Self-adaptive frequency compensation droop control method for micro-grid inverter

Abstract

The invention discloses an adaptive frequency compensation droop control method for a microgrid inverter, and the method is used for the droop control of the inverter when a distributed microgrid DG module is connected to a power grid. The droop control method comprises the following steps: 1) collecting an alternating current quantity of an output end of a three-phase inverter, and converting the alternating current quantity from an abc coordinate system to a dq coordinate system through abc / dq coordinate conversion; 2) performing power calculation to obtain active power P and reactive power Q actually output by the inverter; 3) taking P and Q as input, and calculating the output voltage Um of the inverter and the frequency omega of the inverter through a droop equation; 4) obtaining a compensated frequency through an adaptive PI frequency compensation algorithm based on a fal function; calculating a voltage and current loop reference value according to the compensated frequency; 5) inversely transforming into ioa, b and c values under an abc coordinate system through a voltage loop, a current loop and a dq / abc coordinate, and taking the ioa, b and c values as space pulse width vector modulation input reference values; and 6) outputting the six-path PWM duty ratio of the three-phase inverter bridge through space pulse width vector modulation, and controlling the output of the inverter.

Description

Technical Field

[0001] The present invention relates to electric power technology, and in particular to a microgrid inverter adaptive frequency compensation droop control method, which is particularly suitable for microgrids based on clean energy such as solar energy and wind energy. Background Art

[0002] The problem of environmental pollution caused by traditional power generation is becoming increasingly serious. Microgrid technology based on clean energy such as solar energy and wind energy is gaining more and more attention to reduce the pollution of traditional energy generation to the environment. Therefore, since the introduction of distributed generation microgrid technology, it has developed rapidly with its unique advantages, gradually penetrated into the power system, and became an important part of the power grid. [1-6] .

[0003] Distributed power sources convert clean energy such as wind and solar power into electrical energy, and then convert the electrical energy into the AC power required for grid connection through an inverter device composed of power electronic components. Therefore, as the core unit of DC / AC inverter of distributed power sources, the control performance of the inverter plays a key role in the generation of high-quality electricity by distributed energy. It is the key to achieve voltage and frequency stability and reasonable power distribution, and to ensure the safe and reliable operation of microgrids. [7] .

[0004] When distributed power sources are running in parallel, traditional droop control can be used to simulate the droop characteristics of traditional generators, decouple active power, reactive power, voltage and frequency, adjust the voltage and frequency of the system, and reasonably distribute the load power among the distributed power modules, so that the system voltage and frequency can maintain stable operation. [8-10] However, due to the impedance, resistance, voltage amplitude and structural differences of different lines, power coupling problems will occur. When the load fluctuates, the traditional PV / Qf droop control cannot adjust the power distribution in time; large reactive circulating current may be generated between micro-sources; the output current and voltage of the inverter system will also fluctuate. [11-14] .

[0005] In the prior art, CN115065093A discloses "a method and system for adaptive droop control of a microgrid inverter", which introduces a virtual complex impedance to make the output impedance of the microgrid inverter inductive, obtains the droop control equation of the microgrid inverter at rated power, keeps the droop coefficient of the droop characteristic curve unchanged, converts the droop control equation into an adaptive droop control equation corresponding to the real-time power, calculates the difference between the adaptive droop control equation and the droop control equation, obtains the power deviation value of the microgrid inverter, inputs the power deviation value into a proportional-integral controller, obtains the power adjustment value, superimposes the power adjustment value to the rated power, obtains the latest output power of the microgrid inverter for output, thereby reducing the voltage amplitude and frequency fluctuations caused by load changes.

[0006] The main defect of this solution is that when calculating the actual output power according to the droop equation, the frequency parameter does not take into account the frequency change factor. The difference between the inverter output voltage amplitude and frequency and the large power grid, as well as the impact on the frequency parameters when the load changes, although the problem of controlling the inverter output power is solved to a certain extent through droop control, the phase difference between the inverter output voltage and current and the large power grid is not well handled. The inconsistency between the inverter output AC voltage frequency and the large power grid frequency is bound to lead to phase inconsistency. The microgrid grid connection will cause different degrees of pollution to the power grid. Summary of the invention

[0007] In order to solve the problems existing in the prior art, the present invention introduces adaptive frequency compensation based on traditional droop control. The adaptive PI combines the proportional integral link of the PI control with the fal function, takes the voltage frequency of the large power grid as a reference, outputs the compensation frequency parameters, forms the self-tuning optimization of the parameters, improves the system's rapid response and anti-disturbance performance under large-scale disturbances, calculates the voltage outer loop and current inner loop parameters according to the compensated frequency, and controls the output of the microgrid inverter. By controlling the dq coordinate system and the grid-side voltage vector E (grid voltage synthesis vector ) rotate synchronously, so that the inverter output voltage is in phase with the grid voltage, avoiding large current shocks during the grid connection process. It solves problems such as power coupling, power distribution, and distributed power circulation, and improves the stability of system operation. Finally, the rationality and effectiveness of the control strategy are verified through simulation.

[0008] The technical solution of the present invention is described as follows:

[0009] A microgrid inverter adaptive frequency compensation droop control method is provided, which is used for droop control of the inverter of a distributed microgrid DG module when the distributed microgrid DG module is connected to the power grid. The droop control method comprises the following steps:

[0010] 1) Collect the AC quantity at the output end of the three-phase inverter, and transform the AC quantity from the abc coordinate system to the dq coordinate system through abc / dq coordinate transformation;

[0011] 2) Perform power calculation to obtain the active power P and reactive power Q actually output by the inverter;

[0012] 3) The actual output voltage U of the inverter is obtained by using P, Q, rated active power P' and rated reactive power Q' through the droop equation m and the actual frequency ω of the inverter;

[0013] 4) Generate the compensated frequency ω* through frequency compensation; then calculate the voltage and current after frequency compensation;

[0014] 5) Obtaining modulation control signal through voltage and current double closed-loop control;

[0015] 6) A three-phase full-bridge inverter circuit that controls the inverter by pulse width modulation;

[0016] In step 3), the droop equation is as shown in equation 1-1,

[0017]

[0018] In the formula, m represents the active-frequency droop control coefficient, n represents the reactive-voltage droop control coefficient; P′, Q′, U′, ω′ are the rated active power, rated reactive power, rated voltage and rated frequency of the grid output by the inverter under the rated voltage frequency and amplitude of the grid, respectively; P, Q are the active power and reactive power actually output by the inverter. P represents the actual active power output of the inverter; Q represents the actual reactive power output of the inverter; u represents the output voltage of the inverter; ω represents the output voltage frequency of the inverter; U m =u,U m is the three-phase inverter output voltage u a,b,c The resultant vector magnitude of

[0019] In step 4), first, the grid voltage e is collected a,b,c And through abc / dq coordinate transformation, the grid voltage e a e b e c The coordinates are transformed into a dq coordinate system that rotates synchronously with the inverter output voltage. The angular velocity of the dq coordinate system is calculated according to the droop control output frequency, so that the dq coordinate system and the inverter output voltage form a synthetic vector Synchronous rotation. The inverter outputs a three-phase AC voltage with a frequency ω that is equal to the angular velocity of the dq coordinate system rotation.

[0020] The angular velocity of the dq coordinate system is calculated by the frequency ω, so that the dq coordinate system and the inverter output voltage synthesize the vector Synchronous rotation;

[0021] Control grid voltage synthesis vector Phase make The q-axis component of the vector is zero, and the direction of the vector coincides with the d-axis of the rotating dq coordinate system. q Greater than zero, indicating The phase leads the d-axis, then the angular velocity of the dq coordinate system is increased; when e q If it is less than zero, it means that the phase lags the d-axis, then the angular velocity of the dq coordinate system is controlled to decrease; finally, e q is zero, and the PI controller parameter completes the droop control frequency compensation ω*; ω* and ω are added and output.

[0022] In this article, u oa,b,c Equivalent to u oa 、u ob and u oc ;u od,q Equivalent to u od and u oq ;i oa,b,c Equivalent to i oa 、i ob and i oc ;u a,b,c Equivalent to u a 、u b and u c ;i od,q Equivalent to i od and i oq ;e a,b,c Equivalent to e a 、e b and e c ;e d,q Equivalent to e d and e q .

[0023] In particular, in response to the shortcomings of existing specific technologies, the present invention takes the voltage frequency of the large power grid as a reference, adopts adaptive PI combined with the fal function for frequency compensation, adjusts the droop control frequency parameters, and uses the compensated frequency parameters as the basis for calculating the input parameters of the inverter three-phase inverter bridge, controls the inverter output, and synchronizes the inverter output AC power phase with the large power grid phase through a phase-locked loop. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1-1 Indicates frequency / active power curve;

[0025] Figure 1-2 Indicates voltage / reactive power curve;

[0026] Figure 2-1 It represents the parallel main circuit structure of voltage source type three-phase inverter;

[0027] Figure 2-2 Represents the three-phase stationary coordinate system space vector diagram;

[0028] Figure 3-1 Represents the inverter control system block diagram;

[0029] Figure 3-2 Represents the current loop and voltage loop control block diagram;

[0030] Figure 3-3 Represents the composite vector diagram of the power grid voltage;

[0031] Figure 3-4 Represents the PI adaptive frequency compensation control block diagram;

[0032] Figure 3-5 represents the nonlinear adaptive PI controller structure;

[0033] Figure 4-1 Represents the Simulink simulation model block diagram;

[0034] Figure 4-2a1 and Figure 4-2a2 They represent fal input signal adaptive tracking respectively;

[0035] Figure 4-2b1 and Figure 4-2b2 Respectively represent the inverter output active and reactive power;

[0036] Figure 4-3a1 to Figure 4-3e2 Bode diagrams showing the system transfer function under conventional PI control compensation and the adaptive frequency compensation of the present invention respectively;

[0037] Figure 4-4 Indicates the inverter output voltage harmonic spectrum;

[0038] Figure 4-5a1 and Figure 4-5a2 Respectively represent the output voltage and current waveforms of the three-phase inverter;

[0039] Figure 4-5b Indicates the inverter output phase voltage / current waveform;

[0040] Figure 4-6a and Figure 4-6b They respectively represent the changes in output current and voltage waveforms when the system suddenly adds or removes loads;

[0041] Figure 4-7a It shows the output current change waveform of the inverter when a negative voltage is suddenly added under conventional PI control;

[0042] Figure 4-7b It represents the output current change waveform of the inverter when the adaptive compensation control suddenly adds negative current;

[0043] Figure 4-8a Represents the d-axis component i of the rotating coordinate system d Waveform;

[0044] Figure 4-8b Represents the q-axis component i of the rotating coordinate system q Waveform. DETAILED DESCRIPTION

[0045] The present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0046] Overview

[0047] When the distributed microgrid DG (Distributed Generation) module is connected to the power grid, the inverter is the core component. The traditional inverter adopts droop control, but lacks the damping inertia link. When the system load changes, it will cause the grid frequency and voltage fluctuations, and it is impossible to control the reasonable power output.

[0048] Therefore, the microgrid inverter adaptive frequency compensation droop control method of the present invention introduces an adaptive strategy for traditional droop control, calculates the current active power and reactive power on the basis of the adaptive algorithm, and adjusts the frequency and voltage of the inverter output through coordinate transformation, decoupling control and adaptive frequency compensation. By adjusting the dual closed-loop control current set value, the active and reactive power outputs are distributed, the adaptive algorithm is improved to calculate the inverter output AC frequency compensation value, the PI controller output is adaptively adjusted, and the droop control frequency adaptive compensation is realized, so that the inverter output current and voltage track the grid phase, and the grid stability is enhanced when connected to the grid.

[0049] Finally, the feasibility and advancement of this method are verified through Simulink simulation experiments.

[0050] 1. Droop control

[0051] In the prior art, when the power system fluctuates, it is often subjected to a primary frequency modulation to suppress the fluctuation. Droop control is a means of simulating the primary frequency modulation of the power system.

[15] The droop control controls the output voltage amplitude and frequency of the inverter by simulating the droop characteristic curve of the generator, reasonably distributes the output power, and maintains the stability of the voltage and frequency of the microgrid system.

[0052] When the resistance in the high-voltage line is large, the system output impedance can be approximated as pure inductive. When the system load changes, the relationship between the output active power and frequency, and the reactive power and voltage can be linearly described as follows: Figure 1-1 and Figure 1-2 The relational expression is shown in 1-1:

[0053]

[0054] m and n are the droop control coefficients of active power and reactive power respectively.

[0055] P' represents the rated active power, and P represents the active power actually output by the inverter;

[0056] Q' represents the rated reactive power, Q represents the reactive power actually output by the inverter;

[0057] U' represents the rated voltage, u represents the output voltage of the inverter;

[0058] ω' represents the rated frequency, and ω represents the frequency of the inverter.

[0059] 2. Mathematical model of microgrid parallel inverter

[0060] 2.1 Inverter main circuit structure

[0061] The research object of this invention adopts a three-phase voltage source inverter. The DC side of the inverter is equivalent to a voltage source, and the output is equivalent to a controlled voltage source.

[16] The main circuit adopts a three-phase full-bridge circuit, and the output is an LC filter circuit. oa 、u ob 、u oc is the capacitor voltage, i oa 、i ob 、i oc is the inductor current, Z L The main circuit of the three-phase inverter is as follows Figure 2-1 shown.

[0062] Three-phase inverter output voltage u a 、u b 、u c The phase difference is 120°, U m is their synthetic vector amplitude, and the spatial synthetic vector in the three-phase stationary coordinate system is as follows: Figure 2-2 The corresponding voltage / current calculation formula is shown in 1-2. In formula 1-2, ω is the fundamental frequency of the power supply (the fundamental frequency of the power supply in my country is generally 50Hz),

[0063]

[0064] The six power switch tubes of the three-phase inverter bridge of the inverter are driven by the SPWM signal generated by the sine wave modulation signal after carrier modulation to obtain three-phase AC power. The three sine waves with a phase difference of 120° are compared with the triangular carrier to generate a control signal, which controls the conduction and shutdown of the six IGBTs in the upper and lower bridge arms respectively, thereby generating three-phase AC power with the same frequency as the modulation wave at the load end. The output of the three-phase inverter is connected to the grid through the common connection line PCC, and the load Z L powered by.

[0065] 2.2. Space vector coordinate transformation

[0066] The voltage and current of each phase of the three-phase AC in the power grid are sinusoidal functions that change with time. It is difficult to directly calculate the voltage and current that change at all times. In order to simplify the calculation, Clark transformation and Park transformation are introduced to transform the abc three-phase stationary coordinate system to the αβ two-phase stationary coordinate system, and then transform it to the dq two-phase rotating coordinate system (Park transformation) for analysis. Finally, through the dq / abc inverse transformation, it is restored to the three-phase AC voltage or current.

[0067] 2.2.1 Three-phase stationary coordinate system to two-phase stationary coordinate system

[0068] To perform abc / dq transformation, first transform the abc three-phase coordinate system into the αβ two-phase stationary coordinate system through Clark transformation.

[17] .

[0069] After Clark transformation, we can get formula 2-1:

[0070]

[0071] The transformation matrix C 3 / 2 It is expressed as formula 2-2:

[0072]

[0073] 2.2.2 From two-phase stationary coordinate system to two-phase rotating coordinate system

[0074] The AC quantity in the αβ two-phase stationary coordinate system is transformed into the DC quantity in the dq two-phase rotating coordinate system for analysis through Park transformation.

[0075] i α 、i β 、i d 、i q The relationship is formula 2-3:

[0076]

[0077] Expressing it in matrix form, we have formula 2-4:

[0078]

[0079] Transformation Matrix C abc-dq Formula 2-5:

[0080]

[0081] Through abc / dq transformation, the three-phase capacitor voltage u output by the inverter is converted oa 、u ob 、u oc And the inductor current i oa 、i ob 、i oc Transform to the corresponding dq axis component u d 、u q 、i d 、i q , as shown in Equations 2-6 and 2-7.

[0082]

[0083] Then, in the two-phase dq rotating coordinate system, the expressions of active power and reactive power are as shown in equation 2-8,

[0084]

[0085] The inverse transformation from the dq two-phase rotating coordinate system to the abc three-phase stationary coordinate system is to invert the matrix equation to obtain the inverse transformation from the two-phase rotating coordinate system to the three-phase stationary coordinate system, as shown in Formula 2-9.

[0086]

[0087] 3. Voltage and current dual closed-loop control

[0088] 3.1 Dual Closed-Loop Inverter Control System Model

[0089] Take the neutral point of the grid as the reference ground, and the voltage u of each phase RL branch a and current i a The equation is shown in formula 2-10.

[0090]

[0091] Among them, e a The amplitude and phase of u are determined by the power grid. a Adjustable, so adjust u a When a As the voltage changes, the output current of the inverter can be controlled. a 、e b 、e c Respectively represent the three-phase voltage of the power grid.

[0092] In order to obtain the mathematical model of the three-phase grid-connected inverter, the equations of voltage and current on each phase RL branch are written in matrix form:

[0093]

[0094] Through mathematical derivation, it can be found that the voltage and current equation in the dq coordinate system is:

[0095]

[0096] In this way, the mathematical model of the grid-connected inverter is obtained.

[0097] After coordinate transformation, the current and voltage calculation formulas are shown in equations 2-6 and 2-7.

[0098]

[0099] The active power P and reactive power Q are calculated according to the droop equation to calculate the voltage U of the inner loop controller. m and the inverter frequency ω;

[0100] After adaptive PI frequency compensation, the compensated frequency ω* is generated. According to the compensated frequency ω*, the current and voltage in the stationary coordinate system are calculated by formula 1-1;

[0101] Then calculate the reference voltage u according to the coordinate transformation od,q * ;

[0102] u od,q * With feedback voltage u od,q The current loop reference current i is generated through the voltage (external) loop Ld,q * ;

[0103] i Ld,q * and the feedback inductor current i Ld,q The current (inner) loop generates i in the dq coordinate system od,q value;

[0104] i od,q Then, through the inverse transformation of dq / abc coordinates, it is converted into the three-phase coordinate system i oa,b,c value;

[0105] SVPWM module according to i oa,b,c The six-way PWM duty cycle is obtained to control the on and off of the six power tubes of the three-phase inverter bridge.

[0106] The entire inverter double closed-loop control system block diagram is as follows Figure 3-1 shown.

[0107] 3.2 Voltage-current decoupling based on rotating coordinates

[0108] After the transformation from the abc three-phase stationary coordinate system to the dq two-phase rotating coordinate system, the d-axis current component i d and the q-axis current component i q Mutual coupling, i d The changes in i q The input terminal affects i q In order to d and i q Independent control is performed, and a control quantity that offsets the coupling term of the model itself is added to the input of the inverter (model), and the actual input of the inverter model is transformed into:

[0109]

[0110] Among them, the right side of the equation is the control quantity, which acts together on the input end of the model. Substituting the above equation into formula (2-12), we get:

[0111]

[0112] Due to the cross-coupling terms of the last two terms on the right side of formula (2-13) and the model itself and the d and q axis components of the grid voltage (e d 、e q ) cancel each other out, the control quantity u d * 、u q * Can control each independently d 、i q , add current feedback and PI parameters:

[0113]

[0114] According to the above design, the block diagram of the voltage and current transfer function after decoupling is as follows: Figure 3-2 shown.

[0115] 3.2 Adaptive frequency compensation control

[0116] The output frequency of the traditional droop control inverter changes with the power. There is a phase difference between the output voltage and the grid voltage, which may cause a large impact current. The grid voltage vector The angular velocity of rotation in space is a constant value ω = 2πf, where f is the power frequency of 50Hz. Before connecting to the grid, the frequency is compensated to make the inverter output voltage in phase with the grid voltage.

[0117] The grid voltage e a e b e c The coordinates are transformed into a dq coordinate system that rotates synchronously with the inverter output voltage. The angular velocity of the dq coordinate system is calculated according to the droop control output frequency, so that the dq coordinate system and the inverter output voltage form a synthetic vector Synchronous rotation. Control grid voltage synthesis vector Phase make The q-axis component of is zero, and the vector direction coincides with the d-axis of the rotating dq coordinate system. The vector coordinate diagram is shown in Figure 3-3 As shown;

[0118] When the grid voltage vector The q-axis component of is greater than zero, indicating that the vector The phase leads the d-axis, and the angular velocity ω of the dq coordinate system should be controlled to increase; when The q-axis component of is less than zero, indicating that the phase lags the d-axis. The dq coordinate system ω should be controlled to decrease. The dq rotating coordinate system rotates synchronously with the inverter output composite voltage vector. ω is equivalent to the inverter output three-phase AC voltage angular frequency. Control the grid voltage composite vector The q-axis component of is zero, and the PI controller parameters are adaptively adjusted to complete the droop control frequency compensation. At the same time, the coordinate system is rotated to control value, assign i d * 、i q * value, controls the distribution of active power and reactive power output by the inverter. When the component i on the q axis q When is 0, the inverter outputs active power. The control block diagram is as follows Figure 3-4 shown.

[0119] The proportional-integral link of PI control is combined with the fal function to form the self-tuning optimization of parameters, which improves the system's rapid response and anti-disturbance performance under large-scale disturbance conditions. Figure 3-4 As shown in the figure, the controller structure includes a nonlinear PI controller and a differential follower TD. The nonlinear PI controller is used to adjust the error between the actual voltage value and the given value, and the differential follower realizes fast tracking and regulation control of the inverter output voltage phase. TD input signal The q-axis component e q , real-time tracking of grid voltage vector The q-axis component of the e q Fast tracking without overshoot. When the error changes, the nonlinear PI controller realizes self-tuning of PI parameters

[18] The structure of the nonlinear adaptive PI controller is as follows: Figure 3-5 shown.

[0120] The TD part of the controller is composed of a nonlinear saturation function sin, as shown in equation (3-1):

[0121]

[0122] Where: m is the function variable, and e is the input q , sgn(x) is the sign function:

[0123]

[0124] The improved fal function is shown in formula (3-3):

[0125]

[0126] Where: α is the factor affecting the tracking effect. When it decreases, the filtering effect becomes worse and the tracking effect becomes better; e is the input error. The signal error is inversely proportional to the feedback gain generated by the fal function. The application of the fal function enables the system to have good fast and stable performance.

[0127] The error range of the nonlinear inverter working process is small in steady state. In order to prevent large disturbances in special conditions, which will increase the error and lead to system instability, the fal function is improved and the amplitude is limited within the range of δ, where δ is the amplitude limiting parameter.

[0128] The output value ω* of the nonlinear PI link is:

[0129]

[0130] In the above formula: e is the difference between the frequency feedback value and the reference value; K p , K i are proportional and integral parameters, which are adjusted adaptively by the algorithm. p , G i are the gain coefficients of the proportional link and the integral link, which are adjusted to the best according to the operation results. Here, the set values ​​G p 1.2, G i is 1; 1 , α 0 Represents the tracking effect impact factor, 0<α 1 <α 0 <1;δ 0 , δ 1 Represents the filtering effect factor. 0 , δ 1 The values ​​are 0.01 and 0.015 respectively; α 0 , α 1 The values ​​are 0.15 and 0.3 respectively. The smaller the α value, the stronger the nonlinearity, and the larger the α value, the weaker the nonlinearity; the δ value determines the nonlinear range of the fal(e,α,δ) function. The larger the δ value, the larger the linear range, and the smaller the δ value, the smaller the linear range. The δ value is 1.

[0131] 4. Simulation analysis

[0132] 4.1 Simulation model construction

[0133] according to Figure 3-1 The double closed-loop inverter control system block diagram is shown in the figure. A Simulink simulation model is built. The simulation model structure diagram is as follows Figure 4-1 shown.

[0134] By calculating ω and u, u o * The voltage reference signal is obtained. In order to make the inverter output reach a given value, the voltage and current double closed-loop control is adopted. From the control block diagram, it can be seen that the power calculation is first performed to obtain the P and Q values, and then ω and u are calculated according to the droop control. After adaptive compensation, the output frequency ω* is calculated to obtain u o * ,u o* With voltage feedback u o As the double closed-loop input, the output current control parameters are modulated by the coordinate inverse transformation and the space vector PWM module to output six PWM duty cycles as the inverter input parameters, thereby controlling the inverter output.

[0135] 4.2 Simulation parameters and waveforms

[0136] 4.2.1 Simulation parameters

[0137] The basic parameter settings of the simulation model are shown in Table 4-1.

[0138] Table 4-1 Model parameters

[0139]

[0140] 4.2.2 Simulation waveform

[0141] Please see Figure 4-2a1 to Figure 4-8b .

[0142] 4.3 Simulation waveform analysis

[0143] The inverter control system is a nonlinear system. When the control is in steady state, the error between the output frequency and the set value is within a small range. When there are special circumstances, a large disturbance will occur, which will increase the error and cause the system to be unstable. Through frequency feedback, the change of frequency is controlled and then the change of phase is controlled. The tracking variable is changed from error to error change rate, the input value range is reduced, and the frequency difference is differentiated to achieve accelerated tracking.

[0144] By comparison Figure 4-2a1 and Figure 4-2a2 In the adaptive controller, when the error is greater than 1, the value of the improved fal function is limited to 0. ～ 1, which is obviously smaller than the fal function value, and will reduce the gain caused by large errors.

[0145] The power control controls the current reference signal. d * 、i q * The given value is combined with the current feedback value, and the duty cycle of the three-phase modulation wave is adjusted through double closed-loop control to adjust the inverter output active power and reactive power. The output active power is initially 10kW, jumps to 20kW in the middle, and the reactive power is 0. The simulation waveform is as follows Figure 4-2b1 and Figure 4-2b2 As shown in the figure, the output power can track the given signal well.

[0146] Figure 4-3a1 to Figure 4-3e2The Bode plots of the system transfer function under conventional PI control and the adaptive frequency compensation of the present invention are given. After compensation, the amplitude-frequency characteristic curve of the system loop gain crosses the 0dB line at a frequency of 1500Hz, and the loop gain decreases at -20dB / dec in the low frequency band. The phase margin is 70.6°, which is larger than that before compensation. After compensation, the system stability is increased and has better dynamic characteristics.

[0147] Figure 4-4 The figure shows the harmonic spectrum of the inverter output voltage. It can be seen from the figure that the total harmonic distortion of the voltage spectrum is 1.89%, the maximum 6th harmonic does not exceed 0.8%, the harmonic distortion rate is small, the output waveform is close to a sine wave, and the inverter conversion efficiency and system stability are good.

[0148] Figure 4-5a1 and Figure 4-5a2 The three-phase inverter output voltage and current waveforms are shown below. The voltage peak is 310V, the current peak is 60A, and the phase difference between each phase waveform is 120°. Figure 4-5b The figure shows the voltage and current waveform of one phase of the three-phase inverter output. It can be seen from the figure that the current has a small fluctuation within 0.03s of the initial output, and then the waveform is approximately a smooth sine wave and the waveform is stable.

[0149] When the system suddenly adds or removes load, the output current and voltage waveforms change as follows: Figure 4-6a and Figure 4-6b As shown in the figure, by comparing with the conventional PI control, when the load mutation occurs at 0.1s and recovers at 0.11s, the system tends to be stable at 0.1187s under the conventional PI control, the elapsed time is about 0.0087s, and the overshoot is large; the system tends to be stable at 0.1141s under the adaptive compensation control, the elapsed time is about 0.0041s, and the overshoot is small. The output waveform is as follows Figure 4-7a and 4-7b shown.

[0150] The grid voltage coordinates are transformed into the dq coordinate system of the inverter operation, and the d-axis and q-axis current component waveforms of the rotating dq coordinate system are as follows: Figure 4-8a and Figure 4-8b As shown, the q-axis current component is zero. When a sudden load is added, the output d-axis current waveform recovers within one cycle, and the q-axis component remains at 0 and the waveform is stable. The q-axis component of the dq coordinate system Synchronous rotation, after transformation, the d and q axis current component waves are stable, the dynamic performance is good, it meets the control requirements, and the inverter system is better controlled to distribute active and reactive power for stable output.

[0151] 5. Conclusion

[0152] The microgrid inverter adaptive frequency compensation droop control method of the present invention is used for droop control of the inverter of the distributed microgrid DG module when it is connected to the power grid. The droop control method includes the following steps: 1) collecting the AC quantity at the output end of the three-phase inverter, and transforming the AC quantity from the abc coordinate system to the dq coordinate system through abc / dq coordinate transformation; 2) performing power calculation to obtain the active power P and reactive power Q actually output by the inverter; 3) using P and Q as inputs, calculating the output voltage U of the inverter through the droop equation m and the inverter frequency ω; 4) after the adaptive PI frequency compensation algorithm based on the fal function, the compensated frequency is obtained; then the voltage and current loop reference value is calculated based on the compensated frequency; 5) after the voltage loop, current loop and dq / abc coordinate inverse transformation, it is converted into i in the abc coordinate system oa,b,c Value, as the reference value of space pulse width vector modulation input; 6) Space pulse width vector modulation outputs the six-way PWM duty cycle of the three-phase inverter bridge to control the inverter output.

[0153] The output voltage and current waveforms of the three-phase inverter based on adaptive frequency compensation droop control are smooth and stable during steady-state operation, and have good steady-state characteristics. The dual closed-loop control system has the ability to control the stable operation of the system. The closed-loop control sets the inverter load voltage reference value to 311V, and the actual output voltage is consistent with the voltage reference value. The control strategy has a high tracking accuracy for the system output and can better control the distribution of active power and reactive power output. When the load changes, the adaptive compensation control system can recover quickly, and the system has good dynamic characteristics.

[0154] By adopting the improved fal function and the adaptive compensation algorithm that tracks the frequency error change rate, the system can reach a steady state faster than conventional control. When the load changes suddenly, the adaptive controller enables the inverter output voltage to quickly and accurately track the reference voltage set value generated by the power loop. The dual closed-loop control system based on the adaptive compensation algorithm has a faster dynamic response speed and stronger anti-interference ability. The inverter system has better control accuracy and better steady-state and dynamic performance.

[0155] References:

[0156] [1] Zhang Xing, Li Ming, Guo Zixuan, et al. Review and Prospect of Research on Control Strategies of New Energy Grid-connected Inverters[J]. Global Energy Internet, 2021, 4(05): 506-515.

[0157] [2] Jian Fang, Li Zhiyong. Current status and development of photovoltaic inverter technology[J]. High Power Conversion Technology, 2014, (03): 5-9+19.

[0158] [3] Zou Jianzhang, Chen Qiaofu, Zhang Changzheng. Research on comprehensive performance test platform for photovoltaic inverters[J]. Electrical Measurement and Instrumentation, 2010, 47(08):4.

[0159] [4] Zeng Zheng, Yang Huan, Zhao Rongxiang, et al. A review of research on multifunctional grid-connected inverters[J]. Electric Power Automation Equipment, 2012, 32(08): 5-15.

[0160] [5] Xu Jinming, Xie Shaojun, Zhang Binfeng. A review of current control of LCL filter grid-connected inverters in distributed generation systems[J]. Proceedings of the CSEE, 2015, 35(16): 4153-4166.

[0161] [6] Gong Zanchao, Ju Zhenhe. Review of the development and application of solar photovoltaic power generation[J]. Shandong Industrial Technology, 2017(20):61.

[0162] [7]

[0163] [8] Zhang Xueguang, Fu Zhichao, Chen Wenjia, et al. Improved control method of grid-connected inverter considering the influence of phase-locked loop under weak power grid [J]. Automation of Electric Power Systems, 2018, 42(07): 139-145.

[0164] [9] Wang Siran, Lü Zhengyu. Research on repetitive control method in LCL grid-connected inverter[J]. Proceedings of the CSEE, 2010, 30(27): 69-75.

[0165]

[10] Zha Xiaoyu, Xu Junzhong, Wang Yong. Improved voltage full feedforward strategy for grid-connected inverter under weak power grid[J]. Power Electronics Technology, 2020, 54(04): 84-87.

[0166]

[11] Yang Xingwu, Wang Nannan, Chen Lei, et al. Robust repetitive control strategy for LCL grid-connected inverter under weak power grid[J]. Power Electronics Technology, 2018, 52(10): 15-18.

[0167]

[12] Xu Jinming, Bian Shenyiyang, Qian Hao, et al. Robust control and optimization method of delayed phase-locked loop of single-phase grid-connected inverter under weak power grid [J]. Proceedings of the CSEE, 2020, 40(07): 2062-2070.

[0168]

[13] Xu Jinming, Bian Shenyiyang, Qian Qiang, et al. Single-phase inverter phase-locked loop based on grid current feedforward in weak power grid[J]. Proceedings of the CSEE, 2020, 40(08): 2647-2657.

[0169]

[14] Midtsund T,Suul JA,Undeland T.Evaluation of current controllerperformance and stability for voltage source converters connected to a weakgrid[C] / / The 2nd International Symposium on Power Electronics for DistributedGeneration Systems.June16-18,2010,Hefei,China.IEEE,2010:382-388.

[0170]

[15] U.Borup,F.Blaabjerg,PNEnjeti.Sharing of Nonlinear Load inParallel-Connected Three-PhaseConverters.IEEE Transactions on IndustryApplications,2001,37(6):1817-1823.

[0171]

[16] Zhang Chongwei, Zhang Xing. PWM rectifier and its control. Beijing: Machinery Industry Press, 2003.

[0172]

[17] Guo Xiaoqiang. Research on control technology of three-phase grid-connected inverter in distributed generation system[D]. Qinhuangdao: Yanshan University, 2009.

[0173]

[18] Xu Jiazhu, Cheng Huijie, Huang Wen, et al. Low-frequency oscillation suppression method for high-speed railway vehicle-grid coupling system under adaptive anti-disturbance proportional-integral control[J]. Proceedings of the CSEE, 2018, 38(14): 4035-4045.

Claims

1. A microgrid inverter adaptive frequency compensation droop control method, the method is used for the droop control of the inverter when the distributed microgrid DG module is connected to the power grid, the droop control method comprises the following steps: 1) Collect the AC quantity at the output end of the three-phase inverter, and transform the AC quantity from the abc coordinate system to the dq coordinate system through abc / dq coordinate transformation; 2) Perform power calculation to obtain the active power P and reactive power Q actually output by the inverter; 3) The actual output voltage U of the inverter is obtained by using P, Q, rated active power P' and rated reactive power Q' through the droop equation m and inverter output voltage frequency ω; 4) Through the adaptive PI algorithm, the proportional-integral link of the PI control is combined with the FAL function to generate the compensated frequency ω* according to ω and the natural frequency of the power grid; then the input reference value of the voltage and current double closed loop is calculated according to the compensated frequency ω′; 5) The voltage loop input reference voltage calculated according to ω* generates the SVPWM modulation control signal through the voltage and current loop; 6) The pulse width modulation signal controls the inverter output through the three-phase full-bridge inverter circuit; In step 3), the droop equation is Where m represents the active-frequency droop control coefficient, n represents the reactive-voltage droop control coefficient; P′, Q′, U′, ω′ represent the rated active power, rated reactive power, rated voltage, and rated frequency of the grid output by the inverter under the rated voltage frequency and amplitude of the grid; P, Q represent the actual active power and reactive power output of the inverter, respectively; u represents the output voltage of the inverter; U m =u,U m is the three-phase inverter output voltage u a,b,c The resultant vector magnitude of It is characterized in that in step 4), 4.1) Collecting grid voltage e a,b,c , and transform the grid voltage e by abc / dq coordinate transformation a e b e c The coordinates are transformed into a dq coordinate system that rotates synchronously with the inverter output voltage; 4.2) Control grid voltage synthesis vector Phase make The q-axis component e q is zero, the vector direction coincides with the d-axis of the rotating dq coordinate system; the method is: when e q Greater than zero, indicating The phase leads the d-axis, then the angular velocity of the dq coordinate system is increased; when e q If it is less than zero, it means that the phase lags the d-axis, then the angular velocity of the dq coordinate system is controlled to decrease; finally, e q is zero; 4.3) Control droop control The inverter output voltage frequency ω calculated is synchronized with the angular velocity of the dq coordinate system rotation. Adaptive PI algorithm and voltage and current double closed loop are used to make the dq coordinate system and the inverter output voltage synthesize vector Synchronous rotation; The adaptive PI control completes the droop control frequency compensation and outputs the compensated frequency ω′ by adding ω* and ω.

2. The microgrid inverter adaptive frequency compensation droop control method according to claim 1, characterized in that In step 4), the PI controller is a nonlinear adaptive PI controller, which combines the proportional integral link of the PI control with the fal function to form a self-tuning optimization of the parameters, thereby improving the rapid response and anti-disturbance performance of the system under a wide range of disturbances; the adaptive PI controller includes a differential follower TD and a nonlinear PI controller; In TD, a nonlinear saturation function sin is used according to e q Calculate the output as follows Where: δ represents the limiting parameter, the sign function In the nonlinear PI controller, the frequency compensation ω* of the output of the nonlinear PI controller is: Where: e is the input error, which is inversely proportional to the feedback gain generated by the fal function; μ represents the rate of change of e; G i is the integral link gain coefficient, G p is the proportional link gain coefficient; fal(e,α,δ) is the improved fal function; α is the tracking effect influencing factor. The smaller the α value, the stronger the nonlinearity, and the larger the α value, the weaker the nonlinearity. α0 and α1 represent the tracking effect influencing factors of the integral link and the proportional link, respectively, 0<α1<α0<1. δ0 and δ1 represent the filtering effect influencing factors of the integral link and the proportional link respectively; δ is the limiting parameter, which determines the nonlinear interval range of the fal(e,α,δ) function. The larger the δ value, the larger the linear interval, and the smaller the δ value, the smaller the linear interval.

3. The microgrid inverter adaptive frequency compensation droop control method according to claim 2, characterized in that In step 4), in the nonlinear PI controller, the value of δ in the fal function is 1; G i The value of G is 1. p The value of is 1.2; α0 and α1 are 0.15 and 0.3 respectively; δ0 and δ1 are 0.01 and 0.015 respectively.

4. The microgrid inverter adaptive frequency compensation droop control method according to claim 2, characterized in that Step 1) The collected AC quantity includes the inverter output current i before LC filtering La,b,c , the inverter outputs the voltage u after LC filtering oa,b,c And the current i oa,b,c ;i La,b,c The feedback inductor current i is obtained by abc / dq coordinate transformation Ld,q ;u oa,b,c and oa,b,c The feedback voltage u is obtained by abc / dq coordinate transformation od,q and the feedback voltage i od,q .

5. The microgrid inverter adaptive frequency compensation droop control method according to claim 4 is characterized in that In step 6), according to the i obtained in step 5) oa,b,c The six PWM duty cycles are obtained through space pulse width vector modulation SVPWM, which respectively control the on and off of the six power tubes of the three-phase inverter bridge of the inverter.

Images