Power quality optimization control system and control method for network-forming inverter

Abstract

The invention discloses a grid-forming inverter output electric energy quality optimization control system and control method, and the method comprises the steps: replacing a PWM modulation module with model prediction control in a grid-forming inverter, extracting the voltage of a point of common coupling (PCC) and the harmonic component of a power grid current through the orthogonal characteristic of a sine function, predicting and calculating the harmonic voltage of a power grid, and carrying out the optimization of the output electric energy quality of the grid-forming inverter. The harmonic voltage obtained through calculation is multiplied by a coefficient and added into reference voltage, and model prediction control is used for modulation. The harmonic voltage amplitude on the power grid impedance is controlled by controlling the harmonic voltage multiplication coefficient, and flexible suppression of the network access current harmonic and the PCC voltage harmonic is achieved. The method is simple in structure and convenient to implement, the grid-connected inverter based on the method can achieve higher-quality electric energy output, and the method is suitable for various grid-forming type grid-connected inverters.

Description

Technical Field

[0001] The present invention relates to a grid-connected inverter system, and particularly to an optimized control system and control method for the output power quality of a grid-forming inverter. Background Art

[0002] A new energy distributed grid-connected power generation system includes local loads, energy storage devices, and distributed power generation devices (grid-forming inverters). It not only supplies power to local loads but also needs to be connected to the power grid. Therefore, in a new energy distributed grid-connected power generation system, how to improve the output voltage quality and grid-connected current quality of grid-forming inverters has received extensive attention and discussion.

[0003] Firstly, a new energy distributed grid-connected power generation system needs to supply power to local loads. Some of the local loads are non-linear loads that will generate harmonic currents, which will increase the total harmonic distortion rate (THD) of the PCC voltage and the THD of the grid-connected current. Secondly, a new energy distributed grid-connected power generation system is often located in remote areas, and there may be background harmonics in the power grid, which will also cause an increase in the PCC voltage THD and the grid-connected current THD. How to suppress the harmonics of the output voltage and grid-connected current of grid-forming inverters is a key problem that is currently being solved or will be solved.

[0004] Existing literature has studied the optimized control method for the output power quality of grid-forming inverters. Common methods include using a harmonic suppression scheme with grid-connected current feedforward to reshape the output impedance of the converter, thereby improving the output current quality of the converter; using a voltage feedforward scheme to reshape the output impedance of the converter to achieve suppression of PCC voltage harmonics or grid-connected current harmonics; some scholars also use a combination of voltage feedforward and current feedforward to achieve flexible suppression of PCC voltage harmonics or grid-connected current harmonics. Although the existing optimized control schemes for the output power quality of grid-connected inverters can achieve a certain degree of harmonic suppression, significant digital delay makes the harmonic suppression methods of current feedforward and voltage feedforward have phase errors, and at the same time, the dynamic response of traditional schemes is slow and the harmonic suppression effect is poor.

[0005] Therefore, for grid-forming inverters, it is necessary to study a harmonic suppression scheme with faster control speed, more accurate harmonic extraction and compensation, so as to improve the output power quality of the inverter and achieve high-quality grid connection of the inverter. Summary of the Invention

[0006] The object of the present invention is to provide an optimized control method for the output power quality of a grid-forming inverter. This method can achieve accurate harmonic extraction and suppression through model predictive control, reduce harmonic phase errors, and thus improve the output power quality of the inverter.

[0007] To achieve the above object, the solution of the present invention is:

[0008] A power quality optimization control system for a grid-forming inverter. The control system includes a grid-forming inverter, and the grid-forming inverter includes a bridge inverter topology and an inverter-side inductor L f , a filter capacitor C f , and is connected to the power grid after connecting to a local load.

[0009] Based on the above control system, the present application also provides a control method for it. The control method includes the following steps:

[0010] 1) Sample the inverter output voltage sampling signal u oabc , the grid-connected current sampling signal i gabc , and calculate the grid voltage harmonic signal u rehabc close to the grid voltage through model prediction; and output the optimal switching vector calculated in the previous calculation, and predict the inductor current, output voltage, and output current of the inverter at the next moment according to the sampling signal and the switching state;

[0011] 2) Multiply the harmonic signal u rehabc calculated in step 1) by the coefficient x, add the obtained result to the reference voltage calculated by the grid-forming inverter control, obtain a new reference voltage, and use model predictive control to track the new reference voltage; perform model predictive control calculation using the predicted inductor current, output voltage, and output current of the inverter at the next moment, select the switching vector that makes the output voltage closest to the reference voltage, and save it for output during the next sampling calculation;

[0012] Flexibly suppress the grid-connected current harmonics and PCC voltage harmonics by adjusting the voltage difference on the grid impedance.

[0013] Further, the implementation process of step 1) is specifically as follows:

[0014] 1.1) Use the orthogonality property of the sine function to process the inverter output voltage sampling signal u oabc within a period, and calculate the inverter output voltage harmonic signal u ohabc ;

[0015] 1.2) Use the orthogonality property of the sine function to process the grid-connected current sampling signal i gabc within a period, and calculate the grid-connected current harmonic signal i ghabc ;

[0016] 1.3) Multiply the result i ghabc obtained in step 1.2) by the virtual impedance L s to obtain the virtual inductor voltage harmonic signal u vhabc ;

[0017] 1.4) The inverter output voltage harmonic signal uohabc Subtract the result u obtained in step 1.3) vhabc to obtain the predicted harmonic voltage signal u of the right side of the virtual inductor rehabc .

[0018] Furthermore, the specific steps of step 1.1) to step 1.2) are as follows:

[0019] S11 calculates the sine component amplitude and cosine component amplitude of each harmonic component of the output voltage u of phase A of the inverter according to the orthogonality of trigonometric functions, and includes the voltage signals at times k + 1 and k + 2 in the integral for calculation, obtaining: oa

[0020]

[0021]

[0021] S12 further obtains the expression of W in the discrete calculation of model predictive control s1 as:

[0022]

[0023] where u oa (j) is the instantaneous value of the output voltage of phase A of the inverter at time j, and sin(θ(j)) is the instantaneous phase angle of the output voltage at time j;

[0024] S13 calculates the sine component amplitude W sn of the nth harmonic component of the output voltage of phase A of the inverter and the cosine component amplitude W cn of the nth harmonic component of the output voltage of phase A:

[0025]

[0026] S14 Substitute the inverter output voltage sampling signal u oabc , the grid-connected current sampling signal i gabc into equations (8) and (9) to calculate the inverter output voltage harmonic signal u ohabc and the grid-connected current harmonic signal i ghabc .

[0027] Furthermore, the step 1.3) is:

[0028] Use the relationship between inductor current and voltage to obtain the harmonic voltage across the virtual inductor

[0029]

[0030] Furthermore, the step 1.4) is:

[0031] Obtain the harmonic component of the node voltage on the right side of the virtual inductor according to the KVL equation:

[0032] urehabc =u ohabc -u vhabc (11)

[0033] The grid-connected converter system using the present invention has the following characteristics:

[0034] 1) Model predictive control is used instead of PWM modulation module, and the control system has faster response speed and greater anti-interference ability;

[0035] 2) The adopted harmonic extraction scheme has smaller phase error;

[0036] 3) The compensation control structure is simple and easy to implement;

[0037] 4) Flexible suppression of PCC voltage harmonics and grid current harmonics can be achieved by adjusting the coefficient x. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a main circuit structure diagram of a grid-connected inverter used in the grid-connected inverter power quality optimization control method of the present application, wherein: gabc is the grid voltage; u PCCabc is the grid voltage at the common coupling point; i gabc is the grid current; u oabc is the inverter output voltage; i oabc is the inverter output current; i Labc The LC filter grid-connected inverter consists of a bridge inverter topology, an inverter side inductor L f 、Filter capacitor C f and passive damping R d composition.

[0039] Figure 2 This is the control block diagram of the delay compensation scheme adopted in this application, which includes the first step of sampling and updating the drive signal, using the sampling and switching signals to predict the k+1 switching cycle voltage and current, and finally selecting the optimal switching state so that the output voltage at k+2 is consistent with the reference voltage.

[0040] Figure 3 When the reference output power of the grid-forming inverter is 10kW, the grid impedance is 5mH, there is 3% fifth-order negative-sequence harmonic voltage, 3% seventh-order positive-sequence harmonic voltage, and the nonlinear load power is 1.2kW, the PCC voltage u of the grid-forming inverter without harmonic suppression control is PCC and grid current i g Steady-state test waveform;

[0041] Figure 4When the reference output power of the grid-forming inverter is 10 kW, the grid impedance is 5 mH, there is 3% fifth-order negative-sequence harmonic voltage and 3% seventh-order positive-sequence harmonic voltage in the grid, and the power of the nonlinear load is 1.2 kW, the PCC voltage u of the grid-forming converter when harmonic suppression control is added and x takes 0.3 PCC and the grid current i g Steady-state test waveforms;

[0042] Figure 5 When the reference output power of the grid-forming inverter is 10 kW, the grid impedance is 5 mH, there is 3% fifth-order negative-sequence harmonic voltage and 3% seventh-order positive-sequence harmonic voltage in the grid, and the power of the nonlinear load is 1.2 kW, the PCC voltage u of the grid-forming converter when harmonic suppression control is added and x takes 0.5 PCC and the grid current i g Steady-state test waveforms;

[0043] Figure 6 When the reference output power of the grid-forming inverter is 10 kW, the grid impedance is 5 mH, there is 3% fifth-order negative-sequence harmonic voltage and 3% seventh-order positive-sequence harmonic voltage in the grid, and the power of the nonlinear load is 1.2 kW, the PCC voltage u of the grid-forming converter when harmonic suppression control is added and x takes 0.7 PCC and the grid current i g Steady-state test waveforms;

[0044] Figure 7 When the reference output power of the grid-forming inverter is 10 kW, the grid impedance is 5 mH, there is 3% fifth-order negative-sequence harmonic voltage and 3% seventh-order positive-sequence harmonic voltage in the grid, and the power of the nonlinear load is 1.2 kW, the PCC voltage u of the grid-forming converter when harmonic suppression control is added and x takes 0.95 PCC and the grid current i g Steady-state test waveforms. Specific implementation manners

[0045] The technical solutions of the present invention will be described in detail below with reference to the accompanying drawings.

[0046] Figure 1 This is the main circuit structure diagram of the grid-connected inverter applied to the power quality optimization control method of the grid-forming grid-connected inverter in this application. This application uses model predictive control to replace the PWM modulation module of the grid-forming grid-connected inverter to achieve a higher response speed. The sampled inverter output voltage sampling signal u oabc and the grid-connected current sampling signal i gabc are obtained. The voltage harmonic signal u is calculated by using model prediction refhabcMultiply the predicted voltage harmonic signal by the coefficient x and add it to the reference voltage of the grid-forming inverter to obtain a new reference voltage. Perform model predictive control calculation on the reference voltage to generate the bridge arm drive signal, update the inverter output voltage, reduce the voltage difference on the grid impedance, and control the grid impedance harmonic voltage by changing the magnitude of the coefficient x, thereby achieving flexible suppression of the PCC voltage quality and the grid-connected current quality.

[0047] Embodiment 1:

[0048] This embodiment is an optimized control system for the power quality of the output of a grid-forming inverter. The control system includes a grid-forming inverter, and the grid-forming inverter includes a bridge inverter topology and an inverter-side inductor L f , a filter capacitor C f , and is connected to the power grid after connecting to a local load. As Figure 1 shown, u gabc is the grid voltage; u PCCabc is the grid voltage at the point of common coupling; i gabc is the grid-connected current; u oabc is the inverter output voltage; i oabc is the inverter output current; i Labc is the current of the inverter filter inductor. The LC-filtered grid-connected inverter consists of a bridge inverter topology, an inverter-side inductor L f , a filter capacitor C f and a passive damper R d . The inverter power loop control is a grid-forming virtual synchronous machine control, which combines model predictive control to calculate the reference harmonic voltage and obtain the drive signal of the inverter switching tube.

[0049] Embodiment 2:

[0050] The power quality optimization control method for the grid-forming grid-connected inverter of this application includes the following steps:

[0051] 1) Use model predictive control to replace the traditional PWM modulation module. The specific implementation method is as follows: Equation (1) is the expression of the state equation of the inverter. The zero-order hold is used to transform the state equation of the circuit into a discrete-domain equation (2)

[0052]

[0053] where T s is the sampling time i Labc (k + 1) is the predicted inductor current, u oabc (k + 1) is the predicted inverter output voltage, i Labc (k) is the sampled inductor current, u oabc (k) is the inverter output voltage, i oabc (k) is the output current sampling, uiabc (k) is the output voltage vector of the DC-AC converter.

[0054] The two-level inverter has eight switching vector states. Substituting the expression of each switching vector state into (2) can calculate the output voltage under this switching vector state. The goal of the model predictive control adopted in this application is to achieve stable control of the output voltage of the two-level inverter. Therefore, it is necessary to select the switching vector that makes the output voltage closest to the reference voltage. The square value of the difference between the reference voltage and the alternative output voltage represents the straight-line distance between the reference voltage and the alternative output voltage. Let the cost function be the square value of the difference between the reference voltage and the alternative output voltage. Calculate the cost functions corresponding to the eight switching vectors. Select the one with the minimum cost function, then this alternative voltage is closest to the reference voltage, and the switching vector corresponding to this alternative voltage should be output. The cost function is expressed as:

[0055] J = (u aref - u a (k + 1)) 2 + (u bref - u b (k + 1)) 2 + (u cref - u c (k + 1)) 2 (3)

[0056] where u aref , u bref , u cref are the capacitor reference voltages, and u a (k + 1), u b (k + 1), u c (k + 1) are the predicted output voltages of the inverter.

[0057] Ideally, the predictive control calculation at time k can be completed in an extremely short time, that is, sampling and triggering of the switching device are carried out at the same moment. However, in actual situations, there are influencing factors such as measurement delay, calculation delay, and switching action delay. Therefore, there is a certain delay in the control process, which affects the actual control effect. Therefore, this paper adopts two-step prediction to compensate for the delay, that is, the system outputs the optimal switching vector calculated at the previous moment at time k, and then samples. After predicting the voltage and current at time k + 1, further predict the optimal switching vector solution calculated at time k + 2 and retain it for output at the next sampling. The control block diagram is as Figure 2 shown.

[0058] 2) Adopt the method of recording with a sliding window to record the inverter output voltage sampling signal u oabc and the grid-connected current sampling signal i gabc within a fundamental wave period, and calculate the corresponding harmonic components by using the orthogonality of the sine function.

[0059] According to the Fourier series theorem, it can be obtained that both the output voltage and current of the inverter can be Fourier-expanded into the superposition of countless sine waves. Taking the output voltage u of phase A of the inverter as an example, the Fourier decomposition of the sampling signal can be obtained as follows: oa For example, the Fourier decomposition of the sampling signal can be obtained as follows:

[0060]

[0061] where W s1 is the amplitude of the sine component of the fundamental wave component of the output voltage of phase A, W c1 is the amplitude of the cosine component of the fundamental wave component of the output voltage of phase A, W sn is the amplitude of the sine component of the nth harmonic component of the output voltage of phase A, and W cn is the amplitude of the cosine component of the nth harmonic component of the output voltage of phase A.

[0062] Therefore, according to the orthogonality of trigonometric functions, the amplitudes of the sine and cosine components of each harmonic component of the output voltage u of phase A of the inverter can be calculated. Taking the amplitude W of the sine component of the fundamental wave of u as an example oa For example, taking the amplitude W of the sine component of the fundamental wave of u as an example oa the amplitude W of the sine component of the fundamental wave s1 of u

[0063]

[0064] In digital control, both the sampling and calculation of the inverter are discrete. Therefore, it is necessary to convert Equation (5) into a discrete integral form to realize its application in digital control. The model predictive control delay compensation scheme adopted in this paper is two-step predictive control, that is, the switching vector is sampled and output at time k. The voltage and current signals of the inverter at time k + 1 are predicted using the current sampling signal and the switching vector. Then, the eight possible switching vectors are put into the cost function for rolling optimization to select the best switching vector closest to the reference voltage at time k + 2, and it is output at the beginning of the next sampling period. To reduce the influence introduced by the phase error, the voltage signals at times k + 1 and k + 2 are included in the integral for calculation. Combining the above formula, we get:

[0065]

[0066] Furthermore, the expression of W in the discrete calculation of model predictive control can be obtained as s1 follows:

[0067]

[0068] where u oa (j) is the instantaneous value of the output voltage of phase A at time j, and sin(θ(j)) is the instantaneous phase angle of the output voltage at time j.

[0069] Similarly, the amplitude W of the sine component of the nth harmonic component of the output voltage of phase A in the discrete calculation of model predictive control is given below sn and the amplitude W of the cosine component of the nth harmonic component of the output voltage of phase A cn .

[0070]

[0071] Substitute the sampling signals u oabc and i gabc into equations (8) and (9) respectively, and the harmonic signals u ohabc of the inverter output voltage and i ghabc of the grid-connected current can be calculated

[0072] 3) The harmonic voltage across the virtual inductor can be obtained using the relationship between the inductor current and voltage

[0073]

[0074] Furthermore, the harmonic component of the node voltage on the right side of the virtual inductor can be obtained according to the KVL equation

[0075] u rehabc = u ohabc - u vhabc (11)

[0076] The harmonic voltage u rehabc component on the right side of the virtual inductor is added as a feed-forward compensation to the reference voltage of the grid-forming inverter. By controlling the harmonic voltage component of the PCC voltage to be equal to u rehabc , the harmonic voltage difference of the grid line impedance can be reduced, and cancellation compensation can be achieved, thereby reducing the harmonic content of the grid-connected current

[0077] 4) Multiply the calculated harmonic reference voltage u rehabc by the coefficient x and add it to the reference voltage of the grid-forming converter. Adjusting the coefficient x can achieve flexible suppression of the PCC voltage harmonics and the grid-connected current harmonics Figures 3 to 7 When the reference output power of the grid-forming inverter is 10 kW, the grid impedance is 5 mH, there is a 3% fifth-order negative-sequence harmonic voltage and a 3% seventh-order positive-sequence harmonic voltage in the grid, and the power of the nonlinear load is 1.2 kW, the steady-state test waveforms of the PCC voltage u PCC and the grid current i g of the grid-forming converter with harmonic suppression control are shown. As the coefficient x increases, the harmonic content of the grid-connected current decreases, while the harmonic content of the PCC voltage increases. By adjusting the coefficient x, flexible suppression of the PCC voltage harmonics and the grid-connected current harmonics can be achieved

[0078] The above embodiments are only used to illustrate the technical idea of the present invention, and the protection scope of the present invention cannot be limited thereby. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the present invention.

Claims

1. A grid-type inverter output power quality optimization control system, characterized in that: The control system includes a grid-type inverter, which includes a bridge inverter topology, an inverter side inductor L f 、Filter capacitor C f , connected to the local load and then connected to the grid.

2. The control method of the grid-type inverter output power quality optimization control system according to claim 1, characterized in that: The control method comprises the following steps: 1) Sample the inverter output voltage sampling signal u oabc , grid current sampling signal i gabc , combined with the model prediction calculation, the grid voltage harmonic signal u rehabc ; And output the optimal switching vector calculated last time, and predict the inverter inductor current, output voltage and output current at the next moment according to the sampling signal and the switching state; 2) The harmonic signal u calculated in step 1) is rehabc The result is multiplied by the coefficient x, and the result is added to the reference voltage calculated by the grid-type inverter control to obtain a new reference voltage, and the new reference voltage is tracked by the model predictive control; the model predictive control calculation is performed using the predicted inverter inductor current, output voltage and output current at the next moment, and the switch vector that makes the output voltage closest to the reference voltage is selected, and the switch vector is saved and outputted when the next sampling calculation is performed; By adjusting the voltage difference on the grid impedance, flexible suppression of grid current harmonics and PCC voltage harmonics can be achieved.

3. The control method according to claim 2, characterized in that: The implementation process of step 1) is specifically as follows: 1.1) Using the orthogonal property of the sine function to sample the inverter output voltage signal u within the cycle oabc Processing is performed to calculate the inverter output voltage harmonic signal u ohabc ; 1.2) Using the orthogonal property of the sine function to sample the grid current within the period gabc Processing is performed to calculate the grid current harmonic signal i ghabc ; 1.3) Result i obtained in step 1.2) ghabc Multiply by the virtual impedance L s Get the virtual inductor voltage harmonic signal u vhabc ; 1.4) The inverter output voltage harmonic signal u in step 1.1) ohabc Subtract the result u obtained in step 1.3) vhabc Get the predicted voltage harmonic signal u on the right side of the virtual inductor rehabc .

4. The control method according to claim 2, characterized in that: The steps 1.1) to 1.2) are specifically as follows: S11 calculates the inverter A phase output voltage u according to the orthogonality of trigonometric functions oa The sine component amplitude and cosine component amplitude of each harmonic component are calculated by taking the voltage signals at time k+1 and k+2 into account: S12 further obtains the discrete calculation of model predictive control s1 The expression is: where u oa (j) is the instantaneous value of the inverter A phase output voltage at time j, sin(θ(j)) is the instantaneous phase angle of the output voltage at time j; S13 calculates the amplitude W of the sinusoidal component of the nth harmonic component of the inverter A phase output voltage sn And the cosine component amplitude W of the nth harmonic component of the A phase output voltage cn : S14 Inverter output voltage sampling signal u oabc , grid current sampling signal i gabc Substituting into equations (8) and (9) to calculate the inverter output voltage harmonic signal u ohabc And the grid current harmonic signal i ghabc .

5. The control method according to claim 2, characterized in that: The step 1.3) is: The harmonic voltage across the virtual inductor is obtained using the relationship between the inductor current and voltage.

6. The control method according to claim 2, characterized in that: The step 1.4) is: According to the KVL equation, the harmonic component of the node voltage on the right side of the virtual inductor is obtained: in rehabc =in ohabc -in vhabc (11).

Images