A frequency-adaptive APF control method under grid voltage imbalance

Abstract

This invention discloses an APF control method with frequency adaptation under grid voltage imbalance. The inner loop (voltage loop) uses PI control to stabilize the DC bus voltage, achieving stable energy exchange between the two sides of the converter. The outer loop (current loop) employs a composite control algorithm combining proportional-integral (PI), proportional-complex-integral (PCI), and proportional-resonant (PR) control to achieve precise control of the compensation current. When the grid voltage is balanced, only PI plus PCI control is used, resulting in a large stability margin and fast dynamic response. When the grid voltage is unbalanced, a full-pass filter separates the reference signal into positive and negative sequences. The positive sequence component still uses PI plus PCI control, while the negative sequence component uses PI plus PR control, addressing the poor stability of single PCI control under grid voltage imbalance. A frequency-locked loop is introduced into the system to track and control the grid frequency in real time, improving current control accuracy and simplifying the computational complexity of the control algorithm. This method is applicable to other three-phase inverter control systems.

Description

Technical Field

[0001] This invention belongs to the field of harmonic current control technology, specifically relating to an APF control method with frequency self-adaptation under grid voltage imbalance. Background Technology

[0002] The significant increase in nonlinear loads in power systems, such as nonlinear inverters, rectifiers, and switching power supplies, has led to increasingly severe harmonic and reactive power problems, commonly known as harmonic pollution. This poses a significant challenge to the stability of power systems, communication security, and the safety of electrical appliances. The concept of active power filters has provided a new starting point for solving harmonics and improving power quality. With the development of harmonic detection methods based on instantaneous reactive power theory, harmonic mitigation has reached a new milestone. How to quickly and accurately track harmonic currents with compensation current has become a research topic for many scholars.

[0003] The most classic current control algorithm is proportional-integral (PI) control. PI control is simple and has a fast dynamic response, but it can only eliminate DC steady-state errors and its effect on AC is not significant. Therefore, it is often used in conjunction with other current control algorithms to achieve composite control. Scholars have subsequently conducted research on composite control algorithms such as proportional-resonance (PR) control, deadbeat control, and repetitive control, each with its own advantages and disadvantages. In the research of grid-connected inverters, some scholars have compared proportional-complex-integral (PCI) with other control algorithms and discovered its superior control capabilities. Therefore, some scholars have attempted to apply PCI to active power filters (APFs).

[0004] However, this scholar conducted simulations under ideal conditions and did not consider that the proportional-complex integral oscillator's anti-disturbance characteristic for negative-sequence components is not zero under grid voltage imbalance, leading to inaccurate control of harmonic currents. Furthermore, the resonant bandwidth of the PCI algorithm is smaller than other control algorithms. When the grid frequency fluctuates, it cannot guarantee that the controller gain at the resonant point is sufficiently large. Some scholars have increased the bandwidth of the controller's resonant point by modifying the algorithm, ensuring that even if the resonant point shifts when the grid fundamental frequency deviates, the gain at the resonant point remains sufficiently large. However, this increases the computational load of the algorithm. Summary of the Invention

[0005] Objective: To address the shortcomings of existing technologies, this invention provides an APF control method with frequency adaptation under grid voltage imbalance. It utilizes an all-pass filter to separate the positive and negative sequences of harmonic currents, thereby enabling stable and rapid current control under different grid voltage conditions. Furthermore, a frequency-locked loop is introduced, which can detect the grid frequency in real time and provide feedback, solving the frequency offset problem at its source without increasing the computational load through algorithmic changes.

[0006] Technical Solution: To achieve the objectives of this invention, the technical solution adopted is: an APF control method with frequency adaptation under grid voltage imbalance, comprising the following steps:

[0007] First, the harmonic current components in the two-phase stationary coordinate system are obtained using the harmonic current detection method based on the instantaneous reactive power theory.

[0008] Next, the harmonic current components are passed through an all-pass filter to obtain the positive-sequence and negative-sequence components of the harmonic current components.

[0009] Then, a current control algorithm combining PI control and PCI control is used to control the positive sequence component of the harmonic current; the negative sequence component of the harmonic current is controlled by detecting whether the grid voltage is balanced.

[0010] If the grid voltage is balanced, the PI superimposed PCI current control algorithm is used to control the negative sequence component of the harmonic current. If the grid voltage is unbalanced, the PR superimposed PI current control algorithm is used to control the negative sequence component of the harmonic current.

[0011] Finally, a frequency-locked loop based on a second-order generalized integrator is used to track the fundamental frequency of the power grid in real time. By adjusting the resonant frequency of the second-order generalized integrator to follow the actual frequency of the power grid, the frequency error is controlled to become 0, thus achieving frequency locking.

[0012] Furthermore, the generation process of harmonic currents is as follows:

[0013] The load current signal I in the three-phase stationary coordinate system a ,I b ,I c The harmonic current components are transformed to a two-phase stationary coordinate system using Clark transformation, resulting in the α-axis component I. α and β-axis component I β ;

[0014] A phase-locked loop is used to obtain the sine and cosine signals of the grid voltage in phase with the required Park transform; then I α and I β Active component I transformed into a two-phase rotating coordinate system d and reactive component I q ;

[0015] Active component I d and reactive component I q After passing through a second-order Butterworth low-pass filter, the fundamental active component I is obtained. db and fundamental reactive component I qb ;

[0016] Active component I d Subtract the fundamental active component I db In addition, the output value of the inner voltage loop is used to obtain the active component of the harmonic current I. dh ;The reactive component I q Subtract the fundamental reactive component I qb Obtain the reactive component I of the harmonic current qh ;

[0017] Finally, I dh and I qh The harmonic current component I in the α-β coordinate system is obtained by inverse Park transform. αh and I βh The compensation current I generated by the inverter in the abc coordinate system ca ,I cb ,I cc The compensation current Ic in the α-β coordinate system is obtained after Clark transformation. α and Ic β .

[0018] Furthermore, the process of separating the positive and negative sequences of harmonic current components using an all-pass filter is as follows:

[0019] The harmonic current component I in the α-β coordinate system αh ,I βh Compensation current Ic in α-β coordinate system α ,Ic β Input an all-pass filter, and use the 90° phase shift function of the all-pass filter to... αh ,I βh and Ic α ,Ic β Perform positive and negative order separation to obtain the positive order component I. αh + ,I βh + ,Ic α + and Ic β + and negative order component I αh - ,I βh - ,Ic α - and Ic β - ;

[0020] The decomposition formulas for the positive and negative sequence components in a two-phase stationary coordinate system are as follows:

[0021]

[0022] Where q is the 90° phase lag factor in the time domain.

[0023] Furthermore, the current control algorithm steps under grid voltage imbalance conditions are as follows:

[0024] (1) Positive sequence component I αh + and Ic α + Subtraction yields the error value X α + Orthogonal component I βh + and Ic β + Subtraction yields the error value X β + Negative-order component I αh - and Ic α - Subtraction yields the error value X α - Negative-order component I βh - and Ic β - Subtraction yields the error value X β - ;X α + and X β + The mutually orthogonal signals are fed into the PCI controller as inputs to obtain the positive sequence control signal Y. α + and Y β + ;

[0025] (2) Determine whether the grid voltage is balanced. If the grid voltage is balanced, set the error value X. α - and X β - The mutually orthogonal signals are fed into the PCI controller as inputs to obtain the negative sequence control signal Y. α - and Y β - If the grid voltage is unbalanced, then the error value X will be... α - and X β - The signal is fed into the PR controller to obtain the negative sequence control signal Y.α - and Y β - ;

[0026] (3) The positive and negative sequence control signal Y α + and Y α - The sum is used to obtain the control signal Y. α The positive and negative sequence control signal Y β + and Y β - The sum is used to obtain the control signal Y. β ;

[0027] (4) Control signal Y α and Y β Performing the Clark inverse transform yields the control signal Y in the three-phase stationary coordinate system. a ,Y b and Y c ;

[0028] (5) Transfer the control signal Y a ,Y b ,Y c Compared with the triangular carrier wave, an SPWM wave is generated to control the on and off of the IGBTs in the inverter section of the active power filter, generating the corresponding compensation current I in the abc coordinate system. ca , I cb and I cc .

[0029] Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0030] This invention rationally applies the PCI control algorithm from grid-connected inverters to the harmonic current compensation tracking control of active power filters. The algorithm is more advanced and combines PR and PI control, enabling precise control of the negative sequence component even under grid voltage imbalances, resulting in faster and more stable operation. Furthermore, considering the complex situation of frequency fluctuations in actual grid voltage, and addressing the issue of insufficient resonant gain at the resonant frequency of proportional-complex-integral and proportional-resonant control, this invention directly controls the grid frequency deviation to a sufficiently small level from the source, effectively improving system stability and control speed. Attached Figure Description

[0031] Figure 1 It is the overall control block diagram for separating the positive and negative sequences of harmonic currents, i.e., reference signals;

[0032] Figure 2This is the overall control block diagram for compensating current tracking (harmonic current) reference signals using PCI-PR composite control;

[0033] Figure 3 This is a block diagram of an all-pass filter for separating the positive and negative sequences of harmonic currents. Detailed Implementation

[0034] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] The present invention discloses an APF control method with frequency adaptation under grid voltage imbalance, comprising the following steps: First, harmonic current components in a two-phase stationary coordinate system are obtained using a harmonic current detection method based on instantaneous reactive power theory; second, the harmonic current components are passed through an all-pass filter to obtain the positive-sequence and negative-sequence components of the harmonic current; then, a current control algorithm combining PI control and PCI control is used to control the positive-sequence component of the harmonic current; the negative-sequence component of the harmonic current is controlled by detecting whether the grid voltage is balanced; if the grid voltage is balanced, the current control algorithm combining PI and PCI is used to control the negative-sequence component of the harmonic current; if the grid voltage is unbalanced, the current control algorithm combining PI and PI is used to control the negative-sequence component of the harmonic current; finally, a frequency-locked loop based on a second-order generalized integrator is used to track the fundamental frequency of the grid in real time, and the frequency error is controlled to zero by adjusting the resonant frequency of the second-order generalized integrator to follow the actual frequency of the grid, thus achieving frequency locking.

[0036] Figure 1 This is a block diagram showing the overall control of harmonic currents, i.e., the positive and negative sequences of the reference signal, similar to the i-th harmonic current based on instantaneous reactive power theory. d -i q The first part of the detection method requires obtaining reference signals for the positive and negative sequence components of the current control loop. First, a Hall effect current sensor is used to detect the three-phase load current I. La , I Lb , I Lc Harmonic current components in a two-phase stationary coordinate system are obtained through Clark transformation: α-axis component I α and β-axis component I β The phase-locked loop (PLL) is used to obtain the grid voltage phase information required for the Park transform, and the active power component I in the two-phase rotating coordinate system is obtained through the Park transform. d and reactive component I q Then, the fundamental frequency component is filtered out using a second-order Butterworth low-pass filter (with a set cutoff frequency of 10Hz). The fundamental frequency component is then inverted and added to the original value to obtain the harmonic component I. dh and I qh This requires subtracting the DC bus voltage reference voltage Vd* from the actual measured voltage Vd, then outputting the result after passing it through a PI controller, and finally comparing it with the active component I.d Superposition as a new active component I d Specifically, the active component I d and reactive component I q After passing through a second-order Butterworth low-pass filter, the fundamental active component I is obtained. db and fundamental reactive component I qb Active component I d Subtract the fundamental active component I db In addition, the output value of the inner voltage loop is used to obtain the active component of the harmonic current I. dh Reactive component I q Subtract the fundamental reactive component I qb Obtain the reactive component I of the harmonic current qh . Will I dh and I qh The harmonic current component I in the two-phase rotating coordinate system is obtained by inverse Park transform. αh and I βh The harmonic current component I αh and I βh By utilizing the 90° phase shift capability of an all-pass filter, positive and negative phase sequence separation is performed to obtain the reference signal I for the compensation current. αh + and I βh + (positive-order components) and I αh - and I βh - (Negative-order components).

[0037] The compensation current I in the abc coordinate system generated by the inverter ca ,I cb ,I cc The compensation current Ic in the α-β coordinate system is obtained after Clark transformation. α and Ic β The compensation current Ic in the α-β coordinate system α ,Ic β Input an all-pass filter, and use the 90° phase shift function of the all-pass filter to reduce Ic α ,Ic β Perform positive and negative order separation to obtain the positive order component Ic. α + and Ic β + and negative order component Ic α - and Ic β - .

[0038] A current control algorithm that combines PI control and PCI control is used to control the positive sequence component of harmonic current. Figure 2 This is the overall control block diagram for compensating current tracking (harmonic current) of the reference signal using PCI and PR combined control. First, let's look at the positive sequence component I of the reference signal. αh + and I βh + With the positive sequence component Ic of the compensation current α + and Ic β + The error signal X is obtained by subtraction. α and X β Then, the output result Y is obtained through a PI controller and multiple PCI controllers connected in parallel. α + and Y β + .

[0039] The negative sequence component of the harmonic current is controlled by detecting whether the grid voltage is balanced. If the grid voltage is balanced, the negative sequence component of the harmonic current is controlled by a current control algorithm of PI superimposed PCI. If the grid voltage is unbalanced, the negative sequence component of the harmonic current is controlled by a current control algorithm of PR superimposed PI.

[0040] The steps of the current control algorithm under the condition of grid voltage imbalance are: (1) Positive sequence component I αh + and Ic α + Subtraction yields the error value X α + Orthogonal component I βh + and Ic β + Subtraction yields the error value X β + Negative-order component I αh - and Ic α - Subtraction yields the error value X α - Negative-order component I βh - and Ic β - Subtraction yields the error value X β - ;X α + and X β + The mutually orthogonal signals are fed into the PCI controller as inputs to obtain the positive sequence control signal Y. α + and Y β +(2) Determine whether the grid voltage is balanced. If the grid voltage is balanced, set the error value X. α - and X β - The mutually orthogonal signals are fed into the PCI controller as inputs to obtain the negative sequence control signal Y. α - and Y β - If the grid voltage is unbalanced, then the error value X will be... α - and X β - The signal is fed into the PR controller to obtain the negative sequence control signal Y. α - and Y β - (3) The positive and negative sequence control signal Y α + and Y α - The sum is used to obtain the control signal Y. α The positive and negative sequence control signal Y β + and Y β - The sum is used to obtain the control signal Y. β (4) Transfer the control signal Y α and Y β Performing the Clark inverse transform yields the control signal Y in the three-phase stationary coordinate system. a ,Y b and Y c (5) Transfer the control signal Y a ,Y b ,Y c Compared with the triangular carrier wave, an SPWM wave is generated to control the on and off of the IGBTs in the inverter section of the active power filter, generating the corresponding compensation current I in the abc coordinate system. ca , I cb and I cc .

[0041] Since the harmonics are mostly concentrated in the 5th, 7th, and 11th harmonics, it is only necessary to set the resonant frequency of the PCI resonant point to... n is the corresponding harmonic order. This is the fundamental frequency of the power grid.

[0042] Because the resonant bandwidth of the PCI controller is smaller than that of other controllers, and the actual grid frequency may fluctuate, this can lead to a shift in the resonant point, resulting in insufficient gain at the resonant point. Ignoring frequency shift will lead to poor current tracking control performance. Existing methods use QPR (quasi-proportional resonant) control instead of PR control. QPR has a larger resonant bandwidth than PR, ensuring sufficient gain at the resonant frequency even when the grid frequency shifts, but this increases the computational load. Therefore, if this invention focuses on improving the PCI bandwidth, the computational load will be even greater. Therefore, this invention uses a frequency-locked loop based on a second-order generalized integrator to track the fundamental frequency of the grid in real time. By adjusting the resonant frequency of the second-order generalized integrator to follow the actual grid frequency, the frequency error is reduced to zero, thus achieving frequency locking and solving the frequency shift problem at its source. This ensures that the resonant points of PR and PCI are aligned correctly. Maintaining a large gain within a small frequency bandwidth makes current tracking control more stable.

[0043] Figure 3 The block diagram shows the structure for separating the positive and negative sequence of harmonic currents in an all-pass filter. The decomposition formulas for the positive and negative sequence components in a two-phase stationary coordinate system are:

[0044]

[0045] Where q is the 90° phase lag factor in the time domain. The input and output of an all-pass filter have a 90° phase shift relationship. Therefore, by utilizing the 90° phase shift capability of the all-pass filter and performing some calculations, phase sequence separation can be achieved.

[0046] The active power filter control method with frequency adaptation under grid voltage imbalance described in this invention solves the problem of potential shift in the grid fundamental frequency from the source, simplifying the computational workload of the algorithm. The current tracking control algorithm is simpler than traditional control algorithms and considers the situation under grid voltage imbalance by controlling the positive and negative sequence components separately, thus improving the system's stability and speed.

[0047] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. An APF control method with frequency adaptation under grid voltage imbalance, characterized in that: The method includes the following steps: First, the harmonic current components in the two-phase stationary coordinate system are obtained using the harmonic current detection method based on the instantaneous reactive power theory. Next, the harmonic current components are passed through an all-pass filter to obtain the positive-sequence and negative-sequence components of the harmonic current components. Then, a current control algorithm combining PI control and PCI control is used to control the positive sequence component of the harmonic current; the negative sequence component of the harmonic current is controlled by detecting whether the grid voltage is balanced. If the grid voltage is balanced, the PI superimposed PCI current control algorithm is used to control the negative sequence component of the harmonic current. If the grid voltage is unbalanced, the PR superimposed PI current control algorithm is used to control the negative sequence component of the harmonic current. Finally, a frequency-locked loop based on a second-order generalized integrator is used to track the fundamental frequency of the power grid in real time. By adjusting the resonant frequency of the second-order generalized integrator to follow the actual frequency of the power grid, the frequency error is controlled to become 0, thus achieving frequency locking. The compensating currents I in the a-b-c coordinate system generated by the inverter ca , I cb , I cc are transformed by the Clark transformation to obtain the compensating currents Ic in the α-β coordinate system α and Ic β； The process of separating the positive and negative sequences of harmonic current components using an all-pass filter is as follows: The harmonic current components I αh , I βh in the α-β coordinate system and the compensation current Ic α , Ic β are input into an all-pass filter. Using the function of the all-pass filter to shift the phase by 90°, I αh , I βh and Ic α , Ic β are separated into positive and negative sequence components to obtain the positive sequence components I αh + , I βh + , Ic α + and Ic β + as well as the negative sequence components I αh - , I βh - , Ic α - and Ic β - ; The decomposition formulas for the positive and negative sequence components in a two-phase stationary coordinate system are as follows: , Where q is the 90° phase lag factor in the time domain; The steps of the current control algorithm under grid voltage imbalance conditions are as follows: (1)Positive sequence component I αh + and Ic α + are subtracted to obtain the error value X α + , the positive sequence component I βh + and Ic β + are subtracted to obtain the error value X β + ; the negative sequence component I αh - and Ic α - are subtracted to obtain the error value X α - , the negative sequence component I βh - and Ic β - are subtracted to obtain the error value X β - ; X α + and X β + are mutually orthogonal and sent as inputs to the PCI controller to obtain the positive sequence control signal Y α + and Y β + ; (2) Determine whether the grid voltage is balanced. If the grid voltage is balanced, use the error values X α - and X β - that are orthogonal to each other as inputs to the PCI controller to obtain the negative sequence control signal Y α - and Y β - ; if the grid voltage is unbalanced, then use the error values X α - and X β - as inputs to the PR controller to obtain the negative sequence control signal Y α - and Y β - ; (3) Add the positive and negative sequence control signals Y α + and Y α - to obtain the control signal Y α , add the positive and negative sequence control signals Y β + and Y β - to obtain the control signal Y β ; (4) Carry out the Clark inverse transformation on the control signals Y α and Y β to obtain the control signals Y a , Y b and Y c ; (5) Compare the control signals Y a , Y b , Y c with the triangular carrier wave to generate the SPWM wave, control the conduction and cutoff of the IGBTs in the inverter part of the active power filter, and generate the corresponding compensation current I ca , I cb and I cc .

2. The APF control method with frequency adaptation under grid voltage imbalance according to claim 1, characterized in that: The generation process of harmonic current is as follows: The load current signals I in the three-phase stationary coordinate system a , I b , I c are transformed into harmonic current components in the two-phase stationary coordinate system through Clark transformation, obtaining the α-axis component I α and the β-axis component I β ; The sine and cosine signals in phase with the grid voltage required for Park transformation are obtained through a phase-locked loop; then I α and I β are transformed into the active component I d and the reactive component I q in the two-phase rotating coordinate system; The active component I d and the reactive component I q are passed through a second-order Butterworth low-pass filter to obtain the fundamental active component I db and the fundamental reactive component I qb ; Subtract the active component I d from the fundamental active component I db and then add the output value of the inner voltage loop to obtain the active component of harmonic current I dh ; Subtract the reactive component I q from the fundamental reactive component I qb to obtain the reactive component of harmonic current I qh ; Finally, I dh and I qh are transformed through the Park inverse transformation to obtain the harmonic current components I αh and I βh .

Images