A method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier

Abstract

This invention relates to a method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier, belonging to the field of rectifier control technology in power electronics. In existing technologies, when the three-phase voltage and current sampling of a three-phase PWM rectifier is inaccurate, the secondary ripple in the DC bus voltage and the harmonics of the grid-side current cannot be effectively suppressed. The method proposed in this invention establishes a transfer function between the negative-sequence current and the secondary ripple of the DC voltage based on the power conservation relationship between the AC and DC sides. It adjusts the negative-sequence current through negative feedback of the secondary ripple of the DC voltage to achieve voltage ripple suppression. Using the method described in this invention, rapid and accurate suppression of the secondary ripple of the DC voltage and the harmonics of the grid-side current can be achieved, avoiding reliance on mathematical models and complex calculations. It is not affected by sampling errors and inaccurate system parameters, and exhibits better robustness to system parameters and sampling accuracy.

Description

Technical Field

[0001] This invention belongs to the field of power electronics control technology, specifically relating to a method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier. Background Technology

[0002] PWM rectifiers are widely used in microgrids, motor drives, high-voltage direct current transmission, and electric vehicle charging due to their advantages such as controllable output voltage, low input current harmonic distortion, adjustable power factor, and bidirectional operation. However, most existing control strategies are designed under the assumption of perfect three-phase voltage balance, which is often not true in practical applications. When there is three-phase voltage imbalance, asymmetrical input filters, or deviations in current sampling multiples, traditional control methods may lead to problems such as grid-side input current distortion, DC-side output voltage containing secondary ripple, power fluctuations, and system instability. A common approach is to use a large DC capacitor to mitigate the impact of unbalanced grid on the rectifier output voltage, but a large capacitor not only increases the converter size and reduces power density but also adds unnecessary costs. Existing control methods generally have shortcomings; they cannot suppress DC voltage secondary ripple caused by grid-side voltage and current sampling errors, and they rely on positive / negative sequence decomposition and sampling accuracy, resulting in complex control structures that are unsuitable for most application scenarios.

[0003] Among the improvements to control methods under asymmetric conditions, the six-power robust model predictive control method (see patent CN114649965B), the three-phase PWM converter and its optimized control method under voltage imbalance conditions (see patent CN106411161B), and the literature "Generalized Clarke Transformation and Enhanced Dual-Loop Control Scheme for Three-Phase PWM Converters Under the Unbalanced UtilityGrid" can achieve lower total harmonic distortion of current under grid voltage imbalance. However, these methods require a large number of formula calculations, increasing control complexity, adding additional power devices, reducing the power density of the system, and exhibiting poor suppression of voltage secondary ripple.

[0004] Current methods fall into two categories: one requires additional power devices or apparatus to absorb the secondary ripple of power and eliminate the secondary ripple on the DC voltage, which increases cost and size; the other improves ripple suppression performance by adding constraints to the control method. However, both methods essentially rely on calculations of the rectifier power model and mathematical model, resulting in high computational complexity and limitations on control performance due to parameters such as sampling accuracy and inductance value. Therefore, it is necessary to conduct in-depth research on DC voltage ripple suppression methods for PWM rectifiers to quickly achieve complete suppression of DC voltage secondary ripple, while simultaneously suppressing grid-side current harmonics, and to ensure strong robustness and stability. Summary of the Invention

[0005] To achieve the above objectives, the technical solution adopted in this invention is: a DC voltage ripple suppression method for a high-gain proportional resonant PWM rectifier. This method adds control over the secondary ripple of the DC voltage to the traditional dual-closed-loop vector control. The ripple suppression method generally includes four steps: a DC voltage secondary ripple calculation part, a high-gain proportional resonant controller part, a negative-sequence active current delay part, and a traditional dual-closed-loop vector control part. The specific steps are as follows:

[0006] (1) DC voltage secondary ripple calculation part: extracting DC voltage tracking error u dcref -u dc As the input of a high-gain proportional resonant controller;

[0007] (2) High-gain proportional resonant controller section: The proportional resonant controller was modified to have a greater gain and adjustable bandwidth. After the input secondary ripple passes through the high-gain proportional resonant controller, the reference value of the negative sequence active current on the grid side is output.

[0008] (3) Negative sequence active current delay part: The negative sequence active current is delayed by an inherent time to obtain the negative sequence reactive current, which is then combined with the positive sequence current.

[0009] (4) Traditional dual closed-loop vector control section: includes DC voltage outer loop and AC current inner loop, and the output of the current inner loop enters the vector modulation section.

[0010] Furthermore, in step (1), the DC voltage u dc Divided into DC section and secondary ripple Under stable system conditions, the DC component Equal to DC voltage reference value u dcref Suppress voltage secondary ripple Equivalent to the voltage second ripple reference value The value is 0, through DC voltage u dc and DC voltage reference value u dcrefThe DC voltage ripple component is calculated as follows.

[0011]

[0012] In step (2), a high-gain proportional resonant controller G was used. HGPR (s), whose transfer function is as follows.

[0013]

[0014] Compared to quasi-proportional resonant controllers, high-gain proportional resonant controllers incorporate the parameter k. Properly setting this parameter allows for increased gain without changing the bandwidth, significantly improving controller performance. The remaining parameters are the proportional coefficient K. p resonance coefficient K r Resonant frequency ω0, cutoff frequency ω c The resonant frequency ω0 is the second-order ripple frequency, and the cutoff frequency is set according to the bandwidth requirements. The proportional gain, resonant gain, and parameter k are selected based on the required controller gain.

[0015] When designing a voltage secondary ripple control structure, it is necessary to fully consider various parameters of the actual controlled system. These parameters include, but are not limited to, load characteristics (such as the size of the load and its dynamic changes, as load changes directly affect the generation and variation of voltage ripple in the system), DC capacitor parameters (such as capacitance and equivalent series resistance, which determine the role of the capacitor in filtering and energy storage, thus affecting the amplitude and frequency characteristics of the voltage ripple), and sampling time (the sampling time determines the detection accuracy and response speed of the control system; a suitable sampling time enables the control system to obtain voltage ripple information more accurately for timely and effective control).

[0016] Based on the parameters of these actual controlled systems, the transfer function of the designed voltage second-order ripple control structure is derived through mathematical modeling and analysis. The transfer function reflects the relationship between the input control signal and the output voltage ripple in the voltage second-order ripple control structure. Through this transfer function, it is possible to clearly understand how the control signal affects the voltage ripple, and how the voltage ripple changes with the control signal.

[0017] Then, the controller is designed based on the derived transfer function. Controller design is a crucial step in the entire control system; its purpose is to effectively control the voltage second-order ripple and meet the system's performance requirements through a reasonable control strategy and algorithm. Based on the transfer function, the system's stability and dynamic performance can be analyzed, and the controller parameters can be determined to ensure precise control of the voltage second-order ripple.

[0018] The transfer function of the voltage secondary ripple control structure is as follows. This transfer function is the core mathematical expression of the entire control structure. It integrates the dynamic characteristics of various parts of the system (including the controlled object, controller, etc.) and fully describes the relationship between the input and output of the entire control structure.

[0019]

[0020] Among them, G T (s) is the sampling delay, T samp For the sampling period, G dc (s) is the transfer function corresponding to the DC-side capacitor and the load, R L C is the load, C is the DC capacitor, and G is the load. PI (s) is the transfer function of the DC voltage loop PI controller, K pu K is the proportionality coefficient. iu is the integral coefficient.

[0021] The gain of the high-gain proportional resonant controller is calculated according to the following formula, and the proportional coefficient, resonant coefficient and parameter k of the controller are designed in combination with the control structure transfer function and the second-order ripple suppression requirements.

[0022]

[0023] In step (3), the current reference value is calculated according to the following formula.

[0024]

[0025] in, It is a negative sequence active current. For negative sequence reactive current, e -τs For the delay element, τ is the delay time, which is 7.5ms in this invention, that is, 3 / 4 of the cycle corresponding to twice the power frequency.

[0026] The current reference value is calculated using the following formula.

[0027]

[0028] Among them, i dref This is the active current reference value. This is the reference value for positive sequence active current. i is the negative sequence active current reference value. qref This is the reference value for reactive current. This is the reference value for positive sequence reactive current. This is the reference value for negative sequence reactive current.

[0029] In step (4), the current reference value is input into the current inner loop control section and controlled according to the traditional proportional resonant current loop control method to achieve DC voltage secondary ripple suppression.

[0030] The advantages of this invention are: the system's control structure is simple, requiring no positive or negative sequence decomposition or extensive calculations. It achieves better dynamic and static control performance, rapidly suppressing ripple, with steady-state ripple completely suppressed, and current THD also improved. The suppression effect is unaffected by sampling errors of grid-side voltage and current, or unknown inductance values. No additional power devices are required, thus not affecting the system's power density or cost. Attached Figure Description

[0031] Figure 1 This is a comparison of the gain characteristics of the high-gain proportional resonant controller of the present invention with those of traditional proportional resonant and quasi-proportional resonant controllers.

[0032] Figure 2 This is a diagram of the voltage secondary ripple control structure;

[0033] Figure 3 This is a control structure diagram of the overall system of the present invention;

[0034] Figure 4 The method of this invention is to suppress the dynamic process waveform of DC voltage ripple. Detailed Implementation

[0035] A specific embodiment of the present invention will now be described in conjunction with the accompanying drawings, and the technical principles and effects of the present invention will be further explained.

[0036] The present invention discloses a high-gain proportional resonance PWM rectifier DC voltage ripple suppression method, the gain characteristics of which are compared with those of proportional resonance and quasi-proportional resonance as follows: Figure 1 As shown, it combines the advantages of the latter two, maintaining a high gain while possessing an individually adjustable bandwidth under the same parameters. The closed-loop control method of high-gain proportional resonant control is as follows: Figure 2 As shown, the secondary ripple of DC voltage is used as the controlled object, and the control is achieved by adjusting the negative sequence current. Figure 3 To address the issue of DC voltage ripple in a system control structure, the present invention proposes a DC voltage ripple suppression method comprising the following steps:

[0037] (1) Calculate the magnitude of DC voltage ripple. Since there is a DC voltage loop, the secondary ripple of DC voltage does not need to be calculated separately. Just take the negative of the DC voltage tracking error. When the secondary ripple reference value is set to 0, the negative sign can be canceled. That is, the DC voltage tracking error, which is the input of the DC voltage loop PI controller, is also used as the input of the high gain proportional resonant controller.

[0038] (2) High-gain proportional resonant controller, which reduces DC voltage tracking error u e As input, the high-gain proportional resonant controller outputs a reference value for the negative-sequence active current on the grid side. Because the designed controller has a high gain and a certain bandwidth, the ripple suppression controller here exhibits very small steady-state error and fast dynamic speed. The controller parameters are adjusted according to the required voltage ripple suppression effect. The parameters of the selected high-gain proportional resonant controller are as follows: proportional coefficient K. p The value is 0.05, and the resonance coefficient K is... r The value is 1, the resonant frequency ω0 is 628 rad / s, and the cutoff frequency ω c The gain is 3.14 rad / s, and the parameter k is 0.8. The transfer function of the high-gain proportional resonant controller is shown below:

[0039]

[0040] When implementing this controller digitally, the transfer function in the above equation needs to be discretized and transformed into the z-domain. Let the coefficients of the 0th term of s in the numerator of the transfer function be A0, the 1st term be A1, and the 2nd term be A2; and the coefficients of the 0th term of s in the denominator be B0, the 1st term be B1, and the 2nd term be B2. After transformation into the z-domain, the coefficients of the 0th term of z in the numerator are C0, the -1st term be C1, and the -2nd term be C2; ​​and the coefficients of the 0th term of z in the denominator are D0, the -1st term be D1, and the -2nd term be D2. The form of its transfer function in the discrete domain becomes...

[0041]

[0042] In discrete control systems, it is necessary to record the input and output values ​​of the two control cycles preceding the current control cycle. The input and output of the current control cycle are defined as u, ... e (n) and The input and output of the previous control cycle were u e (n-1) and The input and output of the control cycle before last was u e (n-2) and Before the system starts running, their initial values ​​can be set to 0 to prevent startup shocks. Finally, the formulas for the input and output of the high-gain proportional resonant controller in the discrete control system can be expressed as:

[0043]

[0044] The parameters of the control system are as follows: sampling period T samp The time is 0.1ms, and the load R is... LThe resistance is 130Ω, the DC capacitor C is 200μF, and the proportional coefficient K of the DC voltage loop PI controller is... pu The integral coefficient K is 0.02. iu It is 6.

[0045] (3) The negative sequence reactive current reference value is obtained based on the relationship between negative sequence active and reactive currents. That is, the negative sequence active current reference value is delayed by 7.5ms to obtain the negative sequence reactive current reference value. In a discrete control system, the negative sequence active current reference value of each control cycle can be stored by defining an array. The negative sequence reactive current reference value is the array data before the corresponding control cycle number. In the example of this invention, the number of cycles of the previous data required is 75, which is the multiple of the delay time of 7.5ms to the control cycle of 0.1ms.

[0046] (4) Next, the negative-sequence current is combined with the positive-sequence current. Since the negative-sequence reactive current here is also calculated in the positive-sequence rotating coordinate system, the reference current is the result of directly adding the negative-sequence current and the positive-sequence current. That is, the reference current is:

[0047]

[0048] The reference current here is the positive-sequence active current reference value. The output from the voltage loop PI controller is obtained by multiplying the DC voltage tracking error by the proportional coefficient, multiplying the sum of the values ​​over each control cycle by the integral coefficient, and then adding the two together. This can be expressed by the following formula:

[0049]

[0050] Positive sequence reactive current reference value The negative sequence active current reference value is 0, and the negative sequence reactive current reference value is output by the high-gain proportional resonant controller. Depend on The reference current is obtained with a delay. After obtaining the reference current, the positive-sequence reference current is a DC quantity, while the negative-sequence reference current is an AC quantity at twice the grid frequency, which is not suitable for dual-PI current control. Here, the active reference current i needs to be... dref and reactive reference current i qref Transforming to a two-phase stationary coordinate system, i.e., the αβ coordinate system, the reference current becomes an AC quantity with a 90° phase difference but unequal amplitude, and its frequency is equal to that of the power grid. By employing a dual-PR current control loop, combined with grid voltage feedforward, the grid current can be quickly tracked to the reference current, achieving steady-state tracking with zero steady-state error. The output V of the dual-PR current loop... PRα V PRβ After the grid-side voltage e α ,e β After feedforward, the input to space vector modulation is the reference value v of the rectifier's AC terminal voltage.αref ,v βref The corresponding formula is as follows:

[0051]

[0052] (5) When the reference value of the rectifier AC terminal voltage is input into space vector modulation, two mutually orthogonal AC quantities form a vector. This vector can be linearly represented by the vector corresponding to other switching states. Its range is within a regular hexagon and divided into 6 sectors. It is necessary to determine the position of the sector where this vector is located to decide which two vectors to use to synthesize the target rectifier AC terminal voltage vector, and it is also necessary to determine the magnitude V of the vector. m The phase θ determines the action times t1, t2, and t0 of the two fundamental vectors and the zero vector, and the corresponding calculation method is as follows:

[0053]

[0054] t0 = T samp -t1-t2

[0055] The relationship between the composite vector v and the basic vectors v1 and v2 is as follows:

[0056] v = v1t1 + v2t2

[0057] The duty cycle of each phase power switch is determined by the switching state of the basic vector and its corresponding action time. Let the duty cycle of the upper arm power switch of phase a be D. a The duty cycle of the power switch on the upper arm of phase b is D. b The duty cycle of the power switch on the upper arm of phase c is D. c For ease of calculation, a commonly used duty cycle D is defined. 1～4 as follows:

[0058] D1=t0 / 2T samp

[0059] D2=(t0+2t1) / 2T samp

[0060] D3=(t0+2t2) / 2T samp

[0061] D4=(t0+2t1+2t2) / 2T samp

[0062] The duty cycle corresponding to each sector position is as follows:

[0063]

[0064] After obtaining the duty cycle of each phase, a PWM modulation stage is used to generate drive signals to control the power switches of each phase. Finally, under the regulation of the controller, the DC voltage ripple is almost completely suppressed, and the grid-side current harmonics caused by the second ripple of the DC side voltage are also significantly reduced.

[0065] To demonstrate the significant effects of the present invention, this embodiment provides some results obtained using the embodiments. Figure 4 The dynamic start-up waveform of the DC voltage second-order ripple suppression method of this invention is presented. Within 0-0.4s, without the control method proposed in this invention, the peak-to-peak value of the DC voltage second-order ripple is 5.90V, exhibiting significant ripple fluctuations at twice the power frequency; the effective values ​​of the three-phase voltages are 66V, 110V, and 110V, respectively, and the DC component of the DC bus voltage is 300V. The effective values ​​of the grid-side three-phase currents are 3.47A, 3.30A, and 3.54A, respectively, with total harmonic distortion (THD) rates of 4.0%, 3.6%, and 3.5%, respectively, indicating severe current distortion. After 0.4s, the DC voltage ripple suppression method of this invention begins. Subsequently, the peak-to-peak value of the DC voltage second-order ripple is 0.01V, and the ripple fluctuations at twice the power frequency are effectively suppressed; the THD of the three-phase currents are 2.5%, 2.4%, and 2.4%, respectively, with a significant reduction in current harmonics. Therefore, the method described in this invention can not only effectively suppress the DC voltage second-order ripple but also effectively eliminate grid-side current harmonics.

[0066] Experimental results show that the method described in this invention can not only effectively suppress DC bus voltage ripple, but also effectively eliminate grid-side current harmonics.

[0067] This embodiment demonstrates the implementation effect of DC voltage secondary ripple suppression in a three-phase PWM rectifier. However, this invention is not limited to the three-phase voltage-type full-bridge rectifier topology. This invention is also applicable to DC voltage secondary ripple suppression in any three-phase rectifier topology.

[0068] This invention may be implemented in other specific forms without departing from its spirit or essential characteristics. The described embodiments (e.g., DC voltage secondary ripple closed-loop control, DC voltage ripple value calculation, and high-gain proportional resonant control, etc.) are considered illustrative rather than limiting in all respects. Therefore, the scope of this invention is indicated by the appended claims rather than the foregoing description. All variations falling within the meaning and scope of the equivalent technical solutions of the claims are included within its scope.

Claims

1. A method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier, which adds control of the secondary ripple of DC voltage to the traditional dual-closed-loop vector control. The ripple suppression method generally includes four steps: DC voltage secondary ripple calculation, high-gain proportional resonant controller, negative-sequence active current delay, and traditional dual-closed-loop vector control. The specific steps are as follows: (1) DC voltage secondary ripple calculation part: extracting DC voltage tracking error As the input to the high-gain proportional resonant controller, the DC voltage is divided into a DC component. and secondary ripple Under stable system conditions, the DC section Equal to DC voltage reference value Suppress voltage secondary ripple Equivalent to the voltage second ripple reference value =0, through DC voltage and DC voltage reference value The DC voltage ripple component is calculated using the following formula. ； (2) High-gain proportional resonant controller section: The proportional resonant controller was modified to have a higher gain and adjustable bandwidth, and the input DC voltage tracking error was reduced. After passing through the high-gain proportional resonant controller, the output grid-side negative sequence active current reference value is... The transfer function of the high-gain proportional resonant controller used is: Its transfer function is as follows: ； Compared to quasi-proportional resonant controllers, high-gain proportional resonant controllers incorporate parameters. k , k A value greater than 0 allows the controller gain to increase without changing the bandwidth, significantly improving controller performance. The other parameters are proportional coefficients. resonance coefficient resonant frequency Cutoff frequency The resonant frequency The frequency is the second-order ripple frequency, and the cutoff frequency is set according to the bandwidth requirements, along with the scaling factor, resonant coefficient, and parameters. k Select based on the required controller gain; (3) Negative sequence active current delay part: The negative sequence active current reference value output by the high gain proportional resonant controller. The negative sequence reactive current reference value is obtained after delaying the inherent time. When combined with the positive sequence current, the active current reference value is obtained. and reactive current reference value Entering the inner loop of alternating current; (4) Traditional dual-loop vector control section: including DC voltage outer loop and AC current inner loop. The AC current inner loop adopts the traditional proportional resonant current loop control method, and the active current reference value is... and reactive current reference value It enters the AC current inner loop, and the output of the AC current inner loop then enters the vector modulation section.

2. The method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier as described in claim 1, characterized in that: When designing a voltage secondary ripple control structure, it is necessary to fully consider various parameters of the actual controlled system. These parameters include load characteristics, DC capacitor parameters, including: capacitor size and equivalent series resistance. These parameters determine the role of the capacitor in filtering and energy storage, which in turn affects the amplitude and frequency characteristics of the voltage ripple, as well as the sampling time. The length of the sampling time determines the detection accuracy and response speed of the control system. Based on the parameters of these actual controlled systems, the transfer function of the designed voltage second ripple control structure is derived through mathematical modeling and analysis. The transfer function reflects the relationship between the input control signal and the output voltage ripple in the voltage second ripple control structure. Through this transfer function, it is possible to clearly understand how the control signal affects the voltage ripple and how the voltage ripple changes with the control signal. Then, the controller is designed based on the derived transfer function. The design of the controller is the key link in the entire control system. Its purpose is to effectively control the voltage second ripple and meet the system performance requirements through reasonable control strategies and algorithms. Based on the transfer function, the stability and dynamic performance of the system can be analyzed, and the parameters of the controller can be determined to ensure that the controller can accurately control the voltage second ripple. The overall transfer function of the voltage second-order ripple control structure is as follows. This transfer function is the core mathematical expression of the entire control structure. It integrates the dynamic characteristics of various parts of the system, including the controlled object and the controller, and completely describes the relationship between the input and output of the entire control structure: ； in, For sampling delay, The sampling period is This is the transfer function corresponding to the DC-side capacitor and the load. For load, It is a DC capacitor. The transfer function for the DC voltage loop PI controller. This is the proportionality coefficient. is the integral coefficient.

3. The method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier as described in claim 1, characterized in that: The gain of the high-gain proportional resonant controller can be calculated using the following formula: 。 4. The method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier as described in claim 1, characterized in that: In Part (3), the negative sequence reactive current is calculated according to the following formula. ； in, This is the reference value for negative sequence active current. This is the reference value for negative sequence reactive current. For the delayed process, It is the delay time, which is 7.5 ms, or 3 / 4 of the cycles corresponding to twice the power frequency.

5. The method for suppressing DC voltage ripple in a high-gain proportional resonant PWM rectifier as described in claim 1, characterized in that: In Part (3), the current reference value is calculated according to the following formula. ； in, This is the active current reference value. This is the reference value for positive sequence active current. This is the reference value for negative sequence active current. This is the reference value for reactive current. This is the reference value for positive sequence reactive current. This is the reference value for negative sequence reactive current.

Images