Active power filter virtual impedance control method and device, equipment and medium

Abstract

The present disclosure provides an active power filter virtual impedance control method and device, electronic equipment and medium, and relates to the technical field of power electronics. The specific implementation scheme is: obtaining the output current of each active power filter and the common coupling point voltage; through a band-pass filter, the low-frequency oscillation components in the output current and the common coupling point voltage are extracted respectively, and the phases of the output currents of each active power filter are extracted, and the phase difference between the phases is calculated; based on the low-frequency oscillation components, an oscillation energy function is constructed; by injecting a disturbance signal into the system, the current damping ratio is estimated online according to the open-loop frequency response by measuring; based on the oscillation energy function, the target damping ratio and the current damping ratio, the adaptive virtual resistance is calculated, and based on the change rate of the oscillation energy function and the phase difference, the adaptive virtual inductance is calculated; based on the adaptive virtual resistance and the adaptive virtual inductance, the final modulation voltage is generated based on the adaptive virtual impedance.

Description

Technical Field

[0001] This disclosure belongs to the field of power electronic control technology, specifically relating to a virtual impedance control method and device for an active power filter, an electronic device, and a computer-readable storage medium. Background Technology

[0002] Active power filters (APFs) are core power electronic devices for mitigating grid harmonics and compensating for reactive power. To meet the demands of high-capacity current output and improve system redundancy and reliability, multiple APFs are often operated in parallel in engineering applications. Typical applications include harmonic mitigation in industrial distribution networks and power quality regulation in renewable energy power plants, with up to 16 units or more connected in parallel.

[0003] However, a multi-machine parallel system is equivalent to a high-order, strongly coupled multi-input multi-output system, with complex circulating current channels. During actual engineering commissioning and operation, especially under reactive power offsetting conditions—that is, some APFs output inductive reactive power and some output capacitive reactive power to simulate full-load offsetting tests—the system is highly prone to low-frequency oscillations. These oscillations typically occur in the range of 1Hz to 50Hz, manifesting as periodic large fluctuations in the output current of each module, increased grid-side current distortion, and DC bus voltage oscillations. In severe cases, overcurrent protection can be triggered, leading to a cascading system collapse.

[0004] Traditional virtual impedances often use fixed parameters, which cannot adaptively adjust with changes in grid impedance, load characteristics, and the number of parallel units, resulting in limited damping effect; simply increasing the damping will worsen the dynamic response, making it difficult to balance stability and compensation accuracy.

[0005] Therefore, in scenarios where multiple active power filters operate in parallel, there is a need for a technical solution that can dynamically adjust the damping injection strategy according to the system oscillation state while taking into account the economic efficiency of steady-state operation. Summary of the Invention

[0006] This disclosure provides a method and apparatus for virtual impedance control of an active power filter, an electronic device, and a computer-readable storage medium.

[0007] According to the first aspect, a virtual impedance control method for active power filters is provided, applied to multiple active power filters connected in parallel through a common coupling point. The method includes: acquiring the output current and common coupling point voltage of each active power filter; extracting low-frequency oscillation components from the output current and common coupling point voltage using a bandpass filter with adaptive center frequency adjustment, multiplying the low-frequency oscillation components of the output current with a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter, and calculating the phase difference between the phase of the output current and the phase of the global average current; and calculating the virtual impedance control of each active power filter based on the low-frequency oscillation components of the output current and the common coupling point voltage. The active and reactive circulating current deviations of the power filters are calculated, and an oscillation energy function is constructed based on these deviations. By injecting a disturbance signal into the system, the open-loop frequency response of multiple active power filters is measured, and the current damping ratio of these filters is estimated online based on the open-loop frequency response. An adaptive virtual resistance is calculated based on the oscillation energy function, the target damping ratio, and the current damping ratio, and an adaptive virtual inductance is calculated based on the rate of change and phase difference of the oscillation energy function. An adaptive virtual impedance is calculated based on the adaptive virtual resistance and the adaptive virtual inductance. The adaptive virtual impedance is multiplied by the output current and then superimposed onto the original voltage reference command to generate the final modulation voltage.

[0008] According to the second aspect, another virtual impedance control device for active power filters is provided, applied to multiple active power filters connected in parallel through a common coupling point. The device includes: an acquisition unit configured to acquire the output current and common coupling point voltage of each active power filter; an extraction unit configured to extract low-frequency oscillation components from the output current and common coupling point voltage using a bandpass filter with adaptive center frequency adjustment, multiply the low-frequency oscillation components of the output current with a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter, and calculate the phase difference between the phase of the output current and the phase of the global average current; and a construction unit configured to calculate the phase difference between each active power filter based on the low-frequency oscillation components of the output current and the common coupling point voltage. The system calculates the active and reactive circulating current deviations and constructs an oscillation energy function based on these deviations. An estimation unit is configured to measure the open-loop frequency response of multiple active power filters by injecting a disturbance signal into the system, and to estimate the current damping ratio of the multiple active power filters online based on the open-loop frequency response. An inductance calculation unit is configured to calculate an adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio, and to calculate an adaptive virtual inductance based on the rate of change and phase difference of the oscillation energy function. An impedance calculation unit is configured to calculate an adaptive virtual impedance based on the adaptive virtual resistance and adaptive virtual inductance. A modulation unit is configured to multiply the adaptive virtual impedance by the output current and then superimpose it onto the original voltage reference command to generate the final modulation voltage.

[0009] According to a third aspect, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform a method as described in any implementation of the first aspect.

[0010] According to a fourth aspect, a computer-readable storage medium is provided having computer instructions stored thereon for causing a computer to perform the method as described in any implementation of the first aspect.

[0011] The active power filter virtual impedance control method and apparatus provided in the embodiments of this disclosure are applied to multiple active power filters connected in parallel through a common coupling point. First, the output current of each active power filter and the voltage at the common coupling point are obtained. Second, using a bandpass filter with adaptive center frequency adjustment, low-frequency oscillation components are extracted from the output current and the voltage at the common coupling point, and the low-frequency oscillation component of the output current is multiplied by a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter. The phase difference between the phase of the output current and the phase of the global average current is calculated. Third, based on the low-frequency oscillation components of the output current and the voltage at the common coupling point, the phase difference between each active power filter and the voltage at the common coupling point is calculated. The system calculates the active and reactive circulating current deviations of the active power filters and constructs an oscillation energy function based on these deviations. Then, by injecting a disturbance signal into the system, the open-loop frequency response of multiple active power filters is measured, and the current damping ratio of each filter is estimated online. Subsequently, an adaptive virtual resistance is calculated based on the oscillation energy function, the target damping ratio, and the current damping ratio. An adaptive virtual inductance is calculated based on the rate of change and phase difference of the oscillation energy function. Then, an adaptive virtual impedance is calculated based on the adaptive virtual resistance and inductance. Finally, the adaptive virtual impedance is multiplied by the output current and superimposed onto the original voltage reference command to generate the final modulation voltage. This solves the low-frequency oscillation problem that occurs with multiple (especially 16 or more) active power filters under reactive power suppression conditions, improves the parallel stability and dynamic response performance of the system, and also considers the economic efficiency of steady-state operation.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein: Figure 1 This is a flowchart of an embodiment of the virtual impedance control method for active power filters according to the present disclosure; Figure 2 This is a schematic diagram of a structure of an embodiment of the virtual impedance control device for an active power filter according to the present disclosure; Figure 3 This is a block diagram of an electronic device used to implement the virtual impedance control method for active power filters according to embodiments of the present disclosure. Detailed Implementation

[0014] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0015] The technical solutions of this disclosure are illustrated below through specific embodiments. It should be understood that one or more steps mentioned in this disclosure do not preclude the existence of other methods and steps before or after the combined steps, or that other methods and steps may be inserted between these explicitly mentioned steps. It should also be understood that these examples are for illustrative purposes only and are not intended to limit the scope of this disclosure. Unless otherwise stated, the numbering of each method step is only for the purpose of identifying each method step, and not to limit the order of each method or to limit the scope of implementation of this disclosure. Changes or adjustments to their relative relationships, without substantial changes to the technical content, can also be considered as within the scope of implementation of this disclosure.

[0016] The raw materials and instruments used in the examples are not subject to any specific restrictions on their source; they can be purchased from the market or prepared according to conventional methods known to those skilled in the art.

[0017] In existing technologies, virtual impedance is often used to suppress parallel circulating current, which involves artificially adding virtual resistance or inductance to the control loop. However, the traditional fixed virtual impedance method has the following shortcomings: Dynamic and steady-state inconsistencies: To obtain sufficient damping, a large virtual impedance needs to be set, but this can lead to a significant drop in output voltage and increased losses during steady-state operation; Lack of adaptability to operating conditions: Under special operating conditions such as reactive power countermeasures, the system's equivalent impedance and oscillation modes change significantly, and the fixed impedance cannot provide sufficient positive damping; Insufficient multi-machine coordination capability: When each APF independently adjusts its virtual impedance, inconsistent adjustment directions or even deterioration of mutual coupling may occur.

[0018] Therefore, there is an urgent need for a suppression method that can adaptively sense low-frequency oscillations and dynamically adjust virtual impedance, while also possessing multi-machine collaborative capabilities.

[0019] To address the technical problems in the prior art, this disclosure provides a virtual impedance control method for active power filters. This method uses a bandpass filter with adaptive center frequency adjustment to extract low-frequency oscillation components in real time and track the dominant oscillation frequency. Simultaneously, it estimates the current damping ratio of the system online, and then adaptively adjusts the virtual resistance based on the oscillation energy function and damping margin. It also adaptively adjusts the virtual inductance based on the oscillation energy change rate and phase difference. Finally, it superimposes the adaptive virtual impedance onto the voltage reference command to inject positive damping into the parallel system. Figure 1The flowchart 100 illustrates an embodiment of the virtual impedance control method for active power filters according to this disclosure. This method is applied to multiple active power filters connected in parallel through a common coupling point. The method specifically includes the following steps: Step 101: Obtain the output current and common coupling point voltage of each active power filter among multiple active power filters.

[0020] In this embodiment, the Point of Common Coupling (PCC) refers to the electrical connection point between a distributed power source, power electronic device, or load and the public power grid. In a multi-machine parallel system, the PCC is the confluence node of the output currents of each APF and also the interface between the system and the power grid. The voltage stability at the PCC directly reflects the operating quality of the parallel system; therefore, this disclosure uses the PCC voltage as one of the key signals for monitoring low-frequency oscillations in the system.

[0021] In this embodiment, the digital controller of each active power filter can synchronously sample the current of its respective output branch using a current sensor to obtain the output current signal. Simultaneously, it samples the three-phase voltage at the common coupling point using a voltage sensor and obtains the common coupling point voltage signal after coordinate transformation. The sampled data, after signal conditioning and analog-to-digital conversion, is sent to the digital control unit for subsequent low-frequency oscillation component extraction and virtual impedance adaptive adjustment. As one implementation, the aforementioned digital controller can be based on a digital signal processor and a field-programmable gate array (FPGA), with the sampling period set within the range of 100μs to 500μs.

[0022] Step 102: Using a bandpass filter with adaptive center frequency adjustment, extract the low-frequency oscillation components from the output current and the common coupling point voltage, and multiply the low-frequency oscillation components of the output current with the orthogonal reference signal of the same frequency to extract the phase of the output current of each active power filter, and calculate the phase difference between the phase of the output current and the phase of the global average current.

[0023] In this embodiment, the orthogonal reference signal with the same frequency refers to a pair of sine / cosine reference signals with the same frequency as the target oscillation signal but with a 90° phase difference. By multiplying the extracted low-frequency oscillation component of the output current with the orthogonal reference signal with the same frequency and then low-pass filtering, the instantaneous phase information of the oscillation component can be demodulated.

[0024] In this embodiment, the global average current refers to the arithmetic mean of the output currents of all active power filters (APFs) operating in parallel. In a multi-machine parallel system, the global average current represents the average output level of the system and can be used as a benchmark to measure the balance of the output current of each APF. By comparing the output current of a single APF with the global average current, the current deviation and phase deviation of that module can be obtained, which can then be used for circulating current analysis and virtual impedance coordinated adjustment.

[0025] In this embodiment, the low-frequency oscillation component typically refers to a periodic voltage, current, or power fluctuation component with a frequency range of 1Hz to 50Hz. Its amplitude is significantly higher than that of normal steady-state ripple and may diverge over time. This component is extracted from the original signal using a bandpass filter with an adaptive center frequency.

[0026] In this embodiment, the bandpass filter with adaptive center frequency adjustment extracts low-frequency oscillation components from the output current and the common coupling point voltage, respectively. This includes: inputting the output current and the common coupling point voltage to a preset second-order generalized integrator to generate a first orthogonal signal in phase with the input signal and a second orthogonal signal with a 90-degree phase difference; calculating the deviation between the input signal and the first orthogonal signal; multiplying the deviation by the second orthogonal signal and inputting it to the integrator controller in the frequency-locked loop to calculate the dynamic frequency compensation amount; using the dynamic frequency compensation amount to update the resonant frequency of the second-order generalized integrator in real time, using the updated resonant frequency as the center frequency of the bandpass filter, and directly outputting the extracted low-frequency oscillation components through the second-order generalized integrator.

[0027] Optionally, a bandpass filter with adaptive center frequency adjustment is used to extract low-frequency oscillation components from the output current and the common coupling point voltage, including: performing sliding window sampling on the output current and the common coupling point voltage to obtain a sequence data window of a preset time length; performing time-frequency domain transformation on the sequence data window to obtain the spectral distribution characteristics within the target frequency band of 1Hz~50Hz; extracting the frequency corresponding to the dominant peak with the largest amplitude in the spectral distribution characteristics; performing noise reduction and smoothing processing on the frequency corresponding to the dominant peak through a smoothing filtering algorithm, and using the smoothed frequency as the center frequency of the bandpass filter so that the bandpass filter can extract low-frequency oscillation components.

[0028] Optionally, a bandpass filter with adaptive center frequency adjustment is used to extract low-frequency oscillation components from the output current and the common coupling point voltage, including: acquiring historical time-series characteristic data of the output current and the common coupling point voltage in real time; inputting the historical time-series characteristic data into a lightweight neural network model deployed in the digital controller, the lightweight neural network model being pre-trained using sample data containing low-frequency oscillation characteristics of various reactive power supply conditions; directly outputting the predicted current dominant oscillation frequency through the lightweight neural network model; and using the predicted current dominant oscillation frequency as the center frequency of the bandpass filter so that the bandpass filter can extract the low-frequency oscillation components.

[0029] In this embodiment, the low-frequency oscillation component of the output current is multiplied and demodulated (synchronous detection) with a quadrature reference signal of the same frequency to obtain a quadrature component characterizing the phase of the low-frequency oscillation component. Based on this, the phase of the output current of each active power filter is calculated, and then compared with the global average current phase calculated from the output current of multiple filters to obtain the phase difference of the output current phase of each filter relative to the global average phase. This phase difference reflects the phase misalignment of each active power filter relative to the average operating state of the parallel system, providing a basis for the subsequent adaptive adjustment of the virtual inductor and providing phase deviation information for the subsequent adaptive adjustment of the virtual impedance and the multi-filter cooperative damping injection.

[0030] Step 103: Based on the low-frequency oscillation components of the output current and the common coupling point voltage, calculate the active current deviation and reactive current deviation of each active power filter, and construct the oscillation energy function based on the active current deviation and reactive current deviation.

[0031] In this embodiment, in a multi-machine parallel system, due to differences in the amplitude and phase of the output voltage of each APF, a circulating current will form between the parallel modules without passing through the load. This circulating current is called the circulating current. After decomposing the output current of each APF and the PCC voltage into low-frequency oscillation components, the instantaneous active power and reactive power of each module are calculated based on instantaneous power theory. Then, the average instantaneous power of all modules is subtracted from the instantaneous power of each module to obtain the active power circulating current deviation and reactive power circulating current deviation of that module. These two deviations reflect the degree of imbalance in the distribution of active and reactive power in that module, respectively, and are the fundamental variables for constructing the oscillation energy function.

[0032] In this embodiment, the oscillation energy function is used as a time-domain indicator to quantify the intensity of low-frequency oscillations in the system. It is constructed based on the active power circulation deviation and reactive power circulation deviation. By performing mathematical operations on these two values ​​(such as product integrals, sum-of-squares integrals, or complex power modulus integrals) and accumulating the results within a time window, a scalar energy value is obtained. A larger energy value indicates more severe low-frequency oscillations in the system; conversely, a smaller value indicates that the system is becoming more stable. The value of the oscillation energy function and its rate of change are used to calculate the adaptive virtual resistance and adaptive virtual inductance, respectively.

[0033] In this embodiment, the digital controller calculates the low-frequency oscillation power components of each active power filter based on the low-frequency oscillation components of the output current and common coupling point voltage obtained in step 102, according to the instantaneous power calculation method. Specifically, the instantaneous power is obtained by multiplying the low-frequency oscillation voltage component and the low-frequency oscillation current component, and then decomposed into the active power component and reactive power component of each active power filter. Subsequently, the average active power fluctuation component and the average reactive power fluctuation component of all parallel active power filters are calculated. The active power circulating current deviation is obtained by subtracting the average active power fluctuation component from the active power component of each filter, and the reactive power circulating current deviation is obtained by subtracting the average reactive power fluctuation component from the reactive power component of each filter. The active power circulating current deviation reflects the degree of deviation of the active power filter in active power distribution, and the reactive power circulating current deviation reflects the degree of deviation of the filter in reactive power distribution.

[0034] In this embodiment, the power fluctuation component near the low-frequency oscillation frequency is extracted. Specifically, the low-frequency oscillation components of voltage and current are extracted by a bandpass filter. Then, the power components of each module are calculated according to the instantaneous power calculation method. The average power fluctuation component is then obtained. Finally, the average power fluctuation component is subtracted from each module to obtain the active power circulation deviation and reactive power circulation deviation of each module.

[0035] In this embodiment, the above-mentioned construction of the oscillation energy function based on the active power circulation deviation and the reactive power circulation deviation includes: squaring the active power circulation deviation and the reactive power circulation deviation respectively, summing them according to their respective weighting coefficients, and then integrating them within a time window to construct the oscillation energy function.

[0036] Optionally, the above-mentioned construction of the oscillation energy function based on the active power circulation deviation and the reactive power circulation deviation includes: treating the active power circulation deviation and the reactive power circulation deviation as the real and imaginary parts of the complex power deviation, first obtaining the magnitude (or the square of the magnitude) of the complex power deviation, and then integrating the magnitude within the time window to construct the oscillation energy function.

[0037] In this embodiment, the controller collects the output current of each active power filter and the voltage at the point of common coupling in real time, and extracts the corresponding low-frequency oscillation components from the output current and the voltage at the point of common coupling through a preset low-frequency band-pass filter; then, based on the low-frequency oscillation components, the deviation of each active power filter relative to the average power distribution of the parallel system is calculated, so as to obtain the respective active circulating current deviation and reactive circulating current deviation; further, the active circulating current deviation and the reactive circulating current deviation are combined according to a set weight, and integrated / accumulated in a sliding time window to characterize the oscillation intensity, so as to construct an oscillation energy function for describing the amplitude and sustainability of low-frequency oscillation, which provides a quantitative basis for the subsequent adaptive adjustment of the virtual impedance.

[0038] Step 104: Inject a disturbance signal into the system, measure the open-loop frequency responses of the plurality of active power filters, and estimate the current damping ratios of the plurality of active power filters online according to the open-loop frequency responses.

[0039] In this embodiment, the system refers to the overall multi-machine parallel power electronic system formed by a plurality of active power filters (APFs) connected in parallel to the power grid through a point of common coupling (PCC). The above step 104 includes: injecting a step disturbance signal (or a square wave disturbance signal) into the reactive current reference command of the system, the amplitude of the step disturbance is set within a first preset range of the rated reactive power of the plurality of active power filters (for example, 3%-5% of the rated reactive power), and the duration is set to 12 times the low-frequency oscillation period; taking the voltage at the point of common coupling as the system output, collecting the time-domain response waveform of the voltage at the point of common coupling after the step disturbance is injected; performing band-pass filtering on the time-domain response waveform to retain the oscillation component in the target frequency band, and extracting the instantaneous amplitude envelope of the oscillation component through Hilbert transform; according to two adjacent peaks in the same direction A on the instantaneous amplitude envelope n and A n+1 , calculate the logarithmic decrement rate ( ), and convert the current damping ratio according to the damping theory of the second-order system: .

[0040] Optionally, step 104 may include: injecting a sweeping sinusoidal disturbance signal with a continuously varying frequency within a linear range (e.g., from 0.5 Hz to 20 Hz) into the reactive current reference command of the system, with the sweeping speed set to maintain at least 2 to 3 complete oscillation cycles at each frequency point, and the amplitude of the sweeping disturbance being constant and set within a second preset range of rated reactive power (e.g., 2% to 4%); synchronously recording the injected disturbance signal and the voltage response at the common coupling point at each frequency point, and calculating the amplitude-frequency characteristics and phase-frequency characteristics at each frequency point through cross-correlation analysis or fast Fourier transform to obtain a discrete open-loop frequency response sequence; performing least-squares fitting of the discrete open-loop frequency response sequence with a preset second-order oscillatory link transfer function model, constructing a cost function, and solving for the model parameters that minimize the cost function through gradient descent or linearized least squares method, and directly using the fitted model parameters as the current damping ratio of multiple active power filters.

[0041] Step 105: Calculate the adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio, and calculate the adaptive virtual inductance based on the rate of change and phase difference of the oscillation energy function.

[0042] In this embodiment, the above-mentioned calculation of the adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio includes: setting a first oscillation energy threshold. Second oscillation energy threshold and maximum virtual resistance , where (0 < < ); Calculate the difference between the target damping ratio and the current damping ratio; Calculate the virtual resistance adjustment increment according to different linear gains based on the interval of the oscillation energy function: Specifically, when the value of the oscillation energy function is less than the first oscillation energy threshold, the virtual resistance adjustment increment is set to zero, so that the adaptive virtual resistance is maintained at the steady-state virtual resistance without additional adjustment; when the value of the oscillation energy function is greater than or equal to the first oscillation energy threshold and less than the second oscillation energy threshold, the first difference between the value of the oscillation energy function and the first oscillation energy threshold is calculated, and the second difference between the second oscillation energy threshold and the first oscillation energy threshold is calculated; the first difference is divided by the second difference to obtain the normalized adjustment ratio; the normalized adjustment ratio is multiplied by the first gain coefficient, and then multiplied by the target resistance. The damping ratio deviation between the current damping ratio and the current damping ratio is used to obtain the virtual resistance adjustment increment. When the value of the oscillation energy function is greater than or equal to the second oscillation energy threshold, the second gain coefficient is multiplied by the damping ratio deviation to obtain the virtual resistance adjustment increment. Furthermore, the virtual resistance adjustment increment is hard-limited so that the sum of the steady-state virtual resistance and the virtual resistance adjustment increment does not exceed the preset maximum virtual resistance. The rise and fall rates of the adaptive virtual resistance are limited so that the absolute value of its change per unit time does not exceed the quotient obtained by dividing the difference between the maximum virtual resistance and the steady-state virtual resistance by the preset adjustment time constant. The steady-state virtual resistance is added to the rate-limited adjustment increment to obtain the adaptive virtual resistance.

[0043] Optionally, the above calculation of the adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio includes: calculating the difference between the target damping ratio and the current damping ratio, and using this difference as the closed-loop error signal; calculating the ratio between the value of the oscillation energy function and a preset oscillation energy threshold; multiplying the ratio by the scheduling sensitivity coefficient and adding one to obtain the gain scheduling multiplier; multiplying the reference proportional coefficient by the gain scheduling multiplier to obtain the time-varying proportional gain; multiplying the reference integral coefficient by the gain scheduling multiplier to obtain the time-varying integral gain; multiplying the time-varying proportional gain by the closed-loop error signal to obtain the proportional adjustment term; and adjusting the closed-loop resistance accordingly. The loop error signal is accumulated and integrated over time to obtain the cumulative integral error. The time-varying integral gain is then multiplied by the cumulative integral error to obtain the integral adjustment term. The proportional adjustment term is added to the integral adjustment term to obtain the virtual resistance adjustment increment. The adaptive virtual resistance is limited at the upper and lower limits to ensure that it is not less than the steady-state virtual resistance and not greater than the preset maximum virtual resistance. Furthermore, an integral anti-saturation strategy is adopted: when the adaptive virtual resistance reaches the maximum virtual resistance or the steady-state virtual resistance, the cumulative update of the integral adjustment term in the PI controller is paused. The steady-state virtual resistance is added to the virtual resistance adjustment increment to obtain the adaptive virtual resistance.

[0044] In this embodiment, the above-mentioned calculation of the adaptive virtual inductance based on the rate of change and phase difference of the oscillation energy function includes: preset a phase difference threshold and a hysteresis width; using the phase difference threshold minus half of the hysteresis width as the lower limit of the hysteresis comparison interval, and using the phase difference threshold plus half of the hysteresis width as the upper limit of the hysteresis comparison interval, thus forming the hysteresis comparison interval; comparing the absolute value of the phase difference with the hysteresis comparison interval to generate a binarized adjustment enable signal: when the absolute value of the phase difference increases from less than the lower limit of the hysteresis comparison interval to greater than the upper limit of the hysteresis comparison interval, the adjustment enable signal is enabled; when the absolute value of the phase difference decreases from greater than the upper limit of the hysteresis comparison interval to less than the lower limit of the hysteresis comparison interval, the adjustment enable signal is disabled; in response to the enabled signal, the oscillation energy function is adjusted accordingly. The rate of change of the oscillation energy function is integrated over at least one low-frequency oscillation cycle to obtain the change in oscillation energy within the time window. The change in oscillation energy is equal to the value of the oscillation energy function at the current moment minus the value of the oscillation energy function one low-frequency oscillation cycle ago. The change in oscillation energy is multiplied by a preset virtual inductance adjustment gain coefficient, and then multiplied by the sign of the rate of change of the oscillation energy function to obtain a directional virtual inductance adjustment increment. If the rate of change of the oscillation energy function is positive, the sign is positive; if it is negative, the sign is negative. The steady-state virtual inductance is added to the virtual inductance adjustment increment to obtain a candidate adaptive virtual inductance. The rate of change and upper and lower limits are applied to the candidate adaptive virtual inductance to ensure that the adaptive virtual inductance is not less than a preset minimum virtual inductance and not greater than a preset maximum virtual inductance.

[0045] Optionally, the above calculation of the adaptive virtual inductance based on the rate of change and phase difference of the oscillating energy function includes: directly retaining the positive and negative signs of the phase difference as the closed-loop error signal, without taking the absolute value or setting a preset phase difference threshold; inputting the closed-loop error signal into the PI controller to calculate the first adjustment component, i.e., the feedback component, where the first adjustment component is equal to the proportional coefficient multiplied by the closed-loop error signal, plus the integral coefficient multiplied by the integral value of the closed-loop error signal accumulated over time; multiplying the rate of change of the oscillating energy function by a preset feedforward coefficient to obtain the second adjustment component, i.e., the feedforward component, which is used to provide advance compensation when the oscillating energy changes rapidly; implementing an integral anti-saturation strategy on the first adjustment component, pausing the update of the integral term in the PI controller when the adaptive virtual inductance reaches the preset minimum or maximum virtual inductance; and adding the steady-state virtual inductance, the first adjustment component, and the second adjustment component to obtain the adaptive virtual inductance.

[0046] Step 106: Calculate the adaptive virtual impedance based on the adaptive virtual resistance and adaptive virtual inductance.

[0047] In this embodiment, the digital controller combines the adaptive virtual resistance and adaptive virtual inductance into an adaptive virtual impedance. In the frequency domain, the virtual impedance can be expressed as the sum of the virtual resistance multiplied by the Laplace operator and the adaptive virtual inductance. Within each control cycle of the digital controller, the frequency domain relationship is converted into a time domain implementation, that is, the differential is replaced by the differential, and the multiplication relationship in the frequency domain is converted into calculating the rate of change of the current and then multiplying it by the inductance value, thereby generating the voltage drop command for the virtual impedance.

[0048] Step 107: Multiply the adaptive virtual impedance by the output current and then add it to the original voltage reference command to generate the final modulation voltage.

[0049] In this embodiment, as shown in equation (1), the adaptive virtual impedance is... Multiply by the output current Then superimposed on the original voltage reference command Generate the final modulation voltage : = - (1) Final modulation voltage This causes the APF output impedance to exhibit positive damping characteristics, shifting the system poles to the left half-plane, thereby suppressing low-frequency oscillations.

[0050] In this embodiment, at the current loop output terminal and before the modulation wave is generated, the digital controller multiplies the adaptive virtual impedance by the output current, inverts the result, and then superimposes it onto the original voltage reference command to generate the final modulation voltage. This final modulation voltage, after pulse width modulation, drives the three-phase inverter bridge arm, causing the output impedance of the active power filter to exhibit positive damping characteristics, thereby pushing the system poles to the left half-plane and suppressing low-frequency oscillations in the 1Hz to 50Hz frequency band.

[0051] The virtual impedance control method for active power filters provided in this disclosure first obtains the output current and common coupling point voltage of each of the multiple active power filters. Second, using a bandpass filter with adaptive center frequency adjustment, the low-frequency oscillation components in the output current and common coupling point voltage are extracted respectively. The low-frequency oscillation components of the output current are multiplied by a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter, and the phase difference between the phase of the output current and the phase of the global average current is calculated. Third, based on the low-frequency oscillation components of the output current and common coupling point voltage, the active circulating current deviation and reactive circulating current of each active power filter are calculated. The system first identifies the deviation and constructs an oscillation energy function based on the active and reactive circulating current deviations. Then, by injecting a disturbance signal into the system, the open-loop frequency response of multiple active power filters is measured. Based on the open-loop frequency response, the current damping ratio of multiple active power filters is estimated online. Subsequently, based on the oscillation energy function, the target damping ratio, and the current damping ratio, an adaptive virtual resistance is calculated, and based on the rate of change and phase difference of the oscillation energy function, an adaptive virtual inductance is calculated. Then, based on the adaptive virtual resistance and adaptive virtual inductance, an adaptive virtual impedance is calculated. Finally, the adaptive virtual impedance is multiplied by the output current and superimposed onto the original voltage reference command to generate the final modulation voltage. This solves the low-frequency oscillation problem that occurs with multiple (especially 16 or more) active power filters under reactive power suppression conditions, improves the parallel stability and dynamic response performance of the system, and also considers the economic efficiency of steady-state operation.

[0052] In some optional implementations of this disclosure, the extraction of low-frequency oscillation components from the output current and the common coupling point voltage by the bandpass filter with adaptive center frequency adjustment includes: setting the initial center frequency of the bandpass filter with adaptive center frequency adjustment; sending the extracted signal into a frequency detector; and adjusting the center frequency of the bandpass filter in real time by a proportional-integral regulator so that the center frequency tracks the main oscillation frequency in the signal to obtain the low-frequency oscillation components from the output current and the common coupling point voltage.

[0053] In this optional implementation, the digital controller feeds the output current signals and common coupling point voltage signals of each active power filter into a bandpass filter with adaptive center frequency adjustment. The initial center frequency of the bandpass filter can be set to a value within the range of 1Hz to 50Hz, for example, 25Hz. The filtered signal is fed into a frequency detector to obtain the dominant oscillation frequency in the current signal, and the center frequency of the bandpass filter is updated in real time by a proportional-integral controller to continuously track the dominant low-frequency oscillation frequency, thereby obtaining the low-frequency oscillation components in the output current and common coupling point voltage, respectively.

[0054] In some optional implementations of this disclosure, the above-mentioned construction of the oscillation energy function based on the active power circulation deviation and the reactive power circulation deviation includes: integrating the product of the active power circulation deviation and the reactive power circulation deviation within a time window to construct the oscillation energy function.

[0055] In this optional implementation, the active power circulating current deviation is calculated based on the instantaneous power of each module. and reactive power circulation deviation Construct the oscillation energy function as shown in equation (2): (t) = (2) In equation (2), the integration window T is taken as 1 to 2 times the period of the low-frequency oscillation.

[0056] In this optional implementation, the product of the active power circulating current deviation and the reactive power circulating current deviation is integrated within a sliding time window to construct an oscillation energy function. This oscillation energy function quantitatively characterizes the intensity and duration of the low-frequency oscillation. The integration window length can be 1 to 2 times the low-frequency oscillation period; for example, when the minimum oscillation frequency of interest is 1Hz, the integration window can be 1 to 2 seconds. Since the oscillation energy function contains both active and reactive power deviation information, it can comprehensively reflect the oscillation state of the parallel system. The oscillation energy threshold can be determined based on the rated current of a single active power filter and the minimum oscillation period; for example, it can be the square of 10% of the effective value of the rated current of a single filter multiplied by half of the minimum oscillation period.

[0057] In some optional implementations of this disclosure, the above-mentioned method of injecting a disturbance signal into the system, measuring the open-loop frequency response of multiple active power filters, and estimating the current damping ratio of the multiple active power filters online based on the open-loop frequency response includes: injecting a periodic disturbance signal into the system, the amplitude of which is set within a first preset range of the rated reactive power of the multiple active power filters; using the disturbance signal as the system input and the measured common coupling point voltage response as the system output to obtain the open-loop frequency response of the multiple active power filters; and calculating the current damping ratio based on the phase margin and correction coefficient of the open-loop frequency response.

[0058] In this optional implementation, a small disturbance signal is applied to the PCC voltage or the total output current, and the open-loop frequency response of the system is measured, as shown in equation (3) based on the phase margin. Convert the current damping ratio : = (3) In equation (3) This is a correction factor, with a value ranging from 0.9 to 1.1.

[0059] In this optional implementation, the digital controller injects a periodic disturbance signal into the reactive current command via software. The frequency of the disturbance signal can be selected as a 2Hz sine wave, and the amplitude is set within the range of 2% to 5% of the rated reactive power of the multiple active power filters, for example, 3%. The injected disturbance signal is used as the system input, and the measured voltage fluctuation response at the common coupling point is used as the system output. The open-loop frequency response of the system is obtained using an online frequency identification method. For example, the recursive least squares method can be used to identify the open-loop transfer function online, and then calculate the phase margin of the open-loop frequency response. The current damping ratio is obtained by converting the phase margin and the correction coefficient. The conversion relationship can be expressed as the current damping ratio equal to the phase margin divided by the square root of 1 plus the square of the phase margin, and then multiplied by the correction coefficient. The correction coefficient can range from 0.9 to 1.1. The current damping ratio reflects the energy attenuation characteristics of the system under disturbance; the smaller the damping ratio, the closer the system is to the instability edge. A conservative minimum damping ratio value, such as 0.05, can be preset at the factory and then automatically corrected based on online estimation results during operation.

[0060] In some optional implementations of this disclosure, the above-mentioned calculation of the adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio includes: calculating the ratio of the value of the oscillation energy function to a preset oscillation energy threshold; multiplying the ratio by a sensitivity coefficient to obtain a product, and using the product as the independent variable of the hyperbolic tangent function to obtain the soft-limiting function value; calculating the damping difference between the target damping ratio and the current damping ratio; multiplying the soft-limiting function value, the damping difference, and the preset virtual resistance adjustment damping coefficient to obtain the virtual resistance adjustment increment; and adding the steady-state virtual resistance to the virtual resistance adjustment increment to obtain the adaptive virtual resistance.

[0061] In this optional implementation, the soft-limiting function value refers to the output value obtained by scaling the ratio of the oscillation energy function to the oscillation energy threshold using a sensitivity coefficient and then inputting it into the hyperbolic tangent function. This value is used as the amplitude limiting factor for the virtual resistance adjustment increment, ensuring smooth saturation of the adjustment amount when the oscillation energy is at its maximum, and preventing a step jump in the virtual resistance that could cause a secondary impact on the system.

[0062] In this optional implementation, the adaptive virtual impedance is defined as shown in equation (4): (t) = (t) + s (t)(4) In equation (4), For adaptive virtual resistance, () represents the adaptive virtual inductance; s is the Laplace operator, sL(t) represents one inductive reactance at frequency, and its value increases linearly with frequency. To achieve this in each control cycle of the digital controller, the frequency domain must be converted into the time domain, and a voltage drop command for a virtual impedance must be generated in the controller. The core is to replace the derivative with the differential, and convert the multiplication relationship sL in the frequency domain into the rate of change of current in the time domain and then multiply it by the inductance value.

[0063] The adaptive virtual resistance (dominant low-frequency damping) is represented by equation (5): (t) = + tanh ( ) ( (5) In equation (5), It is a steady-state virtual resistance, with a value range of 0.05Ω to 0.1Ω; This represents the damping coefficient adjusted by the virtual resistance; tan(h) is the hyperbolic tangent function, used to avoid over-adjustment. The value represents the sensitivity coefficient, ranging from 1 to 5. When α < 1, the response is slow. When α > 5, the hyperbolic tangent function approximates a step function, which can easily cause virtual resistance jumps and cause secondary impacts on the system. This indicates the preset oscillation energy threshold; Represents the oscillation energy function; This represents the target damping ratio, with a value range of 0.2 to 0.3. This indicates the current damping ratio.

[0064] In equation (5), The tuning formula is shown in equation (6): = (6) In equation (6), Determined by voltage sag tolerance. This is the minimum damping ratio measured under the worst operating conditions of the system.

[0065] In this optional implementation, for the adaptive virtual resistor, the ratio of the oscillation energy function value to the preset oscillation energy threshold is first calculated. This ratio reflects the relative magnitude of the current oscillation intensity with respect to the alarm level. The ratio is multiplied by a sensitivity coefficient, which can range from 1 to 5, to obtain the product. This product is then used as the independent variable of the hyperbolic tangent function to determine the soft-limiting function value. The hyperbolic tangent function has saturation characteristics, which can prevent the virtual resistor from undergoing abrupt over-adjustment during severe oscillations, thereby reducing secondary impacts on the system. A sensitivity coefficient less than 1 results in a slow response, while a coefficient greater than 5 approximates a step function, easily causing the virtual resistor to jump. Therefore, a value of 1 to 5 is typically chosen in engineering practice. Subsequently, the damping difference between the target damping ratio and the current damping ratio is calculated. The target damping ratio can range from 0.2 to 0.3. The soft-limiting function value, the damping difference, and the preset virtual resistor adjustment damping coefficient are multiplied to obtain the virtual resistor adjustment increment. The virtual resistance damping coefficient can be adjusted based on the voltage sag tolerance and the measured minimum damping ratio under the worst operating conditions. For example, it can be obtained by dividing the difference between the maximum permissible virtual resistance and the steady-state virtual resistance by the difference between the target damping ratio and the minimum damping ratio. The maximum permissible virtual resistance can be determined by the system's allowable voltage sag tolerance, for example, a voltage sag of 2% of the rated voltage, divided by the effective value of the rated current to obtain the maximum virtual resistance. Finally, the steady-state virtual resistance is added to the virtual resistance adjustment increment to obtain the adaptive virtual resistance. The steady-state virtual resistance can range from 0.05Ω to 0.1Ω. To prevent over-adjustment of the virtual resistance, the rate of increase of the virtual resistance can be limited. For example, the absolute value of the rate of change of the virtual resistance should not exceed the maximum permissible rate of change, which can be obtained by dividing the difference between the maximum permissible virtual resistance and the steady-state virtual resistance by 0.5 seconds. When the oscillation energy function remains less than 0.2 times the oscillation energy threshold for more than 3 seconds, the adaptive virtual resistance can gradually recover to the steady-state virtual resistance at a rate of 0.1Ω / s, thereby reducing steady-state losses.

[0066] In some optional implementations of this disclosure, the above-mentioned calculation of the adaptive virtual inductance based on the rate of change and phase difference of the oscillating energy function includes: calculating the deviation between the absolute value of the phase difference and a preset phase difference threshold; multiplying the deviation by the S-curve steepness coefficient and taking the opposite number, and using the opposite number as the exponent of the natural constant to obtain the exponent value; calculating the reciprocal of the sum of the first preset constant and the exponent value to obtain the S-curve adjustment coefficient; multiplying the rate of change of the oscillating energy function, the S-curve adjustment coefficient, and the preset virtual inductance adjustment gain coefficient to obtain the virtual inductance adjustment increment; and adding the steady-state virtual inductance to the virtual inductance adjustment increment to obtain the adaptive virtual inductance.

[0067] The adaptive virtual inductor (which suppresses high-frequency circulating current and phase misalignment) is represented by equation (7): (t) = + (7) In equation (7), This represents the steady-state virtual inductance, with a value ranging from 0.05mH to 0.2mH; This represents the virtual inductance adjustment gain coefficient. Initial recommended value: 0.0001~0.001H; It represents the rate of change of the oscillation energy function, providing anticipatory adjustment capability; This represents the phase difference between the output current of the nth APF and the average current. This represents the phase difference threshold, and its value range is... ~ A phase difference of 5° corresponds to 0.087 rad, and a time difference of approximately 0.28 ms at 50 Hz. The deviation is detectable. A phase difference of 15° corresponds to a time of approximately 0.83 ms. This represents the S-curve adjustment coefficient, with a value range of 10 to 50. When β < 10, the transition is too smooth; when β > 50, it is easy to overflow in digital implementation, resulting in excessive inductance adjustment and causing secondary impact.

[0068] The tuning formula is shown in equation (8): = (8) In equation (8), Irated is the rated current of a single APF. The lowest oscillation period to focus on; Anti-saturation and recovery mechanisms: Ascent rate limit: | |≤

[0069] Recovery mechanism: When <0.2 Lasting more than 3 seconds Gradually recovering to at a rate of 0.1Ω / s .

[0070] In this optional implementation, for the adaptive virtual inductor, the digital controller calculates the rate of change of the oscillation energy function. This rate of change reflects the development trend of the oscillation energy; a positive value indicates that the oscillation is increasing, and a negative value indicates that the oscillation is decreasing. Simultaneously, the absolute value of the phase difference is calculated as the deviation from a preset phase difference threshold, which can range from 5° to 15°. A 5° phase difference corresponds to a time difference of approximately 0.28ms at 50Hz, which is detectable. A 15° phase difference corresponds to a time difference of approximately 0.83ms; if this value is exceeded, the system is close to instability. The deviation value is multiplied by the S-curve steepness coefficient and the inverse is taken. The S-curve steepness coefficient can range from 10 to 50. A steepness coefficient less than 10 results in a too-smooth transition, while a coefficient greater than 50 can easily overflow in digital implementation. The inverse is used as the exponent of the natural constant to obtain the exponent value. The reciprocal of the sum of the first preset constant and the exponent value is calculated to obtain the S-curve adjustment coefficient. The S-curve adjustment coefficient increases the adjustment weight as the phase difference approaches or exceeds the threshold, while automatically decreasing the weight when the phase difference is small, thus avoiding over-adjustment. The virtual inductance adjustment increment is obtained by multiplying the rate of change of the oscillation energy function, the S-curve adjustment coefficient, and the preset virtual inductance adjustment gain coefficient. The initial suggested value for the virtual inductance adjustment gain coefficient can be 0.0001H to 0.001H. The adaptive virtual inductance is obtained by adding the steady-state virtual inductance to the virtual inductance adjustment increment. The steady-state virtual inductance can range from 0.05mH to 0.2mH. Since the rate of change of oscillation energy is positive during the oscillation development stage, the adaptive virtual inductance increases accordingly, thereby injecting inductive damping in advance and achieving proactive suppression of phase misalignment and circulating current.

[0071] In this optional implementation, after the controller detects a phase difference between the output current of the nth APF and the average current, it first calculates the deviation of the absolute value of the phase difference relative to a preset phase difference threshold. This deviation is then multiplied by the S-curve steepness coefficient, and the inverse of this value is used as the exponent. The natural constant e is then exponentially calculated to obtain the exponent value. Subsequently, the S-curve adjustment coefficient is formed using the reciprocal of the sum of the first preset constant and the exponent value. This ensures that the adjustment weight is higher when the phase difference is closer to or exceeds the threshold, and automatically decreases when the deviation is small to avoid over-adjustment. Based on this, the rate of change of the oscillation energy function is multiplied by the S-curve adjustment coefficient and the preset virtual inductance adjustment gain coefficient to obtain the virtual inductance adjustment increment. This increment is then superimposed on the steady-state virtual inductance to obtain an adaptive virtual inductance. This allows for smoothing and limiting adaptive enhancement of the virtual inductance during the low-frequency oscillation development stage, combined with the degree of phase misalignment, to suppress circulating current and improve the stability of the parallel system.

[0072] In some optional implementations of this disclosure, the method further includes: real-time detection of the reactive power output status of multiple active power filters; in response to detecting that at least one active power filter outputs inductive reactive power and at least one active power filter outputs capacitive reactive power, and the absolute value of the total reactive power of the multiple active power filters is less than a preset proportion of the rated reactive power of a single filter, determining that the multiple active power filters have entered reactive power offset mode.

[0073] In this embodiment, the reactive power offset condition refers to a state in which some modules of multiple APFs operating in parallel output inductive reactive power while others output capacitive reactive power, and the two cancel each other out at a common coupling point, resulting in a total reactive power close to zero. Under this condition, the system's equivalent impedance and power flow are complex, easily inducing low-frequency oscillations. This disclosure automatically identifies this condition by detecting the polarity difference of the reactive power of each module and whether the absolute value of the total reactive power is lower than a preset proportion of the rated reactive power of a single unit.

[0074] In this embodiment, the digital controller can also monitor the reactive power output status of each active power filter in real time. When it is detected that some active power filters are outputting inductive reactive power and some are outputting capacitive reactive power, and the absolute value of the total reactive power of multiple active power filters is less than a preset proportion of the rated reactive power of a single filter (e.g., the absolute value of the total reactive power is less than 20% of the rated reactive power of a single filter), the system is determined to have entered a reactive power suppression condition. Under this condition, the digital controller can automatically execute enhanced regulation measures, such as temporarily increasing the target damping ratio by 0.05 to 0.1, increasing the virtual resistance adjustment damping coefficient by 1.2 to 1.5 times, and reducing the phase difference threshold to 3° to 8° to provide stronger damping suppression capability. After the reactive power suppression condition ends, the above-mentioned enhanced regulation parameters can be gradually restored to normal values. The restoration process can use first-order inertial filtering, and the restoration time constant can be selected between 5 seconds and 7 seconds.

[0075] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides an embodiment of an active power filter virtual impedance control device, which is similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to terminal devices.

[0076] like Figure 2As shown, the active power filter virtual impedance control device 200 provided in this embodiment includes: an acquisition unit 201, an extraction unit 202, a construction unit 203, an estimation unit 204, an inductance calculation unit 205, an impedance calculation unit 206, and a modulation unit 207. The acquisition unit 201 can be configured to acquire the output current and common coupling point voltage of each active power filter among multiple active power filters. The extraction unit 202 can be configured to extract the low-frequency oscillation components in the output current and common coupling point voltage using a bandpass filter with adaptive center frequency adjustment, multiply the low-frequency oscillation components of the output current with a quadrature reference signal of the same frequency, extract the phase of the output current of each active power filter, and calculate the phase difference between the phase of the output current and the phase of the global average current. The construction unit 203 can be configured to calculate the active circulating current deviation and reactive circulating current deviation of each active power filter based on the low-frequency oscillation components of the output current and common coupling point voltage, and construct an oscillation energy function based on the active circulating current deviation and reactive circulating current deviation. The estimation unit 204 can be configured to measure the open-loop frequency response of multiple active power filters by injecting a disturbance signal into the system, and to estimate the current damping ratio of the multiple active power filters online based on the open-loop frequency response. The inductance calculation unit 205 can be configured to calculate an adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio, and to calculate an adaptive virtual inductance based on the rate of change and phase difference of the oscillation energy function. The impedance calculation unit 206 can be configured to calculate an adaptive virtual impedance based on the adaptive virtual resistance and the adaptive virtual inductance. The modulation unit 207 can be configured to multiply the adaptive virtual impedance by the output current and then superimpose it onto the original voltage reference command to generate the final modulation voltage.

[0077] In this embodiment, the specific processing of the following components in the active power filter virtual impedance control device 200, namely the acquisition unit 201, extraction unit 202, construction unit 203, estimation unit 204, inductance calculation unit 205, impedance calculation unit 206, and modulation unit 207, and their resulting technical effects, can be found in reference [reference needed]. Figure 1 The relevant descriptions of steps 101, 102, 103, 104, 105, 106, and 107 in the corresponding embodiments will not be repeated here.

[0078] In some embodiments of this disclosure, the extraction unit 202 is configured to: set the initial center frequency of the bandpass filter with adaptive center frequency adjustment; send the extracted signal to a frequency detector; and adjust the center frequency of the bandpass filter in real time through a proportional-integral regulator so that the center frequency tracks the main oscillation frequency in the signal to obtain the low-frequency oscillation component in the output current and the common coupling point voltage.

[0079] In some embodiments of this disclosure, the above-mentioned building unit 203 is configured to integrate the product of active power circulation deviation and reactive power circulation deviation within a time window to construct an oscillation energy function.

[0080] In some embodiments of this disclosure, the estimation unit 204 is configured to: inject a periodic disturbance signal into the system, the amplitude of which is set within a first preset range of the rated reactive power of the plurality of active power filters; use the disturbance signal as the system input and the measured common coupling point voltage response as the system output to obtain the open-loop frequency response of the plurality of active power filters; and calculate the current damping ratio based on the phase margin and correction coefficient of the open-loop frequency response.

[0081] In some embodiments of this disclosure, the inductance calculation unit 205 is configured to: calculate the ratio of the value of the oscillation energy function to a preset oscillation energy threshold; multiply the ratio by a sensitivity coefficient to obtain a product, and use the product as the independent variable of the hyperbolic tangent function to obtain the soft-limiting function value; calculate the damping difference between the target damping ratio and the current damping ratio; multiply the soft-limiting function value, the damping difference, and a preset virtual resistance adjustment damping coefficient to obtain the virtual resistance adjustment increment; and add the steady-state virtual resistance to the virtual resistance adjustment increment to obtain the adaptive virtual resistance.

[0082] In some embodiments of this disclosure, the inductance calculation unit 205 is further configured to: calculate the deviation between the absolute value of the phase difference and a preset phase difference threshold; multiply the deviation by the S-curve steepness coefficient and take the opposite number, and use the opposite number as the exponent of the natural constant to obtain the exponent value; calculate the reciprocal of the sum of the first preset constant and the exponent value to obtain the S-curve adjustment coefficient; multiply the rate of change of the oscillation energy function, the S-curve adjustment coefficient, and the preset virtual inductance adjustment gain coefficient to obtain the virtual inductance adjustment increment; and add the steady-state virtual inductance to the virtual inductance adjustment increment to obtain the adaptive virtual inductance.

[0083] In some embodiments of this disclosure, the above-mentioned active power filter virtual impedance control device 200 further includes: a detection unit (not shown in the figure), which is configured to: detect the reactive power output status of multiple active power filters in real time; and determine that multiple active power filters have entered reactive power matching mode when it is detected that at least one active power filter outputs inductive reactive power, at least one active power filter outputs capacitive reactive power, and the absolute value of the total reactive power of multiple active power filters is less than a preset ratio of the rated reactive power of a single filter.

[0084] The active power filter virtual impedance control device provided in the embodiment extracts the low-frequency oscillation component of the parallel APF online and estimates the current damping ratio. Then, it adaptively adjusts the virtual resistance and virtual inductance and injects positive damping virtual impedance to quickly suppress low-frequency oscillations under multi-machine parallel operation (especially reactive power supply operation), while taking into account steady-state voltage drop and compensation accuracy.

[0085] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0086] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0087] Figure 3 A schematic block diagram of an example electronic device 300 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their patterns are merely examples and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0088] like Figure 3 As shown, the electronic device 300 includes a computing unit 301, which can perform various appropriate actions and processes based on a computer program stored in a read-only memory (ROM) 302 or a computer program loaded from a storage unit 308 into a random access memory (RAM) 303. The RAM 303 may also store various programs and data required for the operation of the electronic device 300. The computing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0089] Multiple components in electronic device 300 are connected to I / O interface 305, including: input unit 306, such as keyboard, mouse, etc.; output unit 307, such as various types of displays, speakers, etc.; storage unit 308, such as disk, optical disk, etc.; and communication unit 309, such as network card, modem, wireless transceiver, etc. Communication unit 309 allows electronic device 300 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] The computing unit 301 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 301 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 301 performs the various methods and processes described above, such as the active power filter virtual impedance control method. For example, in some embodiments, the active power filter virtual impedance control method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 308. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 300 via ROM 302 and / or communication unit 309. When the computer program is loaded into RAM 303 and executed by the computing unit 301, one or more steps of the active power filter virtual impedance control method described above can be performed. Alternatively, in other embodiments, the computing unit 301 may be configured to perform an active power filter virtual impedance control method by any other suitable means (e.g., by means of firmware).

[0091] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0092] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable active power filter virtual impedance control device, such that when executed by the processor or controller, the patterns / operations specified in the flowcharts and / or block diagrams are implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0093] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0094] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0095] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0096] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0097] The foregoing description of specific exemplary embodiments of this disclosure is for illustrative and explanatory purposes. These descriptions are not intended to limit this disclosure to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of this disclosure and their practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of this disclosure, as well as various different choices and variations. The scope of this disclosure is intended to be defined by the claims and their equivalents.

Claims

1. A method for virtual impedance control of an active power filter, characterized in that, The method, applied to multiple active power filters connected in parallel via a common coupling point, includes: Obtain the output current and common coupling point voltage of each of the plurality of active power filters; The low-frequency oscillation components in the output current and the common coupling point voltage are extracted by a bandpass filter with adaptive center frequency. The low-frequency oscillation component of the output current is multiplied by a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter. The phase difference between the phase of the output current and the phase of the global average current is calculated. Based on the low-frequency oscillation components of the output current and the common coupling point voltage, the active current deviation and reactive current deviation of each active power filter are calculated, and an oscillation energy function is constructed based on the active current deviation and the reactive current deviation. By injecting a disturbance signal into the system, the open-loop frequency response of the plurality of active power filters is measured, and the current damping ratio of the plurality of active power filters is estimated online based on the open-loop frequency response. Based on the oscillation energy function, the target damping ratio, and the current damping ratio, the adaptive virtual resistance is calculated, and based on the rate of change of the oscillation energy function and the phase difference, the adaptive virtual inductance is calculated. Based on the adaptive virtual resistance and the adaptive virtual inductance, calculate the adaptive virtual impedance; The adaptive virtual impedance is multiplied by the output current and then superimposed onto the original voltage reference command to generate the final modulation voltage.

2. The method according to claim 1, characterized in that, The bandpass filter, which adaptively adjusts its center frequency, extracts the low-frequency oscillation components from the output current and the common coupling point voltage, respectively. Set the initial center frequency of the bandpass filter with adaptive center frequency adjustment; The extracted signal is sent to a frequency detector, and the center frequency of the bandpass filter is adjusted in real time by a proportional-integral regulator so that the center frequency tracks the main oscillation frequency in the signal, thereby obtaining the low-frequency oscillation component in the output current and the common coupling point voltage.

3. The method according to claim 1, characterized in that, The construction of the oscillation energy function based on the active power circulation deviation and the reactive power circulation deviation includes: The product of the active power circulation deviation and the reactive power circulation deviation is integrated within a time window to construct an oscillation energy function.

4. The method according to claim 1, characterized in that, The step of injecting a disturbance signal into the system, measuring the open-loop frequency response of the plurality of active power filters, and estimating the current damping ratio of the plurality of active power filters online based on the open-loop frequency response includes: A periodic disturbance signal is injected into the system, the amplitude of which is set within a first preset range of the rated reactive power of the plurality of active power filters; The disturbance signal is used as the input of the system, and the measured voltage response at the common coupling point is used as the output of the system to obtain the open-loop frequency response of the plurality of active power filters; The current damping ratio is calculated based on the phase margin and correction coefficient of the open-loop frequency response.

5. The method according to claim 1, characterized in that, The calculation of the adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio includes: Calculate the ratio of the value of the oscillation energy function to the preset oscillation energy threshold; Multiply the ratio by the sensitivity coefficient to obtain the product, and use the product as the independent variable of the hyperbolic tangent function to obtain the soft limiting function value; Calculate the damping difference between the target damping ratio and the current damping ratio; Multiply the soft limiting function value, the damping difference, and the preset virtual resistance adjustment damping coefficient to obtain the virtual resistance adjustment increment; The adaptive virtual resistance is obtained by adding the steady-state virtual resistance to the virtual resistance adjustment increment.

6. The method according to claim 1, characterized in that, The calculation of the adaptive virtual inductance based on the rate of change of the oscillating energy function and the phase difference includes: Calculate the deviation between the absolute value of the phase difference and a preset phase difference threshold; Multiply the deviation value by the steepness coefficient of the S-curve and take the opposite number. Use the opposite number as the exponent of the natural constant to obtain the exponent value. Calculate the reciprocal of the sum of the first preset constant and the exponential value to obtain the S-curve adjustment coefficient; The virtual inductance adjustment increment is obtained by multiplying the rate of change of the oscillation energy function, the S-curve adjustment coefficient, and the preset virtual inductance adjustment gain coefficient. The adaptive virtual inductance is obtained by adding the steady-state virtual inductance to the virtual inductance adjustment increment.

7. The method according to claim 1, characterized in that, The method further includes: Real-time detection of the reactive power output status of the multiple active power filters; In response to the detection that at least one of the plurality of active power filters outputs inductive reactive power and at least one of the active power filters outputs capacitive reactive power, and the absolute value of the total reactive power of the plurality of active power filters is less than a preset proportion of the rated reactive power of a single unit, it is determined that the plurality of active power filters have entered reactive power offset mode.

8. A virtual impedance control device for an active power filter, characterized in that, The device, applicable to multiple active power filters connected in parallel via a common coupling point, comprises: The acquisition unit is configured to acquire the output current and common coupling point voltage of each of the plurality of active power filters; The extraction unit is configured to extract the low-frequency oscillation components in the output current and the common coupling point voltage through a bandpass filter with adaptive center frequency, and multiply the low-frequency oscillation components of the output current with a quadrature reference signal of the same frequency to extract the phase of the output current of each active power filter, and calculate the phase difference between the phase of the output current and the phase of the global average current. The construction unit is configured to calculate the active current deviation and reactive current deviation of each active power filter based on the low-frequency oscillation components of the output current and the common coupling point voltage, and to construct the oscillation energy function based on the active current deviation and the reactive current deviation. The estimation unit is configured to measure the open-loop frequency response of the plurality of active power filters by injecting a disturbance signal into the system, and to estimate the current damping ratio of the plurality of active power filters online based on the open-loop frequency response. The inductance calculation unit is configured to calculate an adaptive virtual resistance based on the oscillation energy function, the target damping ratio, and the current damping ratio, and to calculate an adaptive virtual inductance based on the rate of change of the oscillation energy function and the phase difference. An impedance calculation unit is configured to calculate an adaptive virtual impedance based on the adaptive virtual resistance and the adaptive virtual inductance; The modulation unit is configured to multiply the adaptive virtual impedance by the output current and then superimpose it onto the original voltage reference command to generate the final modulation voltage.

9. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-7.

Images