Method for reactive power sharing of multi-configuration grid-connected converter under single voltage loop amplitude control

Abstract

A reactive power sharing method for a multi-grid inverter under single-voltage loop amplitude control includes the following steps: In a grid-type inverter system using single-voltage loop amplitude control, the voltage and current signals output by the inverter are acquired, and the actual output active power P is calculated. i With reactive power Q i Processing active power P based on droop control loop i With reactive power Q i Generate voltage amplitude reference value V iref and voltage phase angle θ i Based on voltage amplitude reference value V iref Single-voltage loop inner-loop control is performed, and an initial modulation signal is generated; based on reactive power Q i Construct the feedforward compensation term V comp1 Construct feedback compensation term V comp2 ; the feedforward compensation term V comp1 Feedback and compensation item V comp2 With V iref By superimposing and repeating step S2, the final modulation signal is generated, driving the converter output to achieve power sharing. This significantly improves the system's dynamic response speed and combines the advantages of autonomous operation and coordinated control.

Description

Technical Field

[0001] This invention belongs to the field of power system control, and specifically relates to a reactive power sharing method for multi-grid converters under single voltage loop amplitude control. Background Technology

[0002] Currently, distributed generation systems, represented by wind and solar power, are playing an increasingly important role in microgrid architectures. In the islanded operation mode of microgrids, grid-controlled inverters are often connected in parallel to maintain a stable power supply. Achieving balanced power distribution among parallel inverters, especially precise reactive power allocation, has become a key issue in ensuring the reliable operation of microgrids.

[0003] In multi-machine parallel systems operating in islanded mode with grid-connected converters, the outer-loop control strategy based on droop control has become the most widely used control method due to its ability to achieve autonomous power distribution without relying on communication. However, traditional droop control has significant limitations in practical applications: on the one hand, the mismatch between inverter output impedance and line impedance leads to uneven power distribution, especially when using P-ω / QV droop control, where line impedance differences severely affect the accurate distribution of reactive power; on the other hand, factors such as load fluctuations may further exacerbate power distribution deviations, causing overload of some converters, or even triggering protection actions and leading to cascading faults. In addition, uneven reactive power distribution can also lead to reactive power circulating current, reducing system operating efficiency and stability.

[0004] In grid-based control strategies, inner-loop control is typically divided into single-voltage amplitude control and voltage-current dual-loop control. Existing research largely focuses on achieving power sharing through voltage-current dual-loop control, while research on single-voltage amplitude control is relatively limited. Currently, virtual impedance has clear application value in voltage-current dual-loop control, effectively improving power coupling and reactive power sharing. The effectiveness of this control strategy stems from its mechanism: power coupling and reactive power imbalance are essentially caused by differences in grid connection point voltage, and line impedance is a key factor affecting the grid connection point voltage. Under the voltage-current dual-loop control architecture, virtual impedance adjusts the grid connection point voltage, effectively compensating for line impedance imbalance, thereby optimizing power distribution. In contrast, virtual impedance in single-voltage amplitude control only adjusts the inverter output voltage, effectively compensating for filter impedance, without directly affecting the grid connection point voltage, thus having inherent limitations in solving power sharing problems. This difference essentially stems from the different control targets of the two strategies: dual-loop control targets the grid connection point voltage, while single-voltage control only targets the inverter output voltage. Therefore, the traditional virtual impedance method cannot be applied to achieve power equalization in single voltage amplitude control.

[0005] Because single-voltage amplitude control differs from dual-loop control in output characteristics, its power sharing strategy cannot be directly applied to the dual-loop control method. Therefore, for the power sharing problem of single-voltage amplitude control, more effective optimization schemes still need to be explored to improve the stability and reliability of microgrid islanded operation. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art and provide a power sharing method for multi-grid converters under single voltage amplitude control.

[0007] A reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control includes the following steps:

[0008] S1. In a grid-type inverter system employing single-voltage loop amplitude control, the voltage and current signals output by the inverter are acquired, processed by Park transform and a first-order low-pass filter, and the actual output active power P is calculated. i With reactive power Q i Processing active power P based on droop control loop i With reactive power Q i Generate voltage amplitude reference value V iref and voltage phase angle θ i ;

[0009] S2. Based on voltage amplitude reference value V iref Control the inner loop of the single voltage loop and generate the initial modulation signal;

[0010] S3. Based on reactive power Q i Construct the feedforward compensation term V comp1 ;

[0011] S4. Construct a feedback compensation term V based on average reactive power. comp2 ;

[0012] S5. The feedforward compensation term V comp1 Feedback and compensation item V comp2 With V iref By superimposing and repeating step S2, the final modulation signal is generated, which drives the converter output to achieve power equalization.

[0013] In a grid-type inverter system employing single-voltage loop amplitude control, the voltage and current signals output by the inverter are collected, and the actual output active power P is calculated. i and reactive power Q i Specifically:

[0014] ;

[0015] Among them, U odwith U oq The d-axis and q-axis components of the voltage at the grid connection point of the i-th grid-connected converter are I. od with I oq These are the d-axis and q-axis components of the current at the grid connection point of the i-th grid-connected converter.

[0016] Processing active power P based on droop control loop i and reactive power Q i Generate voltage amplitude reference value V iref and voltage phase angle θ i Specifically:

[0017] The voltage amplitude reference value V is generated by the reactive power-voltage droop circuit. iref The actual angular frequency is generated by the active-frequency droop element, and the voltage phase angle θ is obtained after integration. i The formula is expressed as:

[0018] ;

[0019] ;

[0020] In the formula ω iref ω0 is the actual angular frequency, V0 is the rated angular frequency of the system, and P is the rated amplitude of the grid connection point voltage. ref With Q ref These are the reference values ​​for active power and reactive power, respectively, k q k is the reactive voltage droop factor. p This is the active power droop coefficient.

[0021] Based on voltage amplitude reference value V iref The inner loop of the single voltage loop is controlled, and a modulation signal is generated, specifically as follows:

[0022] Based on voltage amplitude reference value V iref The grid connection point voltage is controlled in a closed loop by a proportional-integral regulator to generate the converter output voltage amplitude d-axis command E. id ; Command the corresponding q-axis E iq =0 for E id With E iq The inverse Park transform is performed to convert it into a modulated waveform, and then a drive signal is generated via pulse width modulation to control the on and off of the inverter IGBT.

[0023] Based on voltage amplitude reference value V iref The grid connection point voltage is controlled in a closed loop by a proportional-integral regulator to generate the converter output voltage d-axis command E. id for:

[0024] ;

[0025] In the formula E id The d-axis command for the converter output voltage, V pcc k represents the actual voltage amplitude at the grid connection point. v This is the voltage regulation coefficient.

[0026] Based on reactive power Q i Construct the feedforward compensation term V comp1 Specifically:

[0027] V comp1 = kQ i ;

[0028] k is the feedforward coefficient;

[0029] After adding the feedforward compensation term, the converter output voltage d-axis command can be expressed as:

[0030] .

[0031] Construct a feedback compensation term V based on average reactive power comp2 Specifically, it includes:

[0032] The reactive power Q of all parallel converters in the system is obtained through the communication network. i (i=1,2,…,n), calculate the average reactive power of the entire network. Then, calculate the reactive power Q of the i-th converter. i Deviation from average reactive power ΔQ i =Q i - Q avg This deviation ΔQ i The input is sent to the integral controller to generate the feedback adjustment quantity V. comp2 :

[0033] ;

[0034] k1 is the integral coefficient.

[0035] The feedforward compensation term V comp1 Feedback and compensation item V comp2 With V iref By superimposing and repeating step S2, the final modulation signal is generated, which drives the converter output to achieve power sharing. Specifically, this includes:

[0036] The feedforward compensation term V comp1 Feedback and compensation item V comp2 Directly superimposed on the voltage amplitude reference value V iref Then, the command is sent to the voltage inner loop controller, and the converter output voltage d-axis command is:

[0037] ;

[0038] Command the corresponding q-axis E iq =0 for E id With E iq The modulated wave m is obtained by performing an inverse Park transform. abc Then, a drive signal is generated through pulse width modulation to control the switching on and off of the inverter's IGBTs.

[0039] The inventors discovered that a necessary condition for achieving precise reactive power allocation by droop coefficient is maintaining equal voltage at the grid connection point of each converter. However, differences in the line impedance connecting each converter to the grid lead to deviations in the grid connection point voltage, thus affecting the accuracy of reactive power allocation. In voltage-current dual closed-loop control, the virtual impedance method can effectively adjust the line impedance, making the equivalent line impedance of each converter tend to be consistent, thereby ensuring equal grid connection point voltage and achieving precise reactive power allocation. However, in a single voltage amplitude control strategy, virtual impedance can only effectively adjust the filter impedance and cannot effectively adjust the grid connection point voltage, thus failing to improve the reactive power equivalence. This discovery reveals the differentiated influence mechanism of virtual impedance on reactive power allocation under different control strategies.

[0040] Given the critical impact of grid connection point voltage on power sharing, this invention proposes a power sharing strategy based on direct adjustment of grid connection point voltage for multi-grid converter systems with single-voltage amplitude control. By utilizing the inherent characteristics of droop control—namely, the negative correlation between grid connection point voltage and reactive power (reactive power increases as voltage decreases)—a self-reactive power feedforward compensation mechanism is introduced into the voltage amplitude control stage. Under this compensation mechanism, converters with lower grid connection point voltages have higher self-reactive power, thus compensating for their higher self-reactive power output; conversely, converters with higher grid connection point voltages are compensated for their lower self-reactive power output. This compensation method does not rely on communication and effectively reduces voltage deviations between converters, thereby significantly improving power sharing accuracy. However, the feedforward compensation coefficient k is difficult to select accurately, leading to insufficient sharing accuracy.

[0041] Based on this, in order to further improve the reactive power distribution accuracy, an average power feedback mechanism is introduced, in which each converter calculates its own reactive power Q. i It exchanges power information with other converters via digital communication, and then calculates the average reactive power Q in real time. avg The voltage compensation amount is further adjusted by the average power and its own reactive power error, so as to achieve accurate voltage compensation and reasonable distribution of reactive power. Attached Figure Description

[0042] Figure 1 Topology diagram of a single grid-type converter in this embodiment;

[0043] Figure 2 This embodiment shows the inner loop voltage amplitude control diagram without any compensation.

[0044] Figure 3 This embodiment introduces a hybrid voltage compensation topology that combines feedforward voltage compensation with average power feedback control.

[0045] Figure 4 The reactive power simulation waveform results in this embodiment only use the feedforward voltage compensation strategy.

[0046] Figure 5 The reactive power simulation waveform results of this embodiment only use the average power feedback control strategy;

[0047] Figure 6 The reactive power simulation results of the collaborative control strategy of feedforward voltage compensation and average power feedback in this embodiment are shown in the figure.

[0048] Figure 7 The active power simulation results of this embodiment using a coordinated control strategy of feedforward voltage compensation and average power feedback are shown in the figure. Detailed Implementation

[0049] This invention addresses the reactive power distribution problem of grid-type converters under single-voltage amplitude control under line impedance mismatch conditions. A hybrid compensation control strategy is proposed: firstly, a feedforward voltage compensation method without communication requirements is proposed to achieve preliminary power balancing through local measurement; secondly, a communication-based collaborative compensation mechanism is introduced to construct a two-stage compensation architecture. This hybrid control strategy has the following technical features:

[0050] (1) It achieves precise allocation of reactive power under impedance mismatch conditions; (2) It significantly improves the dynamic response speed of the system; (3) It has the advantages of both autonomous operation and collaborative control.

[0051] The present invention will be further described below with reference to the accompanying drawings and specific preferred embodiments, but this does not limit the scope of protection of the present invention.

[0052] This invention discloses a reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control, comprising the following steps:

[0053] S1. A grid-connected converter system consisting of two parallel-operating units (e.g.) Figure 1The diagram shows the topology of a single grid-connected inverter, whose control architecture employs a hierarchical design strategy. In the outer power control layer, the system has two loops: active power control and reactive power control. By sampling the grid-connected current and voltage, and processing them through Park transform and a first-order low-pass filter, the actual active power P output by the inverter is calculated in real time. i and reactive power Q i :

[0054] ;

[0055] Among them, U od with U oq The d-axis and q-axis components of the voltage at the grid connection point of the i-th grid-connected converter are I. od with I oq These are the d-axis and q-axis components of the current at the grid connection point of the i-th grid-connected converter.

[0056] Then calculate the actual active power P i and reactive power Q i Compared with given active power and reactive power reference values, the reference signal required for the inner loop is generated through the following two control channels: 1) Active-frequency control channel, which calculates and outputs a phase reference value θ based on the active power deviation through the regulator. i 1) Used for the output voltage phase of synchronous inverters; 2) Reactive power-voltage control channel, which generates a voltage amplitude reference value V after processing reactive power deviation. iref As the reference for inner loop voltage control, the formula is expressed as:

[0057] ;

[0058] ;

[0059] In the formula ω iref ω0 is the actual angular frequency, V0 is the rated angular frequency of the system, and P is the rated amplitude of the grid connection point voltage. ref With Q ref These are the reference values ​​for active power and reactive power, respectively, k q k is the reactive voltage droop factor. p This is the active power droop coefficient.

[0060] S2. Constructing the inner-loop control of a network-based control strategy. Currently, commonly used inner-loop control methods include voltage-current dual closed-loop control and single-voltage amplitude control. This invention adopts an inner-loop architecture design based on single-voltage amplitude control. Figure 2 This is the topology diagram for the inner-loop voltage amplitude control. In this control strategy, the inner loop uses a proportional-integral regulator to achieve the grid-connected point voltage amplitude (V). iCompared to the zero-error tracking of the reactive power control loop reference output, the d-axis modulation signal is directly generated, thus eliminating the need for an inner current loop design. The specific process is as follows: Based on the voltage amplitude reference value V obtained in step S1... iref The grid connection point voltage (V) is controlled by a proportional-integral regulator. pcc Perform closed-loop control to generate the converter output voltage amplitude d-axis command E. id：

[0061] ;

[0062] In the formula E id V is the voltage amplitude at the output terminal of the converter. pcc k represents the actual voltage amplitude at the grid connection point. v This is the voltage regulation coefficient;

[0063] Command the corresponding q-axis E iq If E is 0, then E id With E iq The signal is further fed to the inverse Park transform module to be converted into a modulated waveform. Finally, the pulse width modulation module generates a drive signal to control the IGBT's turn-on and turn-off.

[0064] S3, Based on reactive power Q i Construct the feedforward compensation term V comp1 Specifically:

[0065] V comp1 = kQ i ;

[0066] k is the feedforward coefficient;

[0067] The amplitude control loop after adding a feedforward compensation term can be expressed as:

[0068] ;

[0069] S4. To achieve more precise parallel reactive power sharing control, a communication-based coordination compensation mechanism is configured in the system. The two inverters exchange their respective reactive power information Q via the CAN bus. i (i=1,2,…,n). Calculate the average reactive power of the system based on the received data. And establish a power deviation ΔQ i =Q i - Q avg The integral regulator generates a feedback control component V. comp2：

[0070] ;

[0071] k1 is the integral coefficient. This feedback loop can effectively compensate for the limitations of using only local voltage feedforward compensation V. comp1 At the same time, the problem of insufficient steady-state adjustment accuracy caused by parameter mismatch or disturbance is addressed, thereby improving the accuracy and robustness of reactive power distribution in both dynamic and steady-state dimensions.

[0072] S5. The compensation methods involved in S3 and S4 are superimposed to generate the final voltage command, and then the voltage loop amplitude control stage generates a PWM modulation signal to achieve power equalization.

[0073] Figure 3 The figure shows the topology of the proposed hybrid coordinated control system combining feedforward compensation and average power feedback. The feedforward compensation term V... comp1 Feedback and compensation item V comp2 Directly superimposed on the voltage amplitude reference value V iref Then, the command is sent to the voltage inner loop controller, and the converter output voltage d-axis command is:

[0074] ;

[0075] Specifically, this can be expressed as:

[0076] ;

[0077] In the formula Q avg is the average of the reactive power of the converter itself and the reactive power received, k1 is the integral coefficient introduced by the feedback, and k is the voltage compensation coefficient.

[0078] The final expression can be simplified to:

[0079] ;

[0080] V comp_final V represents the sum of the two compensation methods. comp_final =V comp1 +V comp2 ;

[0081] Command the corresponding q-axis E iq =0 for E id With E iq The modulated wave m is obtained by performing an inverse Park transform. abc Then, a drive signal is generated through pulse width modulation to control the switching on and off of the inverter's IGBTs.

[0082] This invention focuses on the reactive power sharing problem in multi-grid converters using single-voltage-loop amplitude control. Due to the fundamental differences in output characteristics between single-voltage-loop amplitude control and dual-loop voltage-current control, traditional virtual impedance methods cannot be directly applied to single-loop control systems to achieve line impedance regulation and reactive power sharing. To address the difficulty of achieving precise reactive power distribution under conditions of converter outlet line impedance mismatch and load fluctuations, this invention provides a detailed comparative analysis of the differences in virtual impedance regulation effects between single-loop and dual-loop control, explaining why single-loop control cannot achieve power sharing by adding traditional virtual impedance. In terms of implementation, firstly, in single-voltage-loop amplitude control mode, voltage and current signals at the grid connection point are acquired, active and reactive power are calculated, and reference values ​​for phase and voltage amplitude are generated based on droop control. Subsequently, a preliminary PWM modulation signal is generated through voltage loop control combined with the phase reference. To further improve power sharing, local reactive power is introduced into the voltage control stage for feedforward compensation, initially correcting the voltage command. Then, the system average reactive power is obtained through communication and used for feedback compensation to further optimize the voltage command. Finally, the feedforward and feedback compensation results are superimposed to form the final voltage command and optimized PWM signal, which drive the converter to achieve precise reactive power sharing. This invention proposes a hybrid voltage feedforward compensation method that effectively achieves precise reactive power sharing within the droop control framework through dynamic and precise compensation of the grid connection point voltage.

[0083] To verify the reactive power sharing performance of the proposed hybrid compensation control strategy in a grid-connected converter with single-voltage loop amplitude control, a high-fidelity system simulation model was constructed based on the MATLAB / Simulink simulation platform. The core of this model is a microgrid system consisting of two grid-connected converters operating in parallel. Its control architecture strictly follows the hierarchical design proposed in this invention: the outer loop uses classic P-ω / QV droop control to generate voltage amplitude and phase reference commands; the inner loop uses the single-voltage loop amplitude control focused on in this invention, directly regulating the voltage amplitude at the grid connection point in a closed loop, eliminating the traditional current inner loop and simplifying the system structure. To accurately simulate the key factors leading to uneven reactive power distribution in actual engineering, we deliberately set differentiated line impedance parameters, artificially creating the core contradiction of voltage deviation at the grid connection point caused by impedance mismatch. Simulation verification employed a control variable comparison method. Under identical system parameters (including droop coefficient, PI regulator parameters, etc., see the parameter table in the manual), the performance of three control strategies—feedforward compensation only, average power feedback compensation only, and a hybrid compensation combining feedforward and feedback—was tested sequentially. The test conditions were designed to consider both steady-state and dynamic processes: the system initially operated at a steady-state point (10kW active load, 2kVar reactive load); subsequently, two load step disturbances with amplitudes of ΔP=10kW and ΔQ=2kVar were applied at t=1.0 seconds and t=1.5 seconds respectively, to rigorously examine the dynamic response and recovery capability of the control system under sudden load changes. Simulation results clearly show that while the single feedforward compensation strategy possesses a fast dynamic response (adjustment time approximately 100ms), its steady-state accuracy is significantly affected by the compensation coefficient, with an error of approximately 3%; while the single average power feedback strategy achieves extremely high steady-state accuracy (error <0.1%), its dynamic response is slow (adjustment time approximately 300ms) and its initial disturbance is large. The hybrid compensation strategy ultimately adopted in this invention successfully combines the advantages of both approaches. It utilizes feedforward compensation to quickly offset most of the initial disturbances, and then uses feedback compensation to finely eliminate steady-state errors, thereby achieving the best balance between dynamic performance and static accuracy. The steady-state error is strictly controlled within 0.5%, the dynamic adjustment time is significantly shortened to 100ms, and it exhibits excellent robustness under continuous disturbances.

[0084] In this invention, the relevant system simulation data are set as follows:

[0085] Table 1. Relevant parameters of the simulation system

[0086]

[0087] Figure 4 The figure shows the result using only feedforward voltage compensation (V). comp1The simulation waveform results are shown in the figure. Experimental data show that when the system introduces a feedforward compensation mechanism at t=0.5s, it can effectively reduce the reactive power distribution error, reducing the error amplitude from 15% before compensation to about 3%. However, it can be seen from the figure that this compensation strategy has two key limitations: 1) the selection of k value is somewhat biased, resulting in insufficient reactive power distribution accuracy; 2) the fixed compensation coefficient k is difficult to adapt to dynamic operating conditions.

[0088] Figure 5 Simulation waveforms using only the average power feedback control strategy are shown. Simulation results indicate that this control strategy has the following typical characteristics: 1) It can achieve accurate reactive power allocation (error <0.1%) under steady-state conditions, verifying the theoretical advantages of feedback control; 2) When a reactive power load is introduced at t=1, the settling time reaches 300ms, significantly longer than the 100ms response time of the feedforward compensation strategy; 3) A maximum instantaneous power deviation of 15% occurs during dynamic processes. These phenomena reveal the inherent limitations of a single control strategy: although average power feedback can guarantee steady-state accuracy, it is difficult to simultaneously meet the requirement of fast response. Therefore, a hybrid control strategy is necessary; only by organically combining the speed of feedforward with the accuracy of feedback can the synergistic optimization of dynamic and static performance be achieved.

[0089] Figure 6 Simulation results of a hybrid strategy combining feedforward voltage compensation and average power feedback control are presented. Comparative analysis reveals the following: 1) In terms of steady-state accuracy, the hybrid control keeps reactive power distribution error below 0.5%, a significant improvement over single feedforward compensation; 2) Dynamic performance is significantly improved, with the settling time under load step disturbances reduced to 100ms, a 66% improvement over the pure average power feedback strategy; 3) The system exhibits good robustness under continuous disturbances, with instantaneous deviations during secondary load switching not exceeding 5%. The fast response characteristics of feedforward compensation effectively suppress initial disturbances, while feedback control ensures steady-state accuracy. Their synergistic effect achieves global optimization of dynamic and static performance.

[0090] Figure 7 The simulation waveform of active power allocation under the hybrid control strategy is shown. It can be seen that active power can also be accurately allocated and has good dynamic performance. Therefore, under the proposed control strategy, accurate allocation of both active and reactive power can be well achieved.

Claims

1. A reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control, characterized in that: Includes the following steps: S1. In a grid-type inverter system employing single-voltage loop amplitude control, the voltage and current signals output by the inverter are acquired, processed by Park transform and a first-order low-pass filter, and the actual output active power P is calculated. i With reactive power Q i Processing active power P based on droop control loop i With reactive power Q i Generate voltage amplitude reference value V iref and voltage phase angle θ i ; S2. Based on voltage amplitude reference value V iref Control the inner loop of the single voltage loop and generate the initial modulation signal; S3. Based on reactive power Q i Construct the feedforward compensation term V comp1 ; S4. Construct a feedback compensation term V based on average reactive power. comp2 ; S5. The feedforward compensation term V comp1 Feedback and compensation item V comp2 With V iref By superimposing and repeating step S2, the final modulation signal is generated, which drives the converter output to achieve power equalization.

2. The reactive power sharing method for multi-grid converters under single-voltage loop amplitude control according to claim 1, characterized in that: In a grid-type inverter system employing single-voltage loop amplitude control, the voltage and current signals output by the inverter are collected, and the actual output active power P is calculated. i and reactive power Q i Specifically: ； Among them, U od with U oq The d-axis and q-axis components of the voltage at the grid connection point of the i-th grid-connected converter are I. od with I oq These are the d-axis and q-axis components of the current at the grid connection point of the i-th grid-connected converter.

3. The reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control according to claim 1 or 2, characterized in that: Processing active power P based on droop control loop i and reactive power Q i Generate voltage amplitude reference value V iref and voltage phase angle θ i Specifically: The voltage amplitude reference value V is generated by the reactive power-voltage droop circuit. iref The actual angular frequency is generated by the active-frequency droop element, and the voltage phase angle θ is obtained after integration. i The formula is expressed as: ； ； In the formula ω iref ω0 is the actual angular frequency, V0 is the rated angular frequency of the system, and P is the rated amplitude of the grid connection point voltage. ref With Q ref These are the reference values ​​for active power and reactive power, respectively, k q k is the reactive voltage droop factor. p This is the active power droop coefficient.

4. The reactive power sharing method for multi-grid converters under single-voltage loop amplitude control according to claim 1, characterized in that: Based on voltage amplitude reference value V iref The inner loop of the single voltage loop is controlled, and a modulation signal is generated, specifically as follows: Based on voltage amplitude reference value V iref The grid connection point voltage is controlled in a closed loop by a proportional-integral regulator to generate the converter output voltage amplitude d-axis command E. id ; Command the corresponding q-axis E iq =0 for E id With E iq The inverse Park transform is performed to convert it into a modulated waveform, and then a drive signal is generated via pulse width modulation to control the on and off of the inverter IGBT.

5. The reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control according to claim 4, characterized in that: Based on voltage amplitude reference value V iref The grid connection point voltage is controlled in a closed loop by a proportional-integral regulator to generate the converter output voltage d-axis command E. id for: ； In the formula E id The d-axis command for the converter output voltage, V pcc k represents the actual voltage amplitude at the grid connection point. v This is the voltage regulation coefficient.

6. The reactive power sharing method for multi-grid converters under single-voltage loop amplitude control according to claim 1, characterized in that: Based on reactive power Q i Construct the feedforward compensation term V comp1 Specifically: V comp1 = kQ i ; k is the feedforward coefficient; After adding the feedforward compensation term, the converter output voltage d-axis command can be expressed as: 。 7. The reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control according to claim 1, characterized in that: Construct a feedback compensation term V based on average reactive power comp2 Specifically, it includes: The reactive power Q of all parallel converters in the system is obtained through the communication network. i (i=1,2,…,n), calculate the average reactive power of the entire network. Then, calculate the reactive power Q of the i-th converter. i Deviation from average reactive power ΔQ i =Q i - Q avg This deviation ΔQ i The input is sent to the integral controller to generate the feedback adjustment quantity V. comp2 : ； k1 is the integral coefficient.

8. The reactive power sharing method for a multi-grid converter under single-voltage loop amplitude control according to claim 1, characterized in that: The feedforward compensation term V comp1 Feedback and compensation item V comp2 With V iref By superimposing and repeating step S2, the final modulation signal is generated, which drives the converter output to achieve power sharing. Specifically, this includes: The feedforward compensation term V comp1 Feedback and compensation item V comp2 Directly superimposed on the voltage amplitude reference value V iref Then, the command is sent to the voltage inner loop controller, and the converter output voltage d-axis command is: ； Command the corresponding q-axis E iq =0 for E id With E iq The modulated wave m is obtained by performing an inverse Park transform. abc Then, a drive signal is generated through pulse width modulation to control the switching on and off of the inverter's IGBTs.

Images