phase, and distortion rate of each frequency of harmonics, and finally form a plurality of frequency spectrum data.

[0033] Specifically, a frequency spectrum analysis algorithm (such as Fourier transform algorithm, wavelet transform algorithm, FFT algorithm, discrete Fourier transform (DFT), short-time Fourier transform (STFT), wavelet packet transform, etc.) that can realize time domain to frequency domain conversion can be used to convert the preprocessed data from time dimension signal to frequency dimension signal. The key parameters (such as harmonic frequency number, amplitude of each frequency harmonic, phase angle, distortion rate ratio relative to fundamental wave, etc.) of each frequency harmonic are identified and extracted from the frequency domain signal, and the fundamental wave signal and the harmonic signal are distinguished (the fundamental wave is the rated frequency signal of the power grid, and the harmonic is the signal of the integer multiple frequency of the fundamental wave). Finally, the extracted harmonic parameters of each frequency are sorted in a standardized format to form multiple frequency spectrum data (such as sorted by frequency from low to high or sorted by amplitude from large to small), ensuring that the data can be directly used for subsequent governance weight allocation.

[0034] S14: Based on the frequency spectrum data, determine the governance weights corresponding to the passive filter unit, the active filter unit, and the reactor unit respectively.

[0035] If only passive filtering is used, high-frequency harmonics cannot be suppressed, if only active filtering is used, the cost is high and it is easy to overload, and if only reactors are used, specific frequency harmonics cannot be suppressed. If no weight is allocated, the three units may simultaneously govern the same harmonic, or no unit governs a certain harmonic, or even mutual interference between units occurs. Therefore, the passive filter unit, the active filter unit, and the reactor unit need to be allocated governance weights in combination with the pre-set mapping relationship.

[0036] It should be noted that the passive filter unit can be composed of passive components and does not include power electronic devices for active control. Typical components include filter capacitors, filter reactors, damping resistors, and switching switches. These components are connected in series or parallel according to capacitance, inductance, and resistance to form single-tuned, double-tuned, or high-pass filter branches, form low-impedance channels for specific frequency harmonics (such as 5th and 7th), and achieve harmonic shunt. The active filter unit takes power electronic devices as the core and includes active control and detection modules. Typical components include main power circuits, DC side energy storage elements, detection circuits, and control units. By real-time detection of harmonic current, the inverter output is controlled to be equal in size and opposite in phase to the harmonic current, and active cancellation of harmonics is achieved. The reactor unit takes inductive elements as the core to assist in stabilizing the impedance of the power grid. Typical components include core reactors, air-core reactors, and adjustment mechanisms, which are usually connected in series at the power grid inlet end or the load side to reduce the propagation and amplification of harmonic current by increasing system impedance.

[0037] The operation of the intelligent collaborative harmonic suppression method of the present application relies on a subject device (such as a harmonic control device), and the passive filter unit, the active filter unit, and the reactor unit are execution components of the subject device. The subject device is a complete system integrating signal acquisition, data processing, decision control, and execution adjustment functions. In addition to the passive filter unit, the active filter unit, and the reactor unit, the subject device can also include a signal acquisition unit, a data processing unit, and a decision control unit. Similarly, the intelligent collaborative harmonic suppression method of the present application also needs to be realized through the hardware actions of the three units. The passive filter unit shunts fixed frequency harmonics, the active filter unit compensates for wideband harmonics, and the reactor unit stabilizes impedance. The three units work collaboratively under the coordination of the subject device according to the weights set by the method, and none of them can be omitted.

[0038] The mapping relationship between harmonic characteristics and governance units is pre-set, such as preferentially allocating low-frequency harmonics to the passive filter unit, preferentially allocating high-frequency harmonics to the active filter unit, and preferentially allocating the reactor unit to assist in stabilizing system impedance to improve overall governance effect. At the same time, resource optimization rules such as cost priority, efficiency priority, and device load balancing priority can be added to the mapping relationship.

[0039] Based on key parameters (such as the amplitude ratio of each frequency harmonic and the distortion rate) in multiple frequency spectrum data, combined with pre-set rules, the governance weights of the three units are calculated through decision algorithms (such as logical judgment algorithms, weight distribution models, optimization algorithms, etc.). For example, high-amplitude low-frequency harmonics correspond to high passive weights, and high-amplitude high-frequency harmonics correspond to high active weights.

[0040] The calculated weights are preliminarily verified, such as whether the weights meet the maximum load of the device and whether they can cover all frequency harmonics. If there is a conflict (such as excessive passive weight leading to device overload), fine-tuning is performed to finally determine reasonable governance weights.

[0041] S15: Based on the governance weights corresponding to the passive filter unit, the active filter unit, and the reactor unit respectively, the capacitance parameters of the passive filter unit, the compensation current parameters of the active filter unit, and the impedance parameters of the reactor unit are adjusted respectively.

[0042] The passive filter unit has a significant advantage in dealing with low-frequency harmonics (2-11 times). Based on the LC resonance principle, it can provide a low-impedance path for specific low-frequency harmonics, achieving efficient shunt and absorption. Passive filtering has the characteristics of low cost, large capacity, small loss, high reliability, etc., and is especially suitable for dealing with low-frequency harmonics with high amplitude and fixed frequency (such as 5 times and 7 times). It is a basic unit in the harmonic control system. It should be noted that relying entirely on passive filtering (i.e. weight 100%) will introduce unacceptable system risks, and retaining a certain proportion (such as 30%) by active filtering and reactor is to build a robust, adaptive and safe system.

[0043] The active filter unit is more skilled at dealing with high-frequency harmonics (13-100 times) and interharmonics. It dynamically cancels harmonics by real-time detection and injection of reverse compensation current, with the characteristics of wide frequency band, high precision and fast response (microsecond level). Active filtering is not affected by system impedance and can adapt to load changes and harmonic frequency fluctuations. It is especially suitable for dealing with high-frequency, complex and changing harmonic components.

[0044] The reactor unit, as an auxiliary control unit, mainly suppresses harmonic amplification and high-frequency harmonic propagation by adjusting system impedance, ensuring the stable operation of passive and active filter units. It has high impedance characteristics in the high-frequency band, which can effectively prevent the spread of high-frequency harmonics and avoid resonance between passive filter units and the grid, improving the reliability and stability of the overall system.

[0045] Specifically, according to the three unit control weights determined in step S04, the key parameters of each unit are adjusted. The passive filter unit adjusts the capacitance parameters (such as parameters affecting the resonance point), the active filter unit adjusts the compensation current parameters (such as parameters affecting the compensation current / voltage), and the reactor unit adjusts the impedance parameters (such as parameters affecting the system impedance), finally allowing the three units to play an actual harmonic suppression role according to the allocated weights. The adjustment here is on-demand adjustment, not limited to fixed adjustment methods.

[0046] According to the passive filter unit control weight, adjust the capacitance parameters affecting the filtering effect (such as adjust the resonance point related parameters: capacitive reactance, inductive reactance parameters, or adjust the parameters of controllable elements (such as capacitance group capacitance, inductance group inductance), to ensure that the passive filter unit can suppress the target frequency harmonics according to the weight (such as high weight to strengthen the matching degree of the resonance point and the target harmonics).

[0047] According to the active filter unit control weight, adjust the compensation current parameters affecting the compensation effect (such as adjust the amplitude, phase, frequency parameters of compensation current / voltage, or adjust the driving parameters of switching devices, not limited to IGBT driving signals), to ensure that the active filter unit can output reverse compensation signals according to the weight to cancel residual harmonics.

[0048] According to the governing weight of the reactor unit, the impedance parameters affecting the impedance characteristics (such as adjusting the magnetic core characteristic parameters, the coil current parameters, the inductance value parameters, not limited to the auxiliary coil current control) are adjusted to ensure that the reactor can stabilize the system impedance according to the weight, and the passive and active governing effects are assisted to improve (such as the weight is high, the impedance adjustment ability is enhanced, and the high-frequency harmonic diffusion is suppressed).

[0049] The intelligent collaborative harmonic suppression method provided by the embodiment of the application can realize wide frequency range harmonic coverage, adapt to load fluctuation and power grid frequency offset, have different weights for different harmonics, make the harmonic suppression effect more stable, optimize the system size and cost, solve the deficiencies of traditional harmonic suppression technology in coverage range, scene adaptability, size and cost, and meet the harmonic suppression needs of different scenes such as industry, new energy and charging piles.

[0050] In some optional implementation manners of the embodiment, after the capacitance parameter of the passive filter unit, the compensation current parameter of the active filter unit and the impedance parameter of the reactor unit are adjusted based on the corresponding governing weights of the passive filter unit, the active filter unit and the reactor unit, the above method further includes the following S21-S25.

[0051] S21: obtaining a load current change rate based on the preprocessed data.

[0052] It should be noted that the load fluctuation is highly random, and the traditional real-time response mode has a delay of 5-10 ms. When the load mutation causes the harmonic amplitude to increase sharply, the real-time adjustment may lag, causing the short-term distortion rate to exceed the standard. The load current change rate refers to the change amount of the load current per unit time, and is used to quantify the severity of the load fluctuation.

[0053] Specifically, a sliding window algorithm (the window length can be configured to 1-10 ms, and is suitable for the power grid fluctuation period) can be used to calculate the load current change rate by the formula , wherein is the load current at t, +Δ t is the load current at t.

[0054] ​S22: inputting the spectrum data and the load current change rate into a load prediction model to obtain predicted spectrum data of the plurality of harmonics, wherein the load prediction model is a model trained by load current change rates and a plurality of historical spectrum data obtained under different load fluctuation scenarios, and the predicted spectrum data is spectrum data in a preset future time period.

[0055] It should be noted that there is a strong correlation between load fluctuation and harmonic change (for example, a 30% increase in load current may cause a 25% increase in the amplitude of the 5th harmonic), but traditional governance cannot predict the harmonic changes caused by this correlation. Through the load prediction model, the frequency and amplitude characteristics of future harmonics can be grasped in advance, providing a target basis for subsequent pre-adjustment and avoiding the delay risk of real-time adjustment.

[0056] Specifically, the load prediction model refers to a prediction model trained based on historical data, which is used to predict future harmonic trends according to current spectrum data, prediction weights, and load change rates. The load prediction model can select a time series prediction model (such as ARIMA, LSTM neural network), or an input-output response model (such as a transfer function obtained through system identification). The input of the model usually includes historical and current multi-frequency spectrum data sequences and load current change rate sequences, and the output is predicted spectrum data (amplitude, phase, etc.) of each frequency in a future period of time. For example, the core expression of the load prediction model can be represented as: wherein, is the predicted spectrum data at the time t+T, is the current frequency spectrum data, is the load current change rate, is a parameter obtained by training the model, is a model mapping relationship (such as an ARIMA, LSTM, etc. time series prediction or system identification model).

[0057] S23: determining governance weights corresponding to the passive filter unit, the active filter unit, and the reactor unit, respectively, based on the predicted spectrum data.

[0058] It should be noted that if no prediction weights are assigned to future harmonics, each governance unit may still operate according to the current weights, which cannot adapt to future harmonic changes (for example, if future high-frequency harmonics increase but the active weight does not increase, resulting in governance failure). Therefore, it is necessary to assign prediction weights in advance to clearly define the future division of labor of each unit and ensure the pertinence of collaborative governance. The predicted spectrum data refers to the amplitude, phase, and other parameters of each frequency harmonic in a future period of time output by the load prediction model.

[0059] ​Specifically, based on the prediction results and the budget rules, the governance weight of each governance unit is calculated in advance, for example, according to the frequency of each frequency harmonic (such as 5 times, 13 times, etc., derived from the fundamental frequency), the amplitude of each frequency harmonic (reflecting the strength of the frequency harmonic), and the scene type, the governance weight can be obtained through the following formula: Governance weight (unit X) = f (predicted harmonic frequency, predicted harmonic amplitude, scene type, unit characteristic constraint). Wherein f is the mapping function corresponding to the preset rule, for example, f (passive) > f (active) in low frequency scene, and vice versa in high frequency scene.

[0060] For example, in the cooperative governance of a cluster of 10 electric vehicle charging piles (10 piles), a cluster composed of 10 60kW direct current charging piles (380V power grid, fundamental frequency 50Hz) is executed.

[0061] The high-frequency current sensor (sampling rate 20kHz) at the inlet end of each charging pile collects real-time current, and the pre-processing module calculates the load current change rate through the sliding window algorithm. The current of a charging pile rises from 40A to 60A in 5ms, with a change rate of 4A / ms. The amplitudes of 5-25 harmonics in this period (5 times 25A, 11 times 18A, 13 times 12A) are extracted synchronously, and finally the load current change rate and harmonic amplitude are obtained.

[0062] The load prediction model is trained based on 100,000+ charging pile load scenarios (including empty load, half load, full load and sudden change scenarios). After inputting "5 times 25A + 11 times 18A + 13 times 12A + 4A / ms change rate", it outputs multiple predicted frequency spectrum data for the next 10ms: 5 times 28A, 11 times 20A, 13 times 14A.

[0063] Combined with the preset rules (low frequency 5-11 times passive, high frequency 13-25 times active), the weights are adjusted according to the prediction data. The passive prediction weight of 5 times harmonic is increased from the base 80% to 82% (more harmonics need to be absorbed), the passive weight of 11 times is increased from 70% to 72%, the active weight of 13 times is increased from 65% to 68%, and the weight of the reactor is maintained at 10% (auxiliary stable impedance).

[0064] According to the prediction weight, the specific parameters are calculated. The 5 times harmonic needs to be passively absorbed 28A x 82% ≈ 23A, and the controllable capacitor value needs to be pre-adjusted from 20μF to 20.5μF (predicted capacitor parameter) through the LC resonance formula. The active needs to compensate 28A x (1-82%) ≈ 5A (predicted compensation current parameter), and the switching frequency of SiC device is maintained at 20kHz; the reactor needs to adjust the impedance from 80Ω to 85Ω (predicted impedance parameter) to suppress the spread of high frequency harmonics.

[0065] Each charging pile controller receives the pre-adjustment instruction, completes parameter setting 10 ms in advance, the passive filter unit adjusts the controllable capacitor group (the capacitance value is smoothly transitioned from 20 mu F to 20.5 mu F) through the PID algorithm, the active unit pre-generates the driving signal of 5A reverse compensation current, and the reactor adjusts the impedance to 85 omega through the auxiliary coil current.

[0066] Through early prediction and pre-adjustment, the application can change the governance action from passive response to active prediction, avoid governance failure caused by delay, and effectively avoid possible short-term over-standard problems in real-time adjustment. At the same time, through high-frequency synchronous acquisition and multi-node coverage, the time consistency and spatial integrity of the data are ensured, and the error rate of the overall governance is reduced.

[0067] In some optional implementations of the embodiment, the prediction spectrum data includes a predicted amplitude. After the spectrum data and the load current change rate are input into the load prediction model to obtain the prediction spectrum data of the plurality of harmonics, the method further includes: if there is a target harmonic with a predicted amplitude greater than a first preset threshold in the plurality of harmonics, replacing the predicted amplitude of the target harmonic with the first preset threshold, and the first preset threshold is the maximum value of the historical amplitude of the target harmonic.

[0068] It should be noted that in the process of prediction, sudden interference (such as transient fluctuation of grid voltage, impact of charging gun plugging) may cause abnormal prediction data. Therefore, it is necessary to avoid governance failure caused by abnormal prediction by statistical analysis of historical data. The first preset threshold is an upper limit value of the harmonic amplitude of each frequency set according to the historical data, which is used to judge whether the prediction result is abnormal. The historical prediction spectrum database is a database for storing historical harmonic prediction and actual data, which is used for data backtracking and correction.

[0069] Specifically, before the governance weight, the prediction result is reasonably checked. A first preset threshold is set for each frequency harmonic, which is usually based on the maximum value of the historical amplitude or the upper limit of a certain confidence interval obtained by long-term statistics. If the predicted amplitude of a certain frequency harmonic exceeds its historical maximum value (i.e. the first preset threshold), it is considered as abnormal prediction, and the historical prediction data under similar working conditions will be automatically retrieved from the historical database to replace the current abnormal value, so as to ensure the accuracy of weight distribution and the robustness of the system.

[0070] For example, in the cooperative management of the above-mentioned electric vehicle charging pile cluster, a first preset threshold is set, the frequency spectrum data of the charging pile cluster within one month is counted, the maximum value of each frequency history amplitude is determined as 5 times 28A, 11 times 22A, and 13 times 16A, and these values are set as the first preset threshold of each target frequency. Due to the plug-in interference of the charging gun, the load prediction model outputs a 5th harmonic prediction amplitude of 30A (exceeding the threshold of 28A), 11 times of 19A (not exceeding 22A), and 13 times of 15A (not exceeding 16A).

[0071] By querying the historical prediction spectrum database, the scenario closest to the current load change rate (3A / ms) is found (1 week ago, the same change rate, 5th harmonic actual amplitude 27A, prediction data 27.5A), and the 5th historical prediction data 27.5A of this scenario is taken as the current corrected prediction data. Based on the correction data, the weight is allocated, the 5th harmonic is allocated with a weight of 27.5A, the passive 80% (22A needs to be absorbed), and the active 20% (5.5A needs to be compensated), avoiding the passive overload (24A>22A maximum capacity) caused by the allocation of 30A. Finally, the subsequent pre-adjustment is performed, the passive capacitor capacity value (20.2μF) and the active compensation current (5.5A) are calculated according to 27.5A, and the pre-adjustment is completed.

[0072] The present application sets the preset threshold and the historical prediction spectrum database, avoids the failure caused by abnormal prediction, ensures the accuracy of weight allocation, and improves the robustness of prediction.

[0073] In some optional implementations of the present embodiment, the above-mentioned determining the management weight corresponding to the passive filter unit, the active filter unit and the reactor unit based on the frequency spectrum data comprises: obtaining the fundamental frequency of the voltage signal and / or the fundamental frequency of the current signal; calculating the frequency of a plurality of harmonics according to the fundamental frequency of the voltage signal and / or the fundamental frequency of the current signal; and determining the management weight corresponding to the passive filter unit, the active filter unit and the reactor unit based on the frequency and the frequency spectrum data.

[0074] It should be noted that the essence of harmonics in the power grid is a sine wave with an integer multiple of the fundamental frequency, and its frequency (such as 5 times, 7 times) must be defined with reference to the fundamental frequency (such as 250 Hz for the 5th harmonic when the fundamental frequency is 50 Hz). The characteristics of different frequency harmonics are significantly different, and low-frequency harmonics (such as 5 times, 7 times) are more suitable for passive filter unit LC resonance circuit governance, and high-frequency harmonics (such as 13 times, 17 times) need to rely on the wideband compensation capability of active filter unit, and the impedance adjustment effect of the reactor unit is also directly related to the harmonic frequency (higher impedance at high frequency). Based on the fundamental frequency to determine the harmonic frequency, avoid incorrect frequency determination, and mismatch between governance unit and harmonic characteristics. Therefore, the frequency is determined based on the fundamental frequency, and the weight is allocated based on the frequency spectrum data, which is the basis for ensuring the adaptability of the governance unit and improving the harmonic suppression efficiency.

[0075] Specifically, the fundamental component is separated from the voltage signal and / or current signal by frequency spectrum analysis technology (such as Fourier transform), and the frequency value is extracted. As the dominant signal of the power grid, the fundamental wave is the sine wave component with the highest energy and the lowest frequency in the frequency spectrum (such as 50 Hz or 60 Hz for industrial power grid), and the fundamental frequency can be determined by identifying the frequency corresponding to the energy peak in the frequency spectrum.

[0076] The frequency of any harmonic strictly follows the integer multiple relationship with the fundamental frequency (such as 100 Hz for the 2nd harmonic and 150 Hz for the 3rd harmonic when the fundamental frequency is 50 Hz), and the harmonic frequency can be directly converted to the frequency through this integer multiple relationship, and the relationship with the fundamental frequency is clear. The harmonic frequency is calculated according to the physical definition of the harmonic, which is an integer multiple (n = 2, 3, 4...) of the fundamental frequency (f1). The frequency of each harmonic is calculated by the formula "harmonic frequency n = harmonic frequency f n / fundamental frequency f1".

[0077] Low-frequency harmonics (usually n≤11) are stable in frequency, have large amplitude, and are suitable for the LC resonance shunt characteristics of passive filter units, so they are allocated a higher weight. High-frequency harmonics (usually n≥13) have a wide frequency range and large amplitude fluctuations, and are suitable for the precise compensation characteristics of active filter units, so they are allocated a higher weight. The impedance of the reactor unit is adjusted according to the harmonic frequency (higher impedance at high frequency), and an auxiliary weight is allocated to stabilize the system, and finally a weight ratio is formed by the cooperation of the three. The governance weight is determined based on the calculated harmonic frequency and the harmonic amplitude, distortion rate, etc. in the frequency spectrum data, and the weight is allocated according to the frequency adaptation characteristics of each governance unit.

[0078] The application extracts the fundamental frequency from the voltage signal and / or the current signal first, then calculates the harmonic frequencies according to the integer multiple relationship between the fundamental frequency and the harmonics, and finally combines the frequencies with the weight of the spectral data (such as harmonic amplitude, distortion rate). Since the fundamental frequency is the core reference for defining the harmonic frequency, this can avoid the mismatch between the treatment unit and the harmonic characteristics caused by frequency misjudgment, ensure the accuracy of harmonic suppression, and fully exert the characteristics of each unit to adapt to different frequency harmonics to optimize resource allocation and adapt to complex operating scenarios such as power grid fundamental frequency offset.

[0079] In some optional implementations of the embodiment, after the passive filter unit, the active filter unit and the reactor unit are respectively adjusted according to the corresponding treatment weights, the method further comprises: when the fundamental frequency is offset, the capacitance parameter of the passive filter unit is corrected to ensure that the resonance point of the passive filter unit matches the offset fundamental frequency.

[0080] Specifically, when the basic frequency is offset, the new actual frequency of the target harmonic is calculated According to the resonance frequency formula of the passive filter: In the case of known target frequency (i.e. ) and capacitance (or inductance ), the new capacitance value (or the new inductance value ) to be adjusted is solved. The formula is or By adjusting the parameters of elements such as capacitance and inductance, the resonance point of the passive filter unit is always matched with the actual frequency of the corresponding frequency harmonic (changed due to fundamental frequency offset), and the treatment efficiency of the passive filter is maintained.

[0081] For example, in the cooperative treatment of the above electric vehicle charging pile cluster, the frequency of the fundamental wave is monitored in real time, and the voltage sensor (accuracy 0.2 level) of each charging pile collects the fundamental frequency 100 times per second. When the fundamental frequency of a certain device decreases from 50Hz to 49.2Hz, a correction instruction is triggered to calculate the offset harmonic frequency. The 5th harmonic frequency = 49.2Hz x 5 = 246Hz, and the 11th harmonic frequency = 49.2Hz x 11 = 541.2Hz. For the 5th harmonic, the passive original resonance point is 250Hz (corresponding to a capacitance of 20μF and an inductance of 10mH), and the capacitance value corresponding to 246Hz is calculated according to the LC resonance formula , and the capacitance value is substituted into =246Hz, L=10mH, and the capacitance value is solved as =20.6μF (capacitance parameter is +0.6μF).

[0082] Finally, real-time correction is performed. The thyristor switch of the controllable capacitor bank is driven by the PID algorithm (a control algorithm that combines proportional, integral and derivative elements) to smoothly adjust the capacitance value from 20μF to 20.6μF (response time 8ms). At the same time, the 5th harmonic current after passive filtering is monitored and reduced from 12A (residual) before correction to 5A (residual).

[0083] This application maintains passive filtering efficiency and avoids harmonic amplification by real-time correction of the capacitance parameters of the passive filter unit, thus preventing a mismatch between the resonant point and the harmonics. Passive correction also reduces system energy consumption.

[0084] In some optional implementations of this embodiment, the aforementioned spectrum data includes the distortion rates of multiple harmonics. The step of determining the mitigation weights corresponding to the passive filter unit, active filter unit, and reactor unit based on the spectrum data includes: calculating the ratio of the distortion rate of each harmonic to the total harmonic distortion rate; and determining the mitigation weights corresponding to the passive filter unit, active filter unit, and reactor unit based on the ratio of the distortion rate of each harmonic to the total harmonic distortion rate.

[0085] It should be noted that harmonic distortion rate refers to the ratio of the root mean square value of the harmonic components to the root mean square value of the fundamental component, usually expressed as a percentage. It is a core indicator for measuring the impact of harmonics on the power quality of the power grid. This includes the single harmonic distortion rate (...). ) and total harmonic distortion (THD). Among them, the single harmonic distortion (THD) The distortion rate of the nth harmonic is given by the formula: in, The voltage amplitude of the nth harmonic is... This represents the amplitude of the fundamental voltage.

[0086] Total Harmonic Distortion (THD) refers to the total distortion of all harmonics, and the formula is: The proportion of harmonic distortion rate refers to the ratio of the distortion rate of a specific harmonic or frequency band to the total harmonic distortion rate. It is used to quantify the contribution of different harmonics to the total distortion. For example, the 5th harmonic has a harmonic voltage amplitude of 0.8, and the fundamental voltage amplitude is 10. This can be obtained using the formula for the single harmonic distortion rate. 8.0, the total harmonic distortion (THD) is 20.0%, then the 5th harmonic distortion accounts for = / THD×100%= =40%.

[0087] Specifically, from the spectral data of multiple frequencies obtained through spectral analysis, the single-order distortion rate of each harmonic is calculated. The total harmonic distortion (THD) is measured to ensure coverage of harmonics from the 2nd to the 100th order across the entire frequency band (data accuracy ≤ ±0.5%). For each harmonic, the percentage of harmonic distortion is calculated using the formula above: "Harmonic distortion percentage = The percentage of harmonics is calculated using " / THD×100%", and the top 5-10 harmonics with the highest percentages are selected (the cumulative percentage is usually ≥80%, which are the core targets for governance).

[0088] For low-frequency harmonics with a high distortion rate (e.g., 5th and 7th harmonics, accounting for ≥20%), a higher weight (e.g., 60%-80%) is allocated due to their compatibility with the LC resonance characteristics of passive filter units, leveraging the advantages of passive, low-cost, and high-capacity filtering for priority mitigation. For high-frequency harmonics with a high distortion rate (e.g., 13th and 17th harmonics, accounting for ≥10%), a higher weight (e.g., 50%-70%) is allocated to active filter units because they require wideband precise compensation. For minor harmonics accounting for ≤5% of the total distortion rate: these are assisted in suppression by reactor units through impedance adjustment, allocated a basic weight of 5%-10%.

[0089] Ultimately, ensure that the total weight of each unit is 100% and does not exceed the hardware capability limit (e.g., the weight of a passive single branch should not exceed 30% to avoid resonance risk). Recalculate the distortion rate percentage every 10-50ms (configurable) and update the weight allocation synchronously to adapt to dynamic changes in harmonics (e.g., a sudden increase in the distortion rate percentage of a certain frequency due to load switching).

[0090] For example, in an industrial power grid scenario, by calculating the single harmonic distortion rate (... The total harmonic distortion (THD) is used to determine the percentage of harmonic distortion. The distortion rate of the 5th harmonic frequency accounts for 40%, the distortion rate of the 7th harmonic frequency accounts for 30%, the distortion rate of the 13th harmonic frequency accounts for 15%, the distortion rate of the 17th harmonic frequency accounts for 10%, and the distortion rate of other harmonic frequencies accounts for 5%.

[0091] Harmonics with a cumulative percentage ≥80% were selected based on the distortion rate percentage. The 5th (40%) + 7th (30%) + 13th (15%) = 85% were the core targets for remediation. The 17th (10%) and others (5%) were secondary targets for remediation.

[0092] The governance weight is allocated in combination with the unit characteristics. Among them, for 5th and 7th harmonics (low frequency, cumulative proportion of 70%). 5th and 7th are typical low frequency harmonics, and the LC loop of the passive filter unit can achieve efficient shunt through resonance (such as 5th harmonic corresponding to resonance frequency 250Hz), and the cost is only 1 / 3 of the active filter. The passive filter unit undertakes the main governance task, and allocates 60% of the total weight (of which 5th accounts for 40% x 60% / 70% ≈ 34%, and 7th accounts for 30% x 60% / 70% ≈ 26%).

[0093] Among them, for 13th harmonic (high frequency, 15%). 13th harmonic frequency is 650Hz, and passive filtering is difficult to accurately resonate (easily affected by power grid frequency deviation), while active filter unit can accurately output 650Hz reverse compensation current through power electronic devices. The active filter unit undertakes the main governance task and allocates 25% of the total weight.

[0094] Among them, for 17th and other harmonics (15%). 17th and above harmonics have low amplitude but high frequency, which is easy to amplify through grid impedance, and the reactor unit can increase the system high-frequency impedance to suppress its propagation. The reactor unit undertakes the auxiliary governance task and allocates 15% of the total weight (mainly used to stabilize the high-frequency impedance).

[0095] The application improves the governance efficiency by targeting the harmonics with the largest total distortion contribution. At the same time, it avoids low-hazard harmonics from occupying too many resources and reduces reactive power loss.

[0096] In some optional implementations of the embodiment, after the above determining the governance weight of the passive filter unit, the active filter unit and the reactor unit corresponding to each of the harmonics according to the ratio of the distortion rate of each of the harmonics to the total harmonic distortion rate, the method further comprises: If there is an abnormal harmonic with a distortion rate greater than a second preset threshold among the plurality of harmonics, the governance weight of the passive filter unit is reduced, and the governance weight of the active filter unit is increased until the distortion rate of the abnormal harmonic is less than the second preset threshold.

[0097] It should be noted that during the adjustment of the parameters, the adjusted distortion rate may exceed the standard due to element parameter drift (such as passive capacitor value aging attenuation of 2%) or sudden changes in load. If not dynamically corrected, the total distortion rate will increase, which does not meet the grid standard. Therefore, the correction strategy of reducing passive and increasing active can quickly pull back the distortion rate and avoid long-term over-standard.

[0098] Specifically, the distortion rate of each frequency harmonic in the power grid is monitored in real time, and the monitored value is compared with a second preset threshold (set according to international / national standards such as IEC or self-defined standards). Once the distortion rate of one or more target frequency harmonics is found to be out of standard, the system automatically triggers the weight adjustment strategy. For example, a typical strategy is to reduce the weight of the passive filter unit for that harmonic, while correspondingly increasing the weight of the active filter unit. This strategy is based on the faster and more accurate characteristics of active filtering, which can quickly make up for the decline in treatment effect caused by passive filter element aging, load changes or other unknown factors.

[0099] For example, in the cooperative treatment of the above-mentioned electric vehicle charging pile cluster, a second preset threshold is set. For example, the second preset threshold can be set to ≤2.8%. The distortion rate is monitored in real time by the sensors of each device, and it is found that the 11th harmonic distortion rate of device 7 is 3.2% (exceeding 2.8%). After investigation, it is found that the passive capacitance value of device 7 has decreased from 20 μF to 19.6 μF (decreased by 2%) due to aging, resulting in a decrease in 11th harmonic absorption current from 10.5 A to 9 A, and an increase in residual current from 4.5 A to 11 A, which cannot be covered by the active compensation of 4.5 A. The master edge node issues an instruction to adjust the 11th harmonic weight of device 7 to “passive 60%, active 35%, and reactor 5%” (original 70%, 25%, and 5%). Finally, the parameters are adjusted again, and the capacitance value of the passive filter unit is adjusted to 19.8 μF (absorbing 20 A x 60% = 12 A), and the compensation current of the active filter unit is increased from 4.5 A to 20 A x 35% = 7 A.

[0100] The present application monitors the distortion rate of multiple frequency harmonics, and when the distortion rate exceeds the standard, it is timely corrected to avoid excessive duration and quickly respond to the risk of exceeding the standard. This reduces the time of exceeding the standard and also protects the hardware components, extends the life of the components, and reduces the frequency of replacing the components.

[0101] Further referring to Figure 2 , as an implementation of the method shown in Figure 1 , the present application provides an embodiment of an intelligent cooperative harmonic suppression device. The device embodiment corresponds to the method embodiment shown in Figure 1 , and the device can be applied to various electronic devices.

[0102] As shown in Figure 2 , the intelligent cooperative harmonic suppression device 200 described in the present embodiment includes an acquisition module 201, a preprocessing module 202, an analysis module 202, a processing module 204, and an execution module 205. Among them: The acquisition module 201 is configured to acquire a voltage signal and a current signal, wherein the voltage signal and the current signal are electrical signals including multiple harmonics.

[0103] The preprocessing module 202 is configured to preprocess the voltage signal and the current signal to obtain preprocessing data.

[0104] The analysis module 203 is configured to perform spectral analysis on the preprocessing data to obtain spectral data of the plurality of harmonics.

[0105] The processing module 204 is configured to determine, based on the spectral data, treatment weights corresponding to a passive filter unit, an active filter unit, and a reactor unit, respectively.

[0106] The execution module 205 is configured to adjust, based on the treatment weights corresponding to the passive filter unit, the active filter unit, and the reactor unit, respectively, a capacitance parameter of the passive filter unit, a compensation current parameter of the active filter unit, and an impedance parameter of the reactor unit, respectively.

[0107] The intelligent collaborative harmonic suppression device provided by the application realizes harmonic coverage in a wide frequency range through intelligent adaptive filtering, adapts to load fluctuations and power grid frequency shifts, makes the harmonic treatment effect more stable, solves the problems of traditional harmonic treatment technology in coverage range, scene adaptability, and volume cost, and meets the harmonic treatment needs of different scenes.

[0108] To solve the above technical problems, the application further provides a computer device. Figure 3 , Figure 3 The computer device provided by the application is a basic structure block diagram.

[0109] The computer device 4 includes a memory 41, a processor 42, and a network interface 43 which are connected to each other through a system bus. It should be pointed out that only the computer device 4 with components 41-43 is shown in the figure, but it should be understood that all the shown components are not required to be implemented, and more or fewer components can be alternatively implemented. Among them, those skilled in the art can understand that the computer device herein is a device capable of automatically performing numerical calculation and / or information processing according to pre-set or stored instructions, and its hardware includes but is not limited to a microprocessor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a digital signal processor (DSP), an embedded device, etc.

[0110] The computer device can be a desktop computer, a notebook computer, a palm computer, and a cloud server, etc. The computer device can interact with a user through a keyboard, a mouse, a remote controller, a touchpad, a voice control device, etc.

[0111] The memory 41 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc.

[0112] In some embodiments, the memory 41 can be an internal storage unit of the computer device 4, such as a hard disk or memory of the computer device 4. In other embodiments, the memory 41 can also be an external storage device of the computer device 4, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the computer device 4. Of course, the memory 41 can also include both the internal storage unit and the external storage device of the computer device 4. In this embodiment, the memory 41 is generally used to store the operating system and various application software installed on the computer device 4, such as computer readable instructions of the intelligent collaborative harmonic suppression method, etc. In addition, the memory 41 can also be used to temporarily store various data that have been output or will be output.

[0113] The processor 42 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chip in some embodiments. The processor 42 is generally used to control the overall operation of the computer device 4. In this embodiment, the processor 42 is used to run the computer readable instructions or process data stored in the memory 41, such as the computer readable instructions of the intelligent collaborative harmonic suppression method.

[0114] The network interface 43 can include a wireless network interface or a wired network interface, which is generally used to establish a communication connection between the computer device 4 and other electronic devices.

[0115] The computer device provided in the present application realizes harmonic coverage in a wider frequency range through intelligent adaptive filtering, adapts to load fluctuations and power grid frequency shifts, makes the harmonic control effect more stable, solves the deficiencies of traditional harmonic control technology in coverage range, scene adaptability, volume and cost, and meets the harmonic control needs of different scenes.

[0116] The application also provides another implementation, that is, providing a computer readable storage medium, the computer readable storage medium stores computer readable instructions, the computer readable instructions can be executed by at least one processor, so that the at least one processor executes the steps of the intelligent collaborative harmonic suppression method as described above.

[0117] The computer readable storage medium provided by the application realizes harmonic coverage in a wide frequency range through intelligent adaptive filtering, adapts to load fluctuation and power grid frequency offset, makes the harmonic control effect more stable, solves the problems of traditional harmonic control technology in coverage range, scene adaptability and volume cost, and meets the harmonic control needs of different scenes.

[0118] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other apparatus. Furthermore, embodiments of the application are not described with reference to any particular programming language.

[0119] In the specification provided herein, a large number of specific details are described. However, it can be understood that the embodiments of the application can be practiced without these specific details. Similarly, in order to simplify the application and help understand one or more of the various inventive aspects, in the above description of the exemplary embodiments of the application, various features of the embodiments of the application are sometimes grouped together into a single embodiment, figure, or description thereof. Among them, the claims following the detailed description are hereby expressly incorporated into the detailed description, wherein each claim itself is a separate embodiment of the application.

[0120] Those skilled in the art can understand that the modules in the device in the embodiments can be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments can be combined into one module or unit or component, and in addition, they can be divided into multiple sub-modules or sub-units or sub-components. Except that at least some of such features and / or processes or units are mutually exclusive.

[0121] It should be noted that the above embodiments illustrate the application rather than limit the application, and those skilled in the art can design alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs located between parentheses shall not constitute a limitation on the claims. The word "a" or "an" before an element does not exclude the presence of a plurality of such elements. The application can be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In a unit claim enumerating several means, the several means can be embodied by one and the same item of hardware.

Claims

1. A method of intelligent coordinated harmonic mitigation, comprising: The method comprises: acquiring a voltage signal and a current signal, the voltage signal and the current signal being electric signals comprising a plurality of harmonics; preprocessing the voltage signal and the current signal to obtain preprocessing data; performing spectral analysis on the preprocessing data to obtain spectral data of the plurality of harmonics; based on the spectral data, determining governance weights corresponding to a passive filter unit, an active filter unit and a reactor unit respectively; based on the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively, adjusting a capacitance parameter of the passive filter unit, a compensation current parameter of the active filter unit and an impedance parameter of the reactor unit respectively.

2. The method of claim 1, wherein, The method comprises: based on the preprocessing data, obtaining a load current change rate; inputting the spectral data and the load current change rate into a load prediction model to obtain predicted spectral data of the plurality of harmonics, wherein the load prediction model is a model trained by load current change rates and a plurality of historical spectral data obtained under different load fluctuation scenarios, and the predicted spectral data is spectral data in a preset future time period; based on the predicted spectral data, determining the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively.

3. The method of claim 2, wherein, The predicted spectral data comprises a predicted amplitude, and after the spectral data and the load current change rate are input into the load prediction model to obtain the predicted spectral data of the plurality of harmonics, the method further comprises: if there is a target harmonic with a predicted amplitude greater than a first preset threshold value among the plurality of harmonics, replacing the predicted amplitude of the target harmonic with the first preset threshold value, and the first preset threshold value is a maximum value of historical amplitudes of the target harmonic.

4. The method of claim 1, wherein, The method comprises: obtaining a fundamental frequency of the voltage signal and / or a fundamental frequency of the current signal; calculating frequencies of the plurality of harmonics according to the fundamental frequency of the voltage signal and / or the fundamental frequency of the current signal; determining the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively according to the frequencies and the spectral data.

5. The method of claim 4, wherein, After the capacitance parameter of the passive filter unit, the compensation current parameter of the active filter unit and the impedance parameter of the reactor unit are adjusted based on the governance weights corresponding to the passive filter unit, the active filter unit and the reactor unit respectively, the method further comprises: when the fundamental frequency deviates, correcting the capacitance parameter of the passive filter unit to ensure that the resonance point of the passive filter unit matches the deviated fundamental frequency.

6. The method of claim 1, wherein, The spectral data comprises distortion rates of the plurality of harmonics, and the method comprises: calculating a ratio of a distortion rate of each harmonic to total harmonic distortion rates; According to the ratio of the distortion rate of each harmonic to the total harmonic distortion rate, the governance weight corresponding to the passive filter unit, the active filter unit and the reactor unit is determined.

7. The method of claim 6, wherein, After the governance weight corresponding to the passive filter unit, the active filter unit and the reactor unit is determined according to the ratio of the distortion rate of each harmonic to the total harmonic distortion rate, the method further comprises: If there is an abnormal harmonic with a distortion rate greater than a second preset threshold in the plurality of harmonics, the governance weight of the passive filter unit is reduced, and the governance weight of the active filter unit is increased until the distortion rate of the abnormal harmonic is less than the second preset threshold.

8. An intelligent coordinated harmonic suppression device, characterized by, Comprise: An acquisition module is configured to acquire a voltage signal and a current signal, the voltage signal and the current signal being an electric signal comprising a plurality of harmonics; A preprocessing module is configured to preprocess the voltage signal and the current signal to obtain preprocessed data; An analysis module is configured to perform frequency spectrum analysis on the preprocessed data to obtain frequency spectrum data of the plurality of harmonics; A processing module is configured to determine the governance weight corresponding to the passive filter unit, the active filter unit and the reactor unit based on the frequency spectrum data; An execution module is configured to adjust the capacitance parameter of the passive filter unit, the compensation current parameter of the active filter unit and the impedance parameter of the reactor unit based on the governance weight corresponding to the passive filter unit, the active filter unit and the reactor unit, respectively.

9. A computer device, comprising: The memory and the processor, the memory has computer readable instructions, the processor executes the computer readable instructions and realizes the steps of the intelligent collaborative harmonic suppression method in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by the processor to realize the steps of the intelligent collaborative harmonic suppression method in any one of claims 1 to 7. The computer readable storage medium stores computer readable instructions, and the computer readable instructions are executed by the processor to realize the steps of the intelligent collaborative harmonic suppression method in any one of claims 1 to 7.

Images