CN110048423B - Current control method for immune power grid voltage harmonic interference - Google Patents

Abstract

The invention discloses a current control method for immune power grid voltage harmonic interference, which is characterized in that under the condition of power grid voltage harmonic influence, grid-connected current waveforms are distorted, so that the tracking performance of current is influenced, and finally the quality of grid-connected current is reduced. In order to solve the problem, the invention provides a harmonic compensation and improved voltage phase-locked loop control method. According to the method, a moving average filter MAF module is added in a traditional three-phase-locked loop to eliminate the influence of harmonic waves on phase locking. In addition, the grid voltage feedforward is adopted to ensure the stable work and operation of the grid current of the inverter under the condition of a non-ideal grid. The inverter side adjusts the current through a fundamental wave control loop and a harmonic compensation module, wherein the harmonic compensation module solves the influence of harmonic waves on current distortion, the fundamental wave control loop enables the system to be more stable, and finally the reliability of the grid-connected current of the inverter is guaranteed.

Description

A Current Control Method Immune to Power Grid Voltage Harmonic Interference

technical field

The invention belongs to the technical field of new energy power generation and micro-grid control in electric power systems, and relates to a method for controlling grid-connected current when new energy power generation is connected to a grid and the grid contains harmonics.

Background technique

In recent years, with the continuous consumption of non-renewable energy, the phenomenon of energy shortage is widespread. The development and utilization of new energy sources such as photovoltaics and wind power are accelerating day by day, and distributed power generation technology is receiving more and more attention. In the distributed generation system, the inverter plays an interface role in the energy conversion process between the renewable energy source and the grid, and also plays an extremely important role in the distributed generation system. However, due to the presence of harmonics in the grid voltage, grid-connected current distortion of three-phase grid-connected inverters generally occurs. At the same time, the normal operation of the grid-connected inverter requires accurate grid voltage phase and frequency information. Therefore, it is necessary to optimize the grid-connected current waveform and improve the tracking performance of the current to improve the power quality.

The traditional grid-connected inverter control strategy adopts the proportional-integral (PI) control method in the dq synchronous coordinate system to realize the tracking of the voltage phase frequency in a steady state without static error. However, in practical applications, it is often affected by factors such as DC side fluctuations, dead time of switching devices, and grid voltage harmonics, which distorts the grid-connected current waveform, weakens the ability of the current to track grid information, and causes serious harmonics to the grid. pollute. The Chinese patent with the authorized announcement number CN105763094A proposes an inverter control method based on voltage feedforward control and composite current control. This method uses the voltage prediction value of the common connection point PCC in the process of obtaining the reference voltage according to deadbeat control. As a voltage feed-forward, compound current control is adopted to avoid resonance of the LCL filter and improve the control accuracy of the control algorithm. However, this method does not involve the modulation of the current fundamental wave signal, nor does it mention the grid voltage phase-locking problem under the influence of grid voltage harmonics; the Chinese patent with the authorized announcement number CN104037800A proposes a photovoltaic grid-connected inverter current Control Method. This method uses PID current control to remove the differential component sensitive to interference signals and the integral component that cannot achieve zero static error of AC signals, so as to enhance the inverter's resistance to grid distortion and harmonic interference. However, this method does not mention the impact of grid voltage harmonics on the realization of accurate grid voltage fundamental wave amplitude, frequency and phase information extraction. Since these similar algorithms do not fully consider the impact of grid voltage harmonics on the entire grid-connected system, especially the impact on grid-connected current and grid voltage phase-locking links, it is necessary to study a three-phase grid voltage harmonic The current control technology in the environment can realize the immunity of grid-connected current to harmonics and can be widely used in power conversion technology, distributed grid-connected systems and other occasions.

Contents of the invention

The purpose of the present invention is to solve the deficiencies of the prior art, and propose a current control method immune to grid voltage harmonic interference. The technical problem to be solved by this method is the distortion of grid-connected current under the influence of grid voltage harmonics and the inability to accurately obtain grid voltage phase information. Moreover, the algorithm proposed by the invention realizes the elimination of grid-connected current waveform distortion, and the improved PLL module in the algorithm ensures the accuracy of obtaining grid voltage information and the synchronization of grid-connected current and grid phase information. Thereby improving the robustness of the whole control algorithm.

The specific technical solution of the present invention is: a current control method immune to grid voltage harmonic interference, specifically comprising the following steps:

S1, using the grid voltage E _a , E _b , E _c as the input signal of the three-phase phase-lock loop (PLL) module, and converting it to the synchronous rotating coordinate system to obtain the voltage E _d , E _q ;

和

S2. Establish a moving average filter (MAF) module in the PLL module, and then input E _d and E _q into the moving average filter module to eliminate harmonics, and obtain an output voltage after filtering

and

Further, the specific implementation method of converting the three-phase grid voltages E _a , E _b , E _c in S1 and S2 to E _{d , E q in} the synchronous rotating coordinate system is as follows:

Among them, θ is the grid voltage phasor θ _PLL sampled by the phase-locked loop, and E ₀ is the zero-sequence component. Establish the MAF module after E _d and E _q respectively, and its transfer function is:

Formula (2) is a continuous domain expression, formula (3) is a discrete domain expression. T _ω is the window length of the MAF, where formula (3) is a discrete time domain expression, where T _ω =NT _S , T _S is the sampling time, and N is the number of samples within a window length of the MAF. Substituting s=jω into (2), we get as follows:

where |G _m | is the gain factor of MAF. Therefore, the following conclusions can be obtained from formula (4):

From formula (5), it can be obtained that when ω=0, the gain of the MAF module is 1, and when f=k/T _ω (k=±1,±2,±3…), the gain is zero. In particular, the simulation results show that when the window length T _ω is equal to T and T/2, the filtering effect is more obvious, and the 5th, 7th, 11th, and 13th harmonics are basically filtered out.

S3. The purpose of establishing the phase lead compensator is to solve the problem of phase delay caused by MAF. The phase lead module is serially connected to the back of the MAF module, thereby effectively speeding up the system response capability and controlling the positive sequence component of the grid voltage to a certain extent. Compensation on the phase;

The transfer function of the phase lead compensator shown is:

In the formula, r is the attenuation factor, and its range is r∈[0,1), and k=(1-r ^N )/(1-r) is a normalized DC sampling gain.

S4. Input the grid voltage fundamental wave signal that has passed through the moving average filter (MAF) and the phase lead compensator to the proportional product integral controller (PI), and obtain the current cycle grid voltage frequency offset Δω _i . Then add Δω _i to the ideal grid voltage frequency ω ₀ to obtain the grid voltage frequency value ω _i in this cycle, and input the frequency value into the integrator to obtain the grid voltage phase value θ _PLL ` before compensation in this cycle; The displacement Δω <sub>i</sub> passes through a constant gain value k <sub>φ</sub> to compensate the phase error of the Park transformation, then the difference between the grid voltage phase value θ <sub>PLL</sub> ` and (k _φ *Δω _i ) before compensation in this cycle is used as the phase of the Park transformation ;

The phase shift caused by MAF can be equivalent to:

Since when the system reaches a steady state, the phase offset of the PLL is θ=θ _PLL `-k <sub>φ</sub> Δω <sub>i</sub> , and the rotation angle of the Park transformation is changed to (θ <sub>PLL</sub> `-k _φ Δω _i ), realizing that in the steady state The input signal of PI is equal to 0. Through the above control, the θ _PLL with zero error phase can be output when the steady state is reached.

S5, use the tracked grid voltage phase θ _PLL as the rotation angle of the three-phase grid voltage E _a , E _b , E _c to perform Park transformation, and obtain the voltage feedforward amount E _dq in the synchronous rotating coordinate system.

S6, the three-phase inverter-side current data i _a , i _b , and _ic obtained by sampling are transformed by Clark into the inverter-side current i _αβ on the αβ coordinate axis, and the grid voltage phase θ _PLL obtained in S3 is used as i _αβ Carry out the phase of Park transformation, convert to obtain the current i _dq in the synchronous rotating coordinate system, subtract the i _dq sampled by the inverter side from the ideal rated value current i _dq ^* to obtain the deviation Δi _dq , and input it to the proportional-integral controller (PI) and then output the inverter side fundamental wave voltage modulation signal ΔV _dq ;

Further, the implementation method of obtaining ΔV _dq in step S6 is:

The expression of the s domain of the proportional controller is: G _PI (s)=K _p +K _i /s.

S7, add the voltage feed-forward amount E _dq obtained in S5 to the inverter-side fundamental voltage modulation signal ΔV _dq obtained in S6, and then perform Park inverse transformation, and use the grid voltage phase θ _PLL obtained in S4 as its transformation phase. After inverse transformation, the voltage fundamental wave modulation signal V _αβ is obtained;

In S8, the ideal current i _dq ^* is subjected to inverse Park transformation from rotating coordinates to i αβ ^* under the αβ coordinate axis, and then i _{αβ *} is subtracted from the current _{i αβ} ^obtained by sampling the inverter side in S6 to obtain The current deviation Δi _αβ under the αβ coordinate axis;

S9, using the current deviation Δi _αβ obtained in S8 and the grid voltage phase θ _PLL obtained in S4 as the input quantity of the harmonic compensator. Multiply the current deviation Δi _α by the cosine value of -5 times the measured grid voltage phase θ _PLL , multiply the current deviation Δi _β by the sine value of -5 times, and then add the two product values to obtain Δi _q ; Multiply the current deviation Δi _α by the sine value of -5 times the measured grid voltage phasor θ _PLL , multiply the current deviation Δi _β5 by the cosine value of -5 times, and then subtract the two product values to obtain Δi _d5 ;

The implementation method of the S9 step is:

S10, input the Δi _dq5 obtained in S9 into the proportional-integral controller (PI), output the 5th harmonic modulation signal under the rotating coordinate system, and input the modulation signal obtained in this cycle into the saturation limiter to ensure that the amplitude is within a certain range Inside. If the amplitude is within the rated range, the output of the saturation limiter is zero, otherwise the excess will be adjusted to the rated range by the proportional-integral controller to finally obtain the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^;

The specific implementation of the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^ obtained in step S10 is:

Among them, G _PI (t)=K _p e(t)+K _i ∫e(t)dt, where K _p is the proportional coefficient, and K _i is the integral time constant.

The expression of the saturation limiter in the S9 step is as follows:

The value of u ₁ is the feedback current value, and u ₂ is a constant value of 40. The meaning of formula (11) is that if u ₁ ∈ (-u ₂ , u ₂ ), the feedback value u is zero, and the original value of u ₁ is output. Otherwise, the difference between u ₁ and u ₂ is PI adjusted until u ₁ is controlled within the limit range u ₂ .

S11, multiply the 5th harmonic modulation signal ΔV _d5 ^ obtained in S10 by the cosine value of -5 times the measured grid voltage phasor θ _PLL , and multiply the harmonic modulation signal ΔV _q5 ^ by the sine value of -5 times Multiply, and then subtract the two product values to get ΔV _α5 ^; multiply the 5th harmonic modulation signal ΔV _d5 ^ with the sine value of -5 times the measured grid voltage phasor θ _PLL , and modulate the signal ΔV Multiply _q5 ^ with the cosine value of -5 times, and then add the two product values to get ΔV _β5 ^;

The implementation of step S11 is:

Further, combined with the previous formulas (9), (10), and (12), the realization method is obtained as follows:

Among them, k is the 5th, 7th, 11th, and 13th harmonic, that is, the modulation methods of each harmonic are similar.

S12, add the voltage fundamental modulation signal V _αβ obtained from the single current closed-loop control and the voltage feedforward modulation signal and the harmonic modulation signal ΔV _αβ ^ obtained from the harmonic compensator, and then undergo Clark inverse transformation to obtain a three-phase voltage modulation signal , and then construct the trigger signal required by the H-bridge IGBT of the inverter after pulse width modulation (Pulse width modulation, PWM).

Compared with prior art, the beneficial effect of the present invention is:

1. In the traditional inverter grid-connected system, the present invention modulates the fundamental wave and harmonic of the grid-connected current respectively. The fundamental wave modulation ensures the stability of the grid-connected current, and the harmonic modulation eliminates the grid voltage harmonics. The impact of waves on the grid-connected current;

2. The improved voltage phase-locked loop includes a MAF module that eliminates the influence of grid voltage harmonics. In addition, a phase compensator is designed to compensate the phase, which realizes accurate extraction of grid voltage information and enhances the robustness of the phase-locked loop. Therefore, the accuracy of the current control of the grid-connected system is improved.

Description of drawings

Figure 1 The overall functional block diagram of the grid-connected current control system of the grid-connected inverter;

Fig. 2 Principle block diagram of grid voltage fundamental wave frequency and phase extraction;

Figure 3 is a schematic block diagram of the topology of a specific harmonic elimination module;

Fig. 4 moving average filter (MAF) when window length value T _ω is equal to T and T/2 filtering effect Bode figure;

Figure 5. The Bode diagram of the phase compensation effect of MAF cascaded with three phase lead compensators with different r values;

Fig. 6 is the simulation waveform of the three-phase grid-connected current (I _A , I _B , I _C ) before adding the harmonic elimination module and grid-side voltage feed-forward;

Figure 7 is the three-phase grid-connected current (I _A , I _B , I _C ) simulation waveform before adding the harmonic elimination module;

Figure 8 is the simulation waveform of the three-phase grid-connected current (I _A , I _B , I _C ) before adding the grid-side voltage feed-forward amount;

Figure 9 is the simulated waveform of the three-phase grid-connected current (I _A , I _B , I _C ) after adding the grid-side voltage feed-forward amount and the harmonic elimination module;

Figure 10 is a comparison diagram of the simulated waveforms of the three-phase grid-connected active power and reactive power before and after adding the grid-side voltage feed-forward amount.

Detailed ways

The embodiments of the present invention are described in detail below in conjunction with the accompanying drawings: this embodiment is implemented on the premise of the technical solution of the present invention, and detailed implementation methods and specific operating procedures are provided, but the protection scope of the present invention is not limited to the following the described embodiment.

Figure 1 shows the overall functional block diagram of the three-phase grid-connected inverter grid-connected current control system. Under the influence of grid voltage harmonics, the grid-connected current waveform will be distorted, which will affect the tracking performance of the current, and eventually lead to the decline of the quality of the grid-connected current. Therefore, it can be seen from Figure 1 that the whole system mainly includes the improved phase-locked loop module on the grid side, the fundamental wave control loop of the inverter side current and the specific harmonic adjustment module. Among them, the improved phase-locked loop module can accurately and quickly extract the phase of the grid voltage; the current fundamental wave control loop corrects the deviation caused by the harmonic voltage; the specific harmonic adjustment module eliminates the harmonic content in the grid-connected current, In turn, the grid-connected current quality meets the grid-connected standards.

According to the principle block diagram in Figure 1, implement according to the following steps:

S1, using the grid voltage E _a , E _b , E _c as the input signal of the three-phase phase-lock loop (PLL) module, and converting it to the synchronous rotating coordinate system to obtain the voltage E _d , E _q ;

和

S2. Establish a moving average filter (MAF) module in the PLL module, and then input E _d and E _q into the moving average filter module to eliminate harmonics, and obtain an output voltage after filtering

and

Further, Fig. 2 is a schematic block diagram of extracting the fundamental frequency and phase of the grid voltage, that is, the control block diagram of the PLL module. The specific implementation method of converting the three-phase grid voltages E _a , E _b , E _c in S1 and S2 to E _{d , E q in} the synchronous rotating coordinate system is as follows:

Among them, θ is the grid voltage phasor θ _PLL sampled by the phase-locked loop, and E ₀ is the zero-sequence component. Establish the MAF module after E _d and E _q respectively, and its transfer function is:

Formula (2) is a continuous domain expression, formula (3) is a discrete domain expression. T _ω is the window length of the MAF, where formula (3) is a discrete time domain expression, where T _ω =NT _S , T _S is the sampling time, and N is the number of samples within a window length of the MAF. Substituting s=jω into (2), we get as follows:

where |G _m | is the gain factor of MAF. Therefore, the following conclusions can be obtained from formula (4):

From formula (5), it can be obtained that when ω=0, the gain of the MAF module is 1, and when f=k/T _ω (k=±1,±2,±3…), the gain is zero. In particular, the simulation results shown in FIG. 4 show that when the window length T _ω is equal to T and T/2, the filtering effect is more obvious, and the 5th, 7th, 11th, and 13th harmonics are basically filtered out.

S3. The purpose of establishing the phase lead compensator is to solve the problem of phase delay caused by MAF. The phase lead module is serially connected to the back of the MAF module, thereby effectively speeding up the system response capability and controlling the positive sequence component of the grid voltage to a certain extent. Compensation on the phase;

The transfer function of the phase lead compensator shown is:

In the formula, r is the attenuation factor, and its range is r∈[0,1), k=(1-r ^N )/(1-r), which is a normalized DC sampling gain.

S4. Input the grid voltage fundamental wave signal that has passed through the moving average filter (MAF) and the phase lead compensator to the proportional product integral controller (PI), and obtain the current cycle grid voltage frequency offset Δω _i . Then add Δω _i to the ideal grid voltage frequency ω ₀ to obtain the grid voltage frequency value ω _i in this cycle, and input the frequency value into the integrator to obtain the grid voltage phase value θ _PLL ` before compensation in this cycle; The displacement Δω <sub>i</sub> passes through a constant gain value k <sub>φ</sub> to compensate the phase error of the Park transformation, then the difference between the grid voltage phase value θ <sub>PLL</sub> ` and (k _φ *Δω _i ) before compensation in this cycle is used as the phase of the Park transformation ;

The phase shift caused by MAF can be equivalent to:

Since when the system reaches a steady state, the phase offset of the PLL is θ=θ _PLL `-k <sub>φ</sub> Δω <sub>i</sub> , and the rotation angle of the Park transformation is changed to (θ <sub>PLL</sub> `-k _φ Δω _i ), realizing that in the steady state The input signal of PI is equal to 0. Through the above control, the θ _PLL with zero error phase can be output when the steady state is reached.

S5, active power has a large jump range, which will cause excessive energy impact and damage the grid-connected system, so voltage feedforward is introduced. As shown in Figure 1, the overall functional block diagram of the grid-connected inverter grid-connected current control system, the tracked grid voltage phase θ _PLL is used as the rotation angle of the three-phase grid voltage E _a , E _b , E _c for Park transformation, Get the voltage feedforward E _dq in the synchronous rotating coordinate system.

S6, the three-phase inverter-side current data i _a , i _b , and _ic obtained by sampling are transformed by Clark into the inverter-side current i _αβ on the αβ coordinate axis, and the grid voltage phase θ _PLL obtained in S3 is used as i _αβ Carry out the phase of Park transformation, convert to obtain the current i _dq in the synchronous rotating coordinate system, subtract the i _dq sampled by the inverter side from the ideal rated value current i _dq ^* to obtain the deviation Δi _dq , and input it to the proportional-integral controller (PI) and then output the inverter side fundamental wave voltage modulation signal ΔV _dq ;

Further, the implementation method of obtaining ΔV _dq in step S6 is:

和

为dq坐标系下的电流额定量。i_a、i_b和i_c为逆变器端三相电流。The expression of the s domain of the proportional controller is: G _PI (s)=K _p +K _i /s, where K _p is the proportional coefficient, K _i is the integral time constant,

and

is the current rating in the dq coordinate system. i _a , i _b and i _c are the three-phase currents at the inverter end.

S7, add the voltage feed-forward amount E _dq obtained in S5 to the inverter-side fundamental voltage modulation signal ΔV _dq obtained in S6, and then perform Park inverse transformation, and use the grid voltage phase θ _PLL obtained in S4 as its transformation phase. After inverse transformation, the voltage fundamental wave modulation signal V _αβ is obtained;

In S8, the ideal current i _dq ^* is subjected to inverse Park transformation from rotating coordinates to i αβ ^* under the αβ coordinate axis, and then i _{αβ *} is subtracted from the current _{i αβ} ^obtained by sampling the inverter side in S6 to obtain The current deviation Δi _αβ under the αβ coordinate axis;

S9, using the current deviation Δi _αβ obtained in S8 and the grid voltage phase θ _PLL obtained in S4 as the input quantity of the harmonic compensator. Multiply the current deviation Δi _α by the cosine value of -5 times the measured grid voltage phase θ _PLL , multiply the current deviation Δi _β by the sine value of -5 times, and then add the two product values to obtain Δi _q5 ; Multiply the current deviation Δi _α by the sine value of -5 times the measured grid voltage phasor θ _PLL , multiply the current deviation Δi _β5 by the cosine value of -5 times, and then subtract the two product values to obtain Δi _d5 ;

As shown in Figure 3, it is the 5th, 7th, 11th, 13th harmonic compensation control block diagram, and the implementation method of the S9 step is as follows:

S10, input the Δi _dq5 obtained in S9 into the proportional-integral controller (PI), output the 5th harmonic modulation signal under the rotating coordinate system, and input the modulation signal obtained in this cycle into the saturation limiter to ensure that the amplitude is within a certain range Inside. If the amplitude is within the rated range, the output of the saturation limiter is zero, otherwise the excess will be adjusted to the rated range by the proportional-integral controller to finally obtain the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^;

The specific implementation of the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^ obtained in step S10 is:

Among them, G _PI (t)=K _p e(t)+K _i ∫e(t)dt, where K _p is the proportional coefficient, and K _i is the integral time constant.

The Fcn module in the control block diagram shown in Figure 3 is the limiting saturator, and the expression of the saturation limiter in the S9 step is as follows:

The value of u ₁ is the feedback current value, and u ₂ is a constant value of 40. The meaning of formula (11) is that if u ₁ ∈ (-u ₂ , u ₂ ), the feedback value u is zero, and the original value of u ₁ is output. Otherwise, the difference between u ₁ and u ₂ is PI adjusted until u ₁ is controlled within the limit range u ₂ .

S11, multiply the 5th harmonic modulation signal ΔV _d5 ^ obtained in S10 by the cosine value of -5 times the measured grid voltage phasor θ _PLL , and multiply the harmonic modulation signal ΔV _q5 ^ by the sine value of -5 times Multiply, and then subtract the two product values to get ΔV _α5 ^; multiply the 5th harmonic modulation signal ΔV _d5 ^ with the sine value of -5 times the measured grid voltage phasor θ _PLL , and modulate the signal ΔV Multiply _q5 ^ with the cosine value of -5 times, and then add the two product values to get ΔV _β5 ^;

The implementation of step S11 is:

Further, combined with the previous formulas (9), (10), and (12), the realization method is obtained as follows:

Among them, k is the 5th, 7th, 11th, and 13th harmonic, that is, the modulation methods of each harmonic are similar.

S12, add the voltage fundamental modulation signal V _αβ obtained from the single current closed-loop control and the voltage feedforward modulation signal and the harmonic modulation signal ΔV _αβ ^ obtained from the harmonic compensator, and then undergo Clark inverse transformation to obtain a three-phase voltage modulation signal , and then construct the trigger signal required by the H-bridge IGBT of the inverter after pulse width modulation (Pulse width modulation, PWM).

In order to verify that the improved phase-locked loop proposed by the present invention can accurately obtain grid voltage phase and frequency information in the presence of grid voltage harmonics, Figures 4 to 5 are simulations of the MAF module and phase lead compensation module of the PLL module renderings. In order to verify the compensation effect of the specific harmonic compensation module on harmonics and the influence of grid-side voltage feedforward on the three-phase grid-connected current, Figure 6 shows the specific harmonic compensation module and grid-side voltage feedforward when other conditions remain unchanged. Waveform diagram of the grid-connected current of the first three phases. Fig. 7 is a waveform diagram of the three-phase grid-connected current before adding a specific harmonic compensation module while keeping other conditions unchanged. Fig. 8 is a waveform diagram of the three-phase grid-connected current before grid-side voltage feed-forward is added while keeping other conditions unchanged. Fig. 9 is a three-phase grid-connected current waveform diagram after adding a specific harmonic compensation module and grid-side voltage feed-forward. In order to verify the influence of grid-side voltage feedforward on three-phase grid-connected active power and reactive power, Figure 10 shows the three-phase grid-connected active power and reactive power before and after adding grid-side voltage feedforward while keeping other conditions unchanged Waveform comparison.

Based on the introduction of the operating conditions in Figures 4 to 10, the dynamic effects and comparative waveforms in Figures 4 to 10 will be described in detail below.

Figure 4 shows the Bode diagram of the voltage phase-locked loop system with MAF module. It can be observed that the phase margin is 43.3°, indicating that the improved phase-locked loop system is stable, and the harmonic components of the power grid are effectively filtered out, but It can be clearly observed that there is a significant phase delay while filtering out harmonics. Figure 5 shows the Bode diagram of the cascaded phase lead compensation module of the MAF module. Curve 1 is the frequency response of the cascaded phase lead compensation module, and the other curves are the frequency response of the MAF module cascaded with the phase lead compensation module . The r value of curve 2 is taken as 0.95, the r value of curve 3 is taken as 0.97, and the r value of curve 4 is taken as 0.99. It can be observed that the phase lead compensation module effectively offsets the phase delay caused by the MAF module and also eliminates harmonics, and the compensation effect of curve 4 is the best. Figure 6 shows the output of the waveform diagram of the three-phase grid-connected current (I _a , I _b , I _c ) lacking harmonic modulation signals and grid-side voltage feedforward signals in the entire system, within the interval of t=0～0.05s , a, b, c three-phase current amplitude fluctuations are large. At the same time, due to the influence of grid voltage harmonics, the three-phase current distortion is very serious, and this distortion always exists as time goes by. Figure 7 shows the output of the three-phase grid-connected current (I _a , I _b , I _c ) of the entire system in the absence of harmonic modulation signals. It can be seen that under the influence of grid voltage harmonics, the three-phase The current is clearly distorted. Figure 8 shows the output of the three-phase grid-connected current (I _a , I _b , I _c ) waveform diagram in the absence of grid-side voltage feedforward in the entire system. It can be seen that in the absence of grid-side voltage feedforward Under normal circumstances, the three-phase current amplitude jump is also very serious in the time period of 0-0.05s, and the jump amplitude of phase c is the largest. In the time period of t=0-0.04s, the three-phase current waveform is also There is distortion, and the degree of distortion is larger than that in Figure 7. However, due to the harmonic modulation signal, the distortion of the three-phase current waveform is gradually eliminated after 0.4s, and the current waveforms of the three phases a, b, and c become Smooth and symmetrical. Figure 9 shows the output waveform of the three-phase grid-connected current waveform of the whole system under the action of harmonic modulation signal and grid-side voltage feed-forward. During the time period of t=0～0.2s, the distortion of the current waveform is basically eliminated, and the situation of the jump of the current amplitude is also eliminated, and the current waveforms of the three phases a, b, and c become smooth and symmetrical.

Figure 10 shows the comparison of three-phase grid-connected active power and reactive power waveforms when the entire system lacks grid-side voltage feedforward. Figure A has a wide range of amplitude jumps of active power and reactive power within the time period of t=0~0.03s, while Figure B shows the waveforms of active power and reactive power after adding grid-side voltage feedforward, which is easy to observe Until the active power and reactive power are gradually stabilized with little fluctuation over time.

The comparison of the three-phase current waveform and the power waveform under the above-mentioned various control conditions shows that the modulation of the fundamental wave and harmonic wave of the grid-connected current proposed by the present invention eliminates the influence of the grid harmonic wave on the current waveform; the voltage feedforward makes The whole system is more stable and reliable; the improved voltage phase-locked loop improves the control precision of the whole system. This current control method immune to grid voltage harmonic interference can be widely used in distributed generation systems.

Those skilled in the art will appreciate that the embodiments described here are to help readers understand the principles of the present invention, and it should be understood that the protection scope of the present invention is not limited to such specific statements and examples. Those skilled in the art can make various other specific modifications and combinations based on the technical revelations disclosed in the present invention without departing from the essence of the present invention, and these modifications and combinations are still within the protection scope of the present invention.

Claims (1)

1. A current control method for immunity grid voltage harmonic interference, is characterized in that, comprises following specific steps: S1, use the grid voltage E _a , E _b , E _c as the input signal of the three-phase phase-locked loop module, and convert the grid voltage to the synchronous rotating coordinate system, namely E _d , E _q ; S2, establish a moving average filter module in the PLL module, and send the grid voltage E _d and E _q to the moving average filter MAF for harmonic elimination, and obtain the output voltage after filtering

and

S3, in order to solve the problem of phase delay caused by MAF, a phase lead compensator is established, and the phase lead module is connected in series behind the MAF module, thereby effectively speeding up the system response capability. In addition, the phase lead module also plays a role in the positive sequence component of the grid voltage To a certain phase compensation effect; S4. Input the grid voltage fundamental wave signal that has passed through the moving average filter and phase lead compensator to the proportional product integral controller PI to obtain the grid voltage frequency offset Δω _i in this cycle, and then compare Δω _i with the ideal grid voltage After the frequency ω ₀ is added, the grid voltage frequency value ω _i of this cycle is obtained, and the frequency value is sent to the integrator to obtain the grid voltage phase value θ _PLL ` before compensation in this cycle; in addition, the grid voltage frequency offset Δω <sub>i</sub> is used to pass A constant gain value k <sub>φ</sub> realizes the compensation of the phase error of the Park transformation, then the difference between the grid voltage phase value θ <sub>PLL</sub> ` before compensation in this cycle and k _φ *Δω _i is used as the phase of the Park transformation; S5, using the tracked grid voltage phase θ _PLL as the rotation angle of the three-phase grid voltage E _a , E _b , E _c to perform Park transformation, and obtain the voltage feedforward amount E _dq in the synchronous rotating coordinate system; S6, the three-phase inverter-side current data i _a , i _b , and _ic obtained by sampling are transformed by Clark into the inverter-side current i _αβ on the αβ coordinate axis, and the grid voltage phase θ _PLL obtained in S3 is used as i _αβ Carry out the phase of Park transformation, convert to obtain the current i _dq in the synchronous rotating coordinate system, subtract the i _dq sampled by the inverter side from the ideal rated value current i _dq ^* to obtain the deviation Δi _dq , and input it to the proportional-integral controller The PI then outputs the inverter-side fundamental wave voltage modulation signal ΔV _dq ; S7, add the voltage feed-forward amount E _dq obtained in S5 to the inverter-side fundamental voltage modulation signal ΔV _dq obtained in S6, and then perform Park inverse transformation, and use the grid voltage phase θ _PLL obtained in S4 as its transformation The phase of the voltage fundamental wave modulation signal V _αβ is obtained after inverse transformation; In S8, the ideal current i _dq ^* is subjected to inverse Park transformation from rotating coordinates to i αβ ^* under the αβ coordinate axis, and then i _{αβ *} is subtracted from the current _{i αβ} ^obtained by sampling the inverter side in S6 to obtain The current deviation Δi _αβ under the αβ coordinate axis; S9, use the current deviation Δi _αβ obtained in S8 and the grid voltage phase θ _PLL obtained in S4 as the input of the harmonic compensator, and use the current deviation Δi _α and the cosine of -5 times the measured grid voltage phase θ _PLL Multiply the current deviation Δi _β by the sine value of -5 times, and then add the two product values to get Δi _q ; the current deviation Δi _α and the measured grid voltage phasor θ _PLL -5 times Multiply the sine value of , multiply the current deviation Δi _β5 by the cosine value of -5 times, and then subtract the two product values to obtain Δi _d5 ; S10, input the Δi _dq5 obtained in S9 into the proportional-integral controller PI, and output the 5th harmonic modulation signal under the rotating coordinate system, and at the same time input the modulation signal obtained in this cycle into the saturation limiter to ensure that the amplitude is within a certain range, If the amplitude is within the rated range, the output of the saturation limiter is zero, otherwise the excess will be adjusted to the rated range by the proportional-integral controller to finally obtain the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^; S11, multiply the 5th harmonic modulation signal ΔV _d5 ^ obtained in S10 by the cosine value of -5 times the measured grid voltage phasor θ _PLL , and multiply the harmonic modulation signal ΔV _q5 ^ by the sine value of -5 times Multiply, and then subtract the two product values to get ΔV _α5 ^; multiply the 5th harmonic modulation signal ΔV _d5 ^ with the sine value of -5 times the measured grid voltage phasor θ _PLL , and modulate the signal ΔV Multiply _q5 ^ with the cosine value of -5 times, and then add the two product values to get ΔV _β5 ^; S12, add the voltage fundamental modulation signal V _αβ obtained from the single current closed-loop control and the voltage feedforward modulation signal and the harmonic modulation signal ΔV _αβ ^ obtained from the harmonic compensator, and then undergo Clark inverse transformation to obtain a three-phase voltage modulation signal , and then construct the trigger signal required by the inverter H-bridge IGBT after pulse width modulation PWM; Steps S1 to S4 are the process of accurately tracking the grid voltage phase. In this control process, the sliding average filter and phase lead compensator are used to eliminate the influence of grid harmonics on it. The specific process is as follows: The specific implementation method of converting the three-phase grid voltages E _a , E _b , E _c in S1 and S2 to E _{d , E q in} the synchronous rotating coordinate system is as follows:

Where θ is the grid voltage phasor θ _PLL sampled by the phase-locked loop, E ₀ is the zero-sequence component, and the MAF module is established after E _d and E _q respectively, and its transfer function is:

Formula (2) is an expression in continuous domain, formula (3) is an expression in discrete domain, T _ω is the window length of MAF, and formula (3) is an expression in discrete time domain, where T _ω = NT _S , T _S is the sampling time, N is the number of samples in a window length of MAF, and substituting s=jω into (2), it is obtained as follows:

where |G _m | is the gain factor of MAF, and the following conclusions can be obtained from formula (4):

From formula (5), it can be obtained that when ω=0, the gain of the MAF module is 1, and when f=k/T _ω (k=±1,±2,±3…), the gain is zero. In particular, The simulation results show that when the window length T _ω is equal to T and T/2, the filtering effect is more obvious, and the 5th, 7th, 11th, and 13th harmonics are basically filtered out; The transfer function of the phase lead compensator in the S3 is:

In the formula, r is the attenuation factor, and its range r∈[0,1), k=(1-r ^N )/(1-r), is a normalized DC sampling gain; The phase shift caused by MAF in S4 can be equivalent to:

In the formula, k _φ is the constant gain value, and Δω _i is the offset between the grid voltage frequency and the rated value. When the system reaches a steady state, the phase offset of the PLL is θ=θ _PLL `-k <sub>φ</sub> Δω <sub>i</sub> , the rotation angle of the Park transformation is changed to θ <sub>PLL</sub> `-k _φ Δω _i , and the input signal of PI is equal to 0 in the steady state. Through the above control, the θ _PLL with zero error phase can be output when the steady state is reached. ; Further, according to the simulation output, it can be seen that the active power output by the system jumps too much, and the energy impact will cause damage to the grid-connected system. Therefore, in step S5, the method of introducing voltage feedforward is used to stabilize the active power of the system; steps S6-S7 is the process of obtaining the fundamental voltage modulation signal and the voltage fundamental modulation signal V _αβ , wherein the specific implementation of step S6 fundamental voltage modulation signal ΔV _dq is:

The expression of the s domain of the proportional controller is: G _PI (s)=K _p +K _i /s, where K _p is the proportional coefficient, K _i is the integral time constant,

and

is the current rating in the dq coordinate system, i _a , i _b and i _c are the three-phase currents at the inverter end; Steps S8-S11 are the implementation methods of obtaining the 5th, 7th, 11th, and 13th harmonic modulation signals ΔV _αβ ^, and the realization methods of each harmonic modulation signal are the same, and the following takes the 5th harmonic modulation signal as an example: The specific implementation method of obtaining Δi _q and Δi _d5 in step S9 is:

The specific implementation of the 5th harmonic modulation signal ΔV _d5 ^ and ΔV _q5 ^ obtained in step S10 is:

Among them, G _PI (t)=K _p e (t)+K _i ∫ e (t)dt, wherein K _p is the proportional coefficient, K _i is the integral time constant; The expression of the saturation limiter in the S9 step is as follows:

The value of u ₁ is the feedback current value, u ₂ is a constant value of 40, the meaning of formula (11) is that if u ₁ ∈ (-u ₂ , u ₂ ), the feedback value u is zero, and the original value of u ₁ is output, otherwise , adjust the difference between u ₁ and u ₂ by PI until u ₁ is controlled within the limit range u ₂ ; Step S11 obtains ΔV _{α 5} ^ and ΔV _{β 5} ^ in a specific implementation manner as follows:

Furthermore, combined with the previous formulas (9), (10), and (12), the realization method of the harmonic modulation signal is obtained as follows:

Among them, k is the 5th, 7th, 11th, and 13th harmonics, that is, the modulation methods of each harmonic are similar; Finally, add the voltage fundamental modulation signal V _αβ obtained from the single current closed-loop control and the voltage feedforward modulation signal and the harmonic modulation signal ΔV _αβ ^ obtained from the harmonic compensator, and then undergo Clark inverse transformation to obtain the three-phase voltage modulation signal , and then construct the trigger signal required by the inverter H-bridge IGBT after pulse width modulation PWM.

Images