A method for suppressing grid-connected current harmonics of an energy storage converter

Abstract

The present application belongs to the technical field of energy storage converter control, and particularly relates to a method for suppressing grid-connected current harmonics of an energy storage converter, comprising: obtaining a three-phase grid voltage sampling signal to solve a real-time angular frequency, and reconstructing a center resonance angular frequency of a harmonic extraction channel; removing a transfer function proportional term and combining a bandwidth angular frequency to update channel coefficients, and separating real-time estimated values of each harmonic component; calculating a phase lead angle based on the number of harmonics, the real-time angular frequency, a control period and total delay taps of a digital link, performing a rotation transformation on the real-time estimated values to generate compensation components; superimposing each compensation component to form a total harmonic compensation amount, and superimposing the total harmonic compensation amount into a given channel of a fundamental current inner loop after being proportionally reduced based on an allowed upper limit to drive output. The present application eliminates the extraction amplitude attenuation when the frequency drifts, and avoids the risk of harmonic reverse amplification caused by the phase misalignment of the compensation amount.

Description

Technical Field

[0001] This invention belongs to the field of energy storage converter control technology, specifically relating to a method for suppressing harmonics in grid-connected current of an energy storage converter. Background Technology

[0002] As the core power conversion interface between energy storage systems and the power grid, energy storage converters are widely used in scenarios such as peak shaving and valley filling, frequency regulation and peak shaving, and grid-connected and off-grid power supply. In grid-connected operation mode, the energy storage converter operates as a current source, and the power quality of the grid-connected current directly affects the safe and stable operation of the power grid. Due to the interaction of high-frequency switching of power devices, DC bus voltage fluctuations, and background harmonics on the grid side, the grid-connected current inevitably contains low-order harmonic components such as the 3rd, 5th, 7th, 9th, and 11th harmonics. These harmonics can cause the harmonic distortion rate of the grid current to exceed the standard, which can harm electrical equipment connected to the same grid connection point.

[0003] To address the aforementioned harmonic issues, existing technologies generally employ harmonic extraction and compensation schemes based on resonant filters or quasi-proportional resonant controllers. The basic idea is to sample the grid-connected current in real time, use a resonant element at a specific harmonic frequency to separate the harmonic components from the sampled signal, and then superimpose the extracted harmonic quantities in reverse phase onto the given value of the inner current loop. Closed-loop regulation forces the converter to output an inverse harmonic current, thereby canceling the original harmonic components in the grid-connected current. This scheme has a clear structure, is easy to implement digitally, and achieves good harmonic suppression under steady-state power frequency conditions. It has already been put into use in some commercial energy storage converter products.

[0004] However, since the actual power grid is not an ideal constant-frequency system, especially in weak grids, areas with a high proportion of renewable energy penetration, and scenarios with drastic load switching, the fundamental frequency of the power grid will continuously drift around the rated value. The center resonant frequency of the existing harmonic extraction stage is pre-set as an integer multiple of the rated power frequency. When the fundamental frequency drifts, the actual frequencies of each harmonic shift out of the passband center of the resonant stage, resulting in significant amplitude attenuation and phase lag in the extracted harmonic components. More importantly, the inherent sampling and pulse width modulation delays of the digital control system of the energy storage converter are further superimposed on this phase deviation, causing the compensation amount injected into the current loop to no longer satisfy the 180-degree out-of-phase relationship with the actual harmonics. In some frequency offset ranges, the compensation amount actually shows a trend of being in phase with the original harmonics, which not only fails to suppress harmonics but also amplifies the harmonic content of the grid-connected current and may even induce current loop oscillations. Summary of the Invention

[0005] This invention provides a method for suppressing harmonics in grid-connected current of an energy storage converter, in order to solve the technical problem that the harmonic compensation amplitude attenuation and phase inaccuracy lead to compensation failure or even reverse amplification under the combined effect of grid frequency drift and digital control delay.

[0006] This invention provides a method for suppressing harmonics in the grid-connected current of an energy storage converter, comprising the following steps: S1. Obtain the three-phase grid voltage sampling signal at the grid connection point of the energy storage converter, perform phase-locked loop operation to calculate the real-time angular frequency, and reconstruct the center resonant angular frequency of the corresponding harmonic extraction channel based on the real-time angular frequency and the number of harmonics to be suppressed. S2, remove the proportional term in the transfer function of the harmonic extraction channel, substitute the center resonant angular frequency into the resonant term, and combine the preset bandwidth angular frequency to discretize and update the channel coefficients, thereby separating the real-time estimated values ​​of each harmonic component in the grid-connected current. S3, based on the harmonic order, real-time angular frequency, control period and pre-set total delay of the digital link, calculates the phase lead angle of the corresponding harmonic, and uses the phase lead angle to perform a rotation transformation on the real-time estimate to generate the phase-corrected compensation component. S4, superimpose the compensation components corresponding to each harmonic to form the total harmonic compensation amount, determine the allowable upper limit based on the rated current amplitude of the converter and the actual amplitude of the current fundamental current, reduce the total harmonic compensation amount proportionally when it reaches the allowable upper limit, and superimpose the final total harmonic compensation amount onto the given channel of the fundamental current inner loop to drive the output compensation current.

[0007] Its effect is that by driving the center frequency of each harmonic extraction channel in real time through the phase-locked loop, the harmonic extraction link always maintains accurate alignment with the actual harmonic frequency under the condition of power grid frequency drift, eliminating the problem of extraction amplitude attenuation that occurs when the frequency is deviated in the existing fixed resonant frequency scheme, and ensuring the integrity of harmonic component extraction in weak power grid and load fluctuation scenarios.

[0008] Further, the three-phase grid voltage sampling signal at the energy storage converter's grid connection point is acquired, and a phase-locked loop operation is performed to calculate the real-time angular frequency, including: The initial angular frequency is obtained by performing a phase-locked loop operation on the three-phase power grid voltage sampling signal; Calculate the change in the initial angular frequency in the current control cycle relative to the previous control cycle; When the change exceeds the preset amplitude limit ratio, the initial angular frequency is amplitude limited using the boundary value corresponding to the amplitude limit ratio, and the real-time angular frequency is finally obtained.

[0009] Its effect is as follows: by judging the periodic change of the initial angular frequency output by the phase-locked loop, and performing amplitude limiting processing when the angular frequency change exceeds the limit, the real-time angular frequency can be obtained. This can suppress the angular frequency jump caused by drastic changes in the power grid frequency and the disturbance of the phase-locked loop, prevent the numerical change of the center resonant angular frequency of the subsequent harmonic extraction channel, avoid the drastic refresh of the channel coefficient causing the harmonic extraction result to jump, and improve the stability of the entire harmonic suppression algorithm in the scenario of drastic fluctuations in the power grid frequency.

[0010] Furthermore, based on the real-time angular frequency and combined with the harmonic orders of the target suppression, the center resonant angular frequency of the corresponding harmonic extraction channel is reconstructed, including: Determine the order of each harmonic that needs to be suppressed; Each harmonic order is multiplied by the real-time angular frequency. The result of the multiplication operation is used as the center resonant angular frequency of the corresponding subharmonic extraction channel, and finally the center resonant angular frequency that slides synchronously with the grid frequency is formed.

[0011] Its effect is that by multiplying each harmonic order by the real-time angular frequency, the passband of each channel always covers the actual harmonic frequency, thus eliminating the attenuation of the extracted amplitude caused by frequency offset from the source.

[0012] Furthermore, the proportional term in the harmonic extraction channel transfer function is removed, and the center resonant angular frequency is substituted into the resonant term. The channel coefficients are then updated using a preset bandwidth angular frequency discretization method, including: Obtain the maximum allowable offset of the power grid frequency; The corresponding angular frequency span is calculated based on the maximum allowable offset of the power grid frequency, and the angular frequency span is used as the bandwidth angular frequency. Substituting the bandwidth angular frequency and the center resonant angular frequency into the transfer function after removing the proportional term, and then mapping it to the discrete domain using a bilinear transformation, finally the discretized updated channel coefficients are obtained.

[0013] Furthermore, based on the harmonic order, real-time angular frequency, control period, and pre-set total digital link delay, the phase lead angle of the corresponding harmonic is calculated, including: Inject a small-amplitude single-frequency test signal with known phase into the current loop, measure the number of control cycles corresponding to the phase lag at the output end and round it to obtain the total delay time of the digital link; The harmonic angular velocity is obtained by multiplying the harmonic order by the real-time angular frequency. The delay time is obtained by multiplying the control cycle by the total delay time of the digital link. Multiplying the harmonic angular velocity by the delay time yields the phase lead angle of the corresponding harmonic.

[0014] Its effect is that by introducing phase advance reconstruction that adaptively adjusts with real-time frequency and harmonic order, the compensation amount of the final injected current loop maintains an accurate anti-phase relationship with the real harmonic under any frequency deviation condition, thus avoiding the risk of harmonic reverse amplification and current loop oscillation caused by phase inaccuracy of the compensation amount.

[0015] Furthermore, a rotational transformation is performed on the real-time estimated value using the phase lead angle to generate a phase-corrected compensation component, including: Transform the real-time estimated values ​​into a two-phase stationary coordinate system; In a two-phase stationary coordinate system, the real-time estimated value is subjected to coordinate rotation transformation according to the phase lead angle; The components after rotation transformation are extracted, and finally the phase-corrected compensation components are generated.

[0016] Furthermore, the allowable upper limit is determined based on the rated current amplitude of the converter and the actual amplitude of the current fundamental current, including: Read the preset rated current amplitude of the energy storage converter; Real-time sampling to obtain the actual amplitude of the current fundamental current; Calculate the difference between the rated current amplitude of the converter and the actual amplitude of the current fundamental current, and finally determine the difference as the allowable upper limit.

[0017] Furthermore, when the total harmonic compensation reaches the allowable upper limit, a proportional reduction is allocated, including: Calculate the sum of the real-time estimated values ​​of each harmonic in all enabled channels; Calculate the ratio of the amplitude of the real-time estimate to the sum of the amplitudes in a single channel; Multiplying the ratio by the allowable upper limit yields the proportionally reduced compensation component for each channel.

[0018] Its effect is that, through amplitude constraints based on the remaining capacity boundary and a reduction mechanism based on proportional allocation, it prioritizes the suppression of the highest harmonic content under extreme operating conditions such as sudden load changes, while ensuring the stability of fundamental power transmission.

[0019] Furthermore, the final total harmonic compensation is superimposed onto the fundamental current inner loop given channel to drive the output compensation current, including: Obtain the fundamental current command given by the power dispatcher; The actual setpoint of the current loop is obtained by subtracting the final total harmonic compensation from the fundamental current command. The actual setpoint is input into the current loop regulator, which ultimately drives the power device to output a compensation current that is out of phase with the original harmonics at the grid connection point.

[0020] Further, the three-phase grid voltage sampling signal at the grid connection point of the energy storage converter is acquired, including: At the start of each current loop control cycle of the energy storage converter, the analog-to-digital converter is triggered to sample. Read the digital signal output by the analog-to-digital converter; The digital signal is filtered to obtain the three-phase power grid voltage sampling signal.

[0021] The beneficial effects are: This invention addresses the differential phase lag caused by digital control link delay to harmonics of different orders by introducing phase advance reconstruction that adaptively adjusts with real-time frequency and harmonic order. This ensures that the compensation amount of the final injected current loop maintains an accurate anti-phase relationship with the real harmonic under any frequency offset condition, significantly improving the effectiveness and stability of harmonic suppression function across the entire frequency offset range.

[0022] This invention is based on the compensation amplitude constraint of the objective boundary of the remaining capacity of the converter. While ensuring the harmonic suppression effect, it also ensures the priority of fundamental power transmission. This allows the solution to be directly embedded into the existing energy storage converter control software architecture without hardware modification, effectively reducing the harmonic distortion rate of the grid-connected current and improving the power quality of the energy storage system connected to the grid. Attached Figure Description

[0023] Figure 1 This is a flowchart of the method for suppressing harmonics in the grid-connected current of the energy storage converter in this invention.

[0024] Figure 2 This is a comparison diagram of the amplitude shape of harmonic extraction under frequency drift in this invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] An embodiment of the grid-connected current harmonic suppression method for energy storage converters provided by this invention: like Figure 1 As shown, the method for suppressing harmonics in the grid-connected current of an energy storage converter includes the following steps: S1 calculates the real-time angular frequency and reconstructs the center resonant angular frequency.

[0027] At the start of each current loop control cycle of the energy storage converter, the analog-to-digital converter (ADC) is triggered to sample the data. The digital signal output by the ADC is read, filtered, and the three-phase grid voltage sampling signal is obtained. A phase-locked loop (PLL) operation is then performed on this sampling signal to calculate the initial angular frequency and real-time phase angle of the grid fundamental frequency within the current control cycle. The change in the initial angular frequency in the current control cycle relative to the previous control cycle is calculated. When the change exceeds a preset limiting ratio, the initial angular frequency is limited using the boundary value corresponding to the limiting ratio, ultimately obtaining the real-time angular frequency. Unlike existing solutions that use phase-locked loop (PLL) results only for grid-connected current orientation, this invention uses real-time angular frequency... As the frequency reference source for all subsequent harmonic extraction channels, the frequency characteristics of the harmonic extraction link slide synchronously with the power grid frequency.

[0028] Harmonic order suppression for the target ( Using low-order odd harmonics (such as 3rd, 5th, 7th, 9th, and 11th harmonics), the harmonic orders requiring harmonic suppression are determined, and an independent extraction channel is constructed for each harmonic. Each harmonic order is multiplied by the real-time angular frequency, and the result of this multiplication is used as the center resonant angular frequency of the corresponding harmonic extraction channel. This ultimately forms the center resonant angular frequency that slides synchronously with the grid frequency, which is set as [value missing]. Instead of a fixed constant based on the rated power frequency, the actual frequencies of different harmonics are scaled down to integer multiples of the fundamental frequency. When the fundamental frequency of the power grid drifts between 49.5Hz and 50.5Hz, the actual frequency of the 5th harmonic changes continuously between 247.5Hz and 252.5Hz. If the center resonant frequency of the extracted channel is still fixed at 250Hz, the frequency deviation at the drift boundary can reach 2.5Hz. For a high-Q resonant circuit, this deviation is sufficient to cause a significant amplitude drop. Changing the center frequency to... Afterwards, the passband of each channel always covers the actual harmonic frequency, eliminating the attenuation of the extracted amplitude caused by frequency offset from the source.

[0029] Furthermore, at the hardware execution level, the digital signal processor of the energy storage converter first updates the phase-locked loop calculation once in each current loop control cycle, and then uses the updated... The discretization coefficients of each harmonic extraction channel are refreshed. To avoid numerical abrupt changes introduced by periodically refreshing the frequency coefficients, an amplitude limit constraint is applied to the updated angular frequency value: the change in angular frequency within a single period must not exceed a preset proportion of the previous period's value. The derivation of this amplitude limit proportion is based on the following: GB / T 15945 stipulates that the rate of change of power grid frequency deviation shall not exceed... Hz / s, in the control cycle Under the corresponding typical engineering parameters, the maximum angular frequency change within a single period is ,in The upper limit of the rate of change of frequency deviation as specified in the national standard, relative to the rated angular frequency. The proportion is directly determined by the above formula. For example, when... Hz / s At ms, this ratio is approximately the rated value. The upper limit of the amplitude is set to a preset ratio of the rated value after taking into account the actual deviation of the project. For example, the preset ratio can be 0.1% of the rated value.

[0030] Thus, the real-time angular frequency output by the phase-locked loop... It serves as both the fundamental frequency orientation reference and the main frequency control source for each harmonic extraction channel. Both share the same phase-locked loop result, avoiding phase accumulation errors between channels caused by inconsistent frequency references, and laying a unified frequency foundation for the accurate separation of subsequent harmonics.

[0031] S2, discretizes and updates the channel coefficients and separates the real-time estimates.

[0032] The continuous domain transfer function of the existing quasi-proportional resonant controller is

[0033] in For proportional gain, For resonant gain, The cutoff bandwidth angular frequency, The resonant center angular frequency, It is a Laplace complex variable. Because of the denominator... As a fixed constant that is an integer multiple of the rated power frequency, when the fundamental frequency drifts, the resonant peak remains pinned to the multiple of the rated frequency and cannot slide with the actual harmonic frequency. This is the direct structural root cause of inaccurate extraction during frequency drift. Furthermore, The proportional term introduces pass-through gain to both the DC and fundamental components. In the harmonic extraction channel, this can introduce non-target frequency components into the output, interfering with the purity of harmonic separation. Therefore, the harmonic extraction channel should remove these components. Only the resonant terms are retained.

[0034] Based on the above analysis, this invention makes targeted improvements to the transfer function: on the one hand, the fixed resonant center angular frequency is... Replace with the already calculated real-time angular frequency With harmonic order On the other hand, removing the proportional term in the harmonic extraction channel transfer function. Furthermore, by substituting the center resonant angular frequency into the resonant term, the channel exhibits frequency selectivity only for the target harmonic frequency. The improved [method / mechanism]... The transfer function of the subharmonic extraction channel is:

[0035] Compare the original transfer function The improvement is reflected in two aspects: the fixed constant in the denominator resonance term. Replace with periodic updates This ensures that the channel peak gain point is always pinned to the actual harmonic frequency; at the same time, it removes... This eliminates the pass-through gain in non-target frequency bands. The maximum permissible offset of the power grid frequency is obtained. Since GB / T 15945 specifies the maximum permissible offset of the power grid frequency, the angular frequency span corresponding to this maximum offset is taken as the bandwidth angular frequency. ,in This is the maximum frequency deviation specified by the standard, thus ensuring that the actual harmonic frequency still falls within the passband even under extreme frequency deviations.

[0036] Next, the bandwidth angular frequency and the center resonant angular frequency are substituted into the transfer function after removing the proportional term, and a bilinear transform is used to map them to the discrete domain, finally obtaining the discretized updated channel coefficients. A bilinear transformation is used to map to the discrete domain. Because... The discretization coefficients change periodically, and within each control cycle, they are based on the current... Recalculate, with parallel operation of each harmonic channel, the 1st The output of each channel is a real-time estimate of the harmonic components separated from the grid-connected current. Each channel operates independently, and channels with higher harmonic content can be selectively enabled based on measured harmonic spectrum results at the grid connection point. Unenabled channels do not participate in the calculation, avoiding unnecessary consumption of processor power. It should be noted that there are many other discretization methods in this field capable of real-time separation of harmonic components. Those skilled in the art can choose different discretization transformation methods according to actual needs, and are not limited to bilinear transformations.

[0037] Thus, each harmonic extraction channel uses the actual harmonic frequency as the peak gain point under any frequency offset condition, outputting real-time estimated values. The amplitude accurately reflects the true content of the corresponding subharmonic in the grid-connected current, providing a precise input basis for the subsequent phase compensation stage.

[0038] S3, calculate the phase lead angle and generate the corrected compensation component.

[0039] Real-time estimated value extracted from the resonant circuit While the amplitude is aligned with the true harmonics, there is a complete digital control link delay between the current sampling and the actual output of the compensation quantity via pulse width modulation. This delay mainly includes the sample-and-hold delay and the modulation update delay, totaling approximately 1.5 control cycles in engineering. Since the phase lag generated after the signal passes through the delay stage is equal to the product of the signal angular velocity and the delay time, and different harmonics have different angular velocities, the phase lag experienced by each harmonic is different: for the fundamental frequency, the lag angle is less than 5 degrees and can be ignored, but for the 11th harmonic, at a typical 10kHz switching frequency, the lag angle is close to 60 degrees. If directly injected in reverse phase, the phase of the synthesized compensation quantity will deviate significantly from the ideal inversion point.

[0040] To address this, the present invention introduces a phase lead reconstruction stage, which calculates the phase lead angle of the corresponding harmonic based on the harmonic order, real-time angular frequency, control period, and pre-set total delay of the digital link.

[0041] in For the first Phase lead angle of subharmonics For harmonic order, For the real-time fundamental angular frequency, For the current loop control cycle, This refers to the total delay steps of the digital link, measured in control cycles. Specifically, a small-amplitude single-frequency test signal with known phase is injected into the current loop. The number of control cycles corresponding to the output phase lag is measured and rounded to obtain the total delay steps of the digital link. The harmonic angular velocity is obtained by multiplying the harmonic order by the real-time angular frequency. The delay time is obtained by multiplying the control cycle by the total delay steps of the digital link. The phase lead angle of the corresponding harmonic is finally obtained by multiplying the harmonic angular velocity by the delay time. This relationship is directly derived from the phase lag relationship of the delay element. , of which The angular velocity of the subharmonic is The delay time is The product of the two is the leading angle that needs to be compensated in advance. Instead of making any assumptions, the values ​​are obtained through frequency sweep measurements during the converter's factory setting phase and then permanently written into the control parameter table as constants after setting.

[0042] Furthermore, a rotation transformation is performed on the real-time estimated value using the phase lead angle to generate the phase-corrected compensation component. Specifically, the real-time estimated value is transformed to a two-phase stationary coordinate system. In this system, a coordinate rotation transformation is performed on the real-time estimated value according to the phase lead angle. The rotated components are then extracted, ultimately generating the phase-corrected compensation component. .because As the power grid frequency changes in real time, It also adaptively adjusts with frequency deviation. A larger value indicates a higher grid frequency, a greater harmonic angular velocity, and a more severe phase lag caused by delay, resulting in a synchronous increase in the lead compensation angle; a smaller value indicates a lower grid frequency, and the lead compensation angle decreases accordingly. Thus, under any frequency deviation condition, it ensures that the compensation amount given by the final injected current loop maintains a precise 180-degree out-of-phase relationship with the actual harmonics in the grid-connected current, completely avoiding the risk of harmonic reverse amplification caused by inaccurate out-of-phase compensation.

[0043] Thus, the compensation component after phase-lead reconstruction Simultaneously, the precise alignment with the actual harmonics in both amplitude and phase dimensions provides a compensating input for the subsequent superposition and injection stage without amplitude attenuation or phase misalignment. This is the core guarantee for the invention to achieve effective harmonic suppression across the entire frequency offset range.

[0044] S4, superimposes the total harmonic compensation and proportionally distributes the reduced drive output.

[0045] The compensation components corresponding to each harmonic Linear superposition is performed to form the total harmonic compensation. The allowable upper limit is determined based on the rated current amplitude of the converter and the actual amplitude of the current fundamental current. Specifically, the rated current amplitude of the energy storage converter is read from the preset value. Real-time sampling to obtain the actual amplitude of the current fundamental current Calculate the difference between the rated current amplitude of the converter and the actual amplitude of the current fundamental current, and determine the allowable upper limit using the following formula. :

[0046] This boundary is objectively determined by the equipment's rated parameters and real-time sampling, without introducing any subjective empirical coefficients. It is reduced proportionally when the total harmonic compensation reaches the allowable upper limit. Specifically, the sum of the amplitudes of the real-time estimates corresponding to each harmonic in all enabled channels is calculated. Calculate the amplitude of the real-time estimate in a single channel. The ratio of the sum of amplitudes to the total amplitude is multiplied by the allowable upper limit to obtain the proportionally reduced compensation component allocated to each channel, i.e., the compensation capacity allocated to each channel. This ratio is objectively determined by the real-time output of each channel's resonant components. The channel with the highest harmonic content receives the largest allocation share, and the fundamental power transmission function is not affected by compensation actions.

[0047] The final total harmonic compensation is superimposed on the fundamental current inner loop given channel to drive the output compensation current. Specifically, the fundamental current command given by the power dispatch is obtained, the final total harmonic compensation is subtracted from the fundamental current command to obtain the actual given value of the current loop, and the actual given value is input to the current loop regulator, which ultimately drives the power device to output a compensation current that is out of phase with the original harmonics at the grid connection point, thereby achieving active harmonic cancellation.

[0048] Next, within each control cycle, the processing flow from S1 to S4 is executed in a fixed-cycle sequence: phase-locked frequency update, reconstruction of coefficients for each channel, harmonic extraction, phase lead correction, superposition injection, and amplitude constraint. The entire process is embedded in the existing current loop interrupt service routine of the energy storage converter, without altering the existing power control architecture. It is compatible with grid-connected charging, grid-connected discharging, and different load rate conditions. Upgrades during engineering deployment only require adding corresponding functional modules to the existing control software. It should be noted that there are many other ways in the field to achieve multi-channel compensation fusion and capacity constraint allocation. Those skilled in the art can choose different allocation methods according to actual needs, and are not limited to the amplitude reduction method.

[0049] Thus, by using amplitude constraints based on the remaining capacity boundary and a reduction mechanism based on proportional allocation, under extreme operating conditions such as sudden load changes, the converter prioritizes the suppression of the highest harmonic content without exceeding its own current output capacity, while ensuring the stability of fundamental power transmission, giving this solution good engineering robustness and practicality.

[0050] like Figure 2 As shown, it can be observed that when the fundamental frequency deviates from the rated value, the amplitude coefficient of the fifth harmonic extraction in the existing fixed frequency method shows a significant downward deviation. The frequency follower method of the present invention maintains a stable state close to the reference value throughout the entire frequency drift range, indicating that after the center resonant angular frequency is reconstructed with the real-time angular frequency, the passband of the extraction channel can continuously cover the actual harmonic frequency.

[0051] The above are all preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Therefore, all equivalent changes made in accordance with the structure, shape and principle of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A method for suppressing harmonics in grid-connected current of an energy storage converter, characterized in that, include: S1. Obtain the three-phase grid voltage sampling signal at the grid connection point of the energy storage converter, perform phase-locked loop operation to calculate the real-time angular frequency, and reconstruct the center resonant angular frequency of the corresponding harmonic extraction channel based on the real-time angular frequency and the number of harmonics to be suppressed. S2, remove the proportional term in the transfer function of the harmonic extraction channel, substitute the center resonant angular frequency into the resonant term, and combine the preset bandwidth angular frequency to discretize and update the channel coefficients, thereby separating the real-time estimated values ​​of each harmonic component in the grid-connected current. S3, based on the harmonic order, real-time angular frequency, control period and pre-set total delay of the digital link, calculates the phase lead angle of the corresponding harmonic, and uses the phase lead angle to perform a rotation transformation on the real-time estimate to generate the phase-corrected compensation component. S4, superimpose the compensation components corresponding to each harmonic to form the total harmonic compensation amount, determine the allowable upper limit based on the rated current amplitude of the converter and the actual amplitude of the current fundamental current, reduce the total harmonic compensation amount proportionally when it reaches the allowable upper limit, and superimpose the final total harmonic compensation amount onto the given channel of the fundamental current inner loop to drive the output compensation current.

2. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, Acquire the three-phase grid voltage sampling signal at the grid connection point of the energy storage converter, and perform phase-locked loop (PLL) calculations to determine the real-time angular frequency, including: The initial angular frequency is obtained by performing a phase-locked loop operation on the three-phase power grid voltage sampling signal; Calculate the change in the initial angular frequency in the current control cycle relative to the previous control cycle; When the change exceeds the preset amplitude limit ratio, the initial angular frequency is amplitude limited using the boundary value corresponding to the amplitude limit ratio, and the real-time angular frequency is finally obtained.

3. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, Based on the real-time angular frequency, and combined with the harmonic orders of the target suppression, the center resonant angular frequency of the corresponding harmonic extraction channel is reconstructed, including: Determine the order of each harmonic that needs to be suppressed; Each harmonic order is multiplied by the real-time angular frequency. The result of the multiplication operation is used as the center resonant angular frequency of the corresponding subharmonic extraction channel, and finally the center resonant angular frequency that slides synchronously with the grid frequency is formed.

4. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, The proportional term in the harmonic extraction channel transfer function is removed, and the center resonant angular frequency is substituted into the resonant term. The channel coefficients are then updated using a preset bandwidth angular frequency discretization method, including: Obtain the maximum allowable offset of the power grid frequency; The corresponding angular frequency span is calculated based on the maximum allowable offset of the power grid frequency, and the angular frequency span is used as the bandwidth angular frequency. Substituting the bandwidth angular frequency and the center resonant angular frequency into the transfer function after removing the proportional term, and then mapping it to the discrete domain using a bilinear transformation, finally the discretized updated channel coefficients are obtained.

5. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, Based on the harmonic order, real-time angular frequency, control period, and pre-set total digital link delay cycles, the phase lead angle of the corresponding harmonic is calculated, including: Inject a small-amplitude single-frequency test signal with known phase into the current loop, measure the number of control cycles corresponding to the phase lag at the output end and round it to obtain the total delay time of the digital link; The harmonic angular velocity is obtained by multiplying the harmonic order by the real-time angular frequency. The delay time is obtained by multiplying the control cycle by the total delay time of the digital link. Multiplying the harmonic angular velocity by the delay time yields the phase lead angle of the corresponding harmonic.

6. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, A rotational transformation is performed on the real-time estimated value using the phase lead angle to generate a phase-corrected compensation component, including: Transform the real-time estimated values ​​into a two-phase stationary coordinate system; In a two-phase stationary coordinate system, the real-time estimated value is subjected to coordinate rotation transformation according to the phase lead angle; The components after rotation transformation are extracted, and finally the phase-corrected compensation components are generated.

7. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, The allowable upper limit is determined based on the rated current amplitude of the converter and the actual amplitude of the current fundamental current, including: Read the preset rated current amplitude of the energy storage converter; Real-time sampling to obtain the actual amplitude of the current fundamental current; Calculate the difference between the rated current amplitude of the converter and the actual amplitude of the current fundamental current, and finally determine the difference as the allowable upper limit.

8. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, When the total harmonic compensation reaches the allowable upper limit, it is reduced proportionally, including: Calculate the sum of the real-time estimated values ​​of each harmonic in all enabled channels; Calculate the ratio of the amplitude of the real-time estimate to the sum of the amplitudes in a single channel; Multiplying the ratio by the allowable upper limit yields the proportionally reduced compensation component for each channel.

9. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, The final total harmonic compensation is superimposed onto the fundamental current inner loop given channel to drive the output compensation current, including: Obtain the fundamental current command given by the power dispatcher; The actual setpoint of the current loop is obtained by subtracting the final total harmonic compensation from the fundamental current command. The actual setpoint is input into the current loop regulator, which ultimately drives the power device to output a compensation current that is out of phase with the original harmonics at the grid connection point.

10. The method for suppressing harmonics in grid-connected current of an energy storage converter according to claim 1, characterized in that, Acquire the three-phase grid voltage sampling signal at the grid connection point of the energy storage converter, including: At the start of each current loop control cycle of the energy storage converter, the analog-to-digital converter is triggered to sample. Read the digital signal output by the analog-to-digital converter; The digital signal is filtered to obtain the three-phase power grid voltage sampling signal.

Images