Energy storage type inverter networking control method considering low-voltage fault

Abstract

The invention discloses an energy storage type inverter networking control method considering a low-voltage fault, and belongs to the technical field of power generation, power transformation or power distribution. The method comprises a peak current prediction hysteresis control strategy, a reactive power droop-current PI control parallel minimum control law strategy and a virtual synchronous machine control strategy. The peak current prediction hysteresis control strategy calculates the current change trend of the next switching period in real time, and outputs a current peak value under the condition of accurately limiting voltage flicker in a single period. The reactive power droop control and the current PI control are used for taking a minimum output mode in parallel, and reactive power regulation and quick response of reactive current during low-voltage faults are maintained. Wherein the output saturation lower limit of the PI regulator is set by a minimum internal potential limiting mechanism, and sinusoidal output of reactive current under the short-circuit fault of the power grid is ensured. The method disclosed by the invention effectively gives consideration to low-voltage ride-through control, has the advantages of good rapidity and high stability, and is suitable for the energy storage type network-building inverter with high-reliability operation.

Description

Technical Field

[0001] The present invention relates to microgrid control technology, and discloses a grid-forming control method for an energy storage inverter that takes into account low grid voltage faults, belonging to the technical field of power generation, power transformation or power distribution. Background Art

[0002] In recent years, with the rapid development of renewable energy power generation technology and distributed energy systems, energy storage grid-forming inverters, as key interface devices, have been widely used in new power system scenarios such as photovoltaic power stations, wind farms, and microgrids. This type of inverter not only needs to complete the function of DC-AC power conversion, but also undertakes important system-level control tasks such as maintaining grid voltage / frequency stability and providing inertia support, and is the core device to ensure the reliable grid connection of new energy. However, in a complex grid environment, the system frequently faces fault conditions such as voltage sags and asymmetric short circuits. These abnormal situations will cause transient shocks and steady-state distortions of the inverter output current, which may not only damage power devices, but also endanger the safe and stable operation of the grid.

[0003] To improve the stability of new energy power generation systems connected to new power systems, when the grid experiences a short-term fault resulting in a voltage drop, the energy storage inverter must be able to maintain grid-connected operation within a specified time. The time-voltage characteristic curve of low voltage ride-through is as Figure 1 shown.

[0004] Through experimental tests, it can be observed that in the extreme condition where the load suddenly changes from no-load to full short circuit, the short-circuit fault ride-through characteristics of a high-power energy storage grid-forming inverter are as Figure 2 shown, and the partial amplification of the short-circuit current waveform is as Figure 3 shown. After the load is short-circuited, the fault current rapidly climbs, and its effective value reaches about 500A. The rapidly increasing short-circuit current is extremely likely to trigger the over-current protection mechanism of the inverter, resulting in low voltage ride-through failure. To ensure the stable operation of the energy storage grid-forming inverter during grid faults, effective short-circuit current limiting measures must be taken. According to the requirements of grid specifications, during low voltage ride-through, the output current of the inverter should be strictly controlled within the range of 1.2 to 1.5 times the rated current. This limitation requirement takes into account both the safe operation boundary of the equipment and the stability requirements of the grid.

[0005] The current fault current limiting control of energy storage grid-forming inverters mainly adopts the following six types of technical solutions:

[0006] The first type is the per-pulse control method: by real-time monitoring the current and setting a hardware protection threshold, the PWM signal is immediately turned off when over-current occurs. Although this method can quickly limit the current amplitude, it belongs to an open-loop protection mechanism and cannot achieve sinusoidal control of the output current. Moreover, frequent hard turn-off will cause voltage spikes and device stress problems;

[0007] The second type is the adaptive hysteresis control method: by dynamically adjusting the hysteresis width to optimize the switching frequency, but there is a delay in calculating the loop width, typically a delay of 2 - 3 switching cycles, which may lead to current out - of - control under grid mutation conditions;

[0008] The third type is the traditional PI control method: it has good current tracking performance under steady - state conditions, but its bandwidth is limited. When facing transient processes such as voltage sags, there is overshoot and a risk of over - current;

[0009] The fourth type is the power - limiting method: it adopts a three - loop closed - loop control architecture, but there are problems such as insufficient system bandwidth and slow dynamic response, and it cannot provide effective over - current protection during short - circuit faults;

[0010] The fifth type is the virtual impedance method: by generating impedance characteristics through algorithms to limit current, there are a fault detection window and time delay, resulting in a lag in impedance adjustment, which may cause current oscillation during deep voltage dips;

[0011] The sixth type is the non - linear control method: such as sliding - mode control, although it has the advantage of strong robustness, there is a problem of high - frequency chattering. The harmonics near the switching frequency increase by 5 - 10%, resulting in the deterioration of current THD and making it difficult to meet the grid - connected power quality standards.

[0012] In summary, the traditional fault current - limiting control technology of energy - storage inverters has slow dynamic response, insufficient transient over - current suppression, and distorted steady - state waveforms, and cannot meet the requirements for limiting the output current during low - voltage ride - through. The present invention aims to propose a grid - forming control method for energy - storage inverters considering low - voltage faults to overcome the above - mentioned defects. Summary of the Invention

[0013] The object of the present invention is to address the deficiencies in the above - mentioned background technology and provide a grid - forming control method for energy - storage inverters considering low - voltage faults. Based on a control architecture of reactive - power droop and current PI parallel minimum control law, introducing a peak - current prediction hysteresis control mechanism and minimum internal - potential limit, it can not only handle reactive - power regulation under normal conditions, but also focus on solving the technical problems of slow dynamic response, large transient over - current impact, and serious distortion of steady - state current waveforms of grid - forming inverters under grid - voltage flicker conditions, achieving the object of reliable grid - forming of energy - storage inverters under grid - voltage flicker conditions.

[0014] To achieve the above - mentioned object of the invention, the present invention adopts the following technical solutions:

[0015] A grid - forming control method for energy - storage inverters considering low - voltage faults, including: a peak - current prediction hysteresis control strategy, a minimum control law strategy of reactive - power droop - current PI control in parallel, and a virtual - synchronous - machine control strategy;

[0016] Peak current prediction hysteresis control strategy, which predicts the peak values of the three-phase inverter currents based on the inverter output voltage, the three-phase grid-connected currents, and the three-phase grid-connected voltages, calculates the duty cycle correction amount according to the difference between the maximum value among the predicted values of the three-phase inverter currents and the set maximum current value, and adjusts the duty cycle of the inverter output voltage according to the duty cycle correction amount;

[0017] Reactive power droop - current PI control parallel minimum control law strategy, which performs droop control on the reactive power injected by the inverter into the grid, accumulates the preset internal potential on the basis of the droop control result to obtain the first voltage reference value, performs minimum internal potential limitation on the output of the current PI regulator of the inverter, accumulates the preset internal potential on the basis of the minimum internal potential limitation result to obtain the second voltage reference value, and takes the minimum value of the first voltage reference value and the second voltage reference value as the internal potential reference;

[0018] Virtual synchronous machine control strategy, which calculates the coordinate rotation angle through the virtual synchronous machine swing equation, and the control command of the inverter is obtained according to the inverter output voltage duty cycle, the internal potential reference, and the coordinate rotation angle.

[0019] As a further optimization scheme of a grid-forming control method for an energy storage inverter considering low-voltage faults, in the peak current prediction hysteresis control strategy, the peak values of the three-phase inverter currents are predicted based on the inverter output voltage, the three-phase grid-connected currents, and the three-phase grid-connected voltages, specifically: Where, i n (k), i n (k + 1) are the predicted peak values of the n-phase current of the inverter at the kth moment and the (k + 1)th moment, n = a, b, c, T s is the sampling period, L is the inductor on the inverter side, v(k) is the inverter output voltage at the kth moment, and u(k) is the grid-connected voltage at the kth moment.

[0020] As a further optimization scheme of a grid-forming control method for an energy storage inverter considering low-voltage faults, in the peak current prediction hysteresis control strategy, the duty cycle correction amount is calculated according to the difference between the maximum value among the predicted values of the three-phase inverter currents and the set current maximum value, specifically: Where, d′ is the duty cycle correction amount, i max (k + 1) is the maximum value among the predicted values of the three-phase inverter currents at the (k + 1)th moment, I max_ref is the set current maximum value, and I max_ref is 1.2 times the rated current.

[0021] As a further optimization scheme of the grid-forming control method for an energy storage inverter considering low-voltage faults, in the peak current prediction hysteresis control strategy, the duty cycle of the inverter output voltage 1-d is adjusted according to the duty cycle correction amount, where d = max(0, d′), and the value range of d is [0, 1].

[0022] As a further optimization scheme of the grid-forming control method for an energy storage inverter considering low-voltage faults, in the minimum control law strategy with reactive power droop-current PI control in parallel, droop control is performed on the reactive power injected by the inverter into the grid. Specifically: ΔU1 = K q (Q * -Q), ΔU1 is the result of droop control, K q is the reactive power droop coefficient, Q * is the desired reactive power, and Q is the reactive power injected by the inverter into the grid.

[0023] As a further optimization scheme of the grid-forming control method for an energy storage inverter considering low-voltage faults, in the minimum control law strategy with reactive power droop-current PI control in parallel, in the current PI regulator of the inverter, the calculation method of the current reference value is: where, I ref is the current reference value, I qref is the reference reactive current injected by the inverter into the grid, and I d is the active current injected by the inverter into the grid.

[0024] As a further optimization scheme of the grid-forming control method for an energy storage inverter considering low-voltage faults, in the minimum control law strategy with reactive power droop-current PI control in parallel, minimum internal potential limitation is performed on the output of the current PI regulator of the inverter. Specifically: According to the current PI control saturation lower limit required for the sinusoidalization of the inverter output current under the grid short-circuit condition, the k value in U * lowlimit =-U0 + kω0LI SC is adjusted, where U * lowlimit is the current PI regulation saturation lower limit, U0 is the preset internal potential, ω0 is the grid rated angular frequency, and I SC is the amplitude of the short-circuit current allowed by the inverter.

[0025] As a further optimization scheme of the grid-forming control method for an energy storage inverter considering low-voltage faults, in the virtual synchronous machine control strategy, the swing equation of the virtual synchronous motor calculates the coordinate system rotation angle by constructing a virtual inertia link that superimposes the grid rated angular frequency on the frequency deviation, where, θ e is the dq rotating coordinate system angle, DP For simulating the damping coefficient of a synchronous generator, J is the rotor inertia of the simulated synchronous generator, P is the active power injected by the inverter into the grid, and P m is the virtual mechanical power.

[0026] An electronic device includes a memory and a processor. A computer program is stored on the memory and runs on the processor. When the processor runs the computer program, it executes the steps of the grid-forming control method for the energy storage type inverter described above.

[0027] A computer-readable storage medium stores a computer program. When the computer program runs, it executes the steps of the grid-forming control method for the energy storage type inverter described above.

[0028] The present invention adopts the above technical solutions and has the following beneficial effects:

[0029] (1) The present invention abandons the traditional voltage-current double closed-loop control architecture and the power-voltage-current triple closed-loop control architecture, and proposes a control mode of reactive power droop and current PI parallel minimum control law. The internal potential reference is obtained through power-current parallel calculation and minimum internal potential limitation, and a high-bandwidth peak current prediction hysteresis control mechanism is introduced to provide a response time window for current PI control, quickly and autonomously realizing the sinusoidal control of the reactive power support current under the condition of grid voltage dip, realizing high-reliability grid formation, and providing a set of grid-forming control solutions that can effectively cope with grid voltage flicker and balance transient response and steady-state performance, meeting the strict grid code requirements on the premise of ensuring the safe operation of the equipment.

[0030] (2) The present invention adopts a parallel architecture of current PI control and reactive power droop control in cooperation with the minimum control law to achieve seamless and rapid switching between normal conditions and fault conditions. Under normal conditions, reactive power droop control is automatically adopted to maintain the grid voltage regulation function. During the fault period, it automatically switches to current PI control to output the reactive current required for low voltage ride-through. At the same time, by setting the saturation lower limit of the PI controller, it is ensured that even in the most severe condition, that is, when the grid is short-circuited, a high-quality sinusoidal reactive power waveform that meets the grid-forming requirements is still output.

[0031] (3) Based on the peak current prediction hysteresis technology, the present invention can complete the dynamic adjustment of the duty cycle in an extremely short time, effectively suppressing the peak value of the output current under the condition of voltage flicker, and providing a response time window for current PI control.

[0032] (4) The present invention calculates the coordinate rotation angle in real time through the swing equation in the virtual synchronous machine control strategy, providing data support for accurately obtaining the inverter control command. Description of the Drawings

[0033] Figure 1 is a time-voltage characteristic curve diagram of the low voltage ride-through of the energy storage type inverter.

[0034] Figure 2 It is the waveform diagram of the short - circuit fault test of a high - power energy - storage inverter.

[0035] Figure 3 is Figure 2 The enlarged diagram of the short - circuit current waveform shown.

[0036] Figure 4 It is the block diagram of the grid - forming control method of the energy - storage inverter considering low - voltage faults proposed by the present invention.

[0037] Figure 5 It is the flow chart of the grid - forming control method of the energy - storage inverter considering low - voltage faults proposed by the present invention.

[0038] Figure 6 It is the waveform diagram of the rated current of a low - power energy - storage inverter provided in an embodiment of the present invention.

[0039] Figure 7 It is the waveform diagram of the low - voltage ride - through process of a low - power energy - storage inverter provided in an embodiment of the present invention.

[0040] Explanation of the reference numerals in the figure: 1. Peak - current - prediction hysteresis control strategy, 2. Minimum - control - law strategy of reactive - power droop - current PI parallel, 3. Virtual - synchronous - machine control strategy. Specific embodiments

[0041] The technical solution of the invention will be described in detail below with reference to the accompanying drawings.

[0042] Figure 4 It is the block diagram of the grid - forming control method proposed by the present invention, mainly including: peak - current - prediction hysteresis control strategy 1, minimum - control - law strategy 2 of reactive - power droop - current PI parallel, and virtual - synchronous - machine control strategy 3. The peak - current - prediction hysteresis control strategy is characterized by an extremely high bandwidth, so it will not interfere with other power and current closed - loop control systems with limited bandwidths. In the initial stage of grid - voltage flicker, the peak - current - prediction hysteresis control strategy strictly limits the peak value of the transient reactive current within the set threshold value through real - time current - trajectory prediction and duty - cycle dynamic adjustment. The typical value of the set threshold is 1.5 times the rated current; in the stage of continuous voltage drop at the grid - connection point, the minimum internal - potential limit module sets the saturation lower limit of the current PI controller to ensure sinusoidal current output under the most severe condition, i.e., grid short - circuit, and meets the requirements of high - quality reactive - power grid connection.

[0043] This experiment was verified on a low - power energy - storage grid - forming inverter platform with a DC - bus voltage of 24V. The specific implementation methods of each strategy will be elaborated below.

[0044] The peak current prediction hysteresis control strategy starts autonomously at the moment when grid voltage flicker is detected. Its extremely high bandwidth design enables this control strategy not to interfere with the power and current closed-loop control systems with limited bandwidth. This strategy predicts the current change trend in the next switching cycle in real time, compares it with the preset current threshold, and dynamically calculates the required correction amount of the output voltage duty cycle, thereby actively adjusting the amplitude of the inverter output voltage before the current exceeds the limit, effectively suppressing the generation of transient impact current. The high bandwidth design provides a sufficient time buffer window for the subsequent minimum control law strategy in parallel with the reactive power droop-current PI control with limited bandwidth, ensuring that during voltage flicker, the system can quickly suppress transient overcurrent, enabling the minimum control law strategy in parallel with the reactive power droop-current PI control with limited bandwidth to maintain a steady-state sinusoidal output current.

[0045] The core idea of the peak current prediction hysteresis control strategy is that when the predicted value of the phase current peak of the energy storage inverter exceeds the set maximum current value, the output voltage duty cycle is dynamically corrected to reduce the amplitude of the output voltage of the energy storage inverter; and whether to reduce the amplitude of the inverter output voltage is determined according to whether the predicted value of the phase current peak in the next switching cycle exceeds the limit.

[0046] Specifically, a mathematical model of the energy storage inverter is established according to Kirchhoff's voltage law, and the grid-connected phase voltage is expressed by the formula:

[0047]

[0048] The forward Euler discretization method is used to perform discrete prediction calculations on the state equations of the three-phase currents of the inverter.

[0049]

[0050] Among them, u n is the grid-connected phase n voltage; v n is the inverter phase n output voltage; i n is the inverter phase n current; L is the inductor on the inverter side; i n (k), i n (k + 1) are the predicted peaks of the inverter phase n current at time k and time k + 1; T s is the sampling period, which is 6.67e-5 s in this experiment; v(k) is the inverter output voltage at time k, and if the influence of delay and high-frequency harmonics is ignored, it can be replaced by the modulation voltage U; u(k) is the grid-connected voltage at time k obtained by sampling.

[0051] Then, the difference is taken between the maximum value of the absolute value of the predicted peaks of the three-phase currents of the inverter and the set maximum current I max_ref to obtain the duty cycle correction amount d′, which is expressed by the formula:

[0052]

[0053] During voltage flicker, considering the safe operating conditions of the droop control type inverter, according to the given requirements of the low voltage ride-through inverter current, the maximum current I is set max_ref Take 1.2 times the rated current value, that is, I max_ref = 1.2I N = 12.0 A.

[0054] The output voltage of the inverter can be expressed as:

[0055] d = max(0, d')

[0056] υ n = υ n (1 - d)n = a, b, c

[0057] 1 - d is the duty cycle of the inverter output voltage. When |i max (k + 1)| is greater than I max_ref but less than 2I max_ref the calculated value range of d' is between 0 - 1, the duty cycle is 1 - d, and the amplitude of the inverter output voltage is dynamically adjusted; when |i max (k + 1)| is greater than 2I max_ref d' takes 1, the duty cycle is 0, and at this time, it is considered that the overcurrent problem is relatively serious, and the driving signal of the bridge arm is blocked; when |i max (k + 1)| is less than I max_ref the calculated d' is less than 0, the duty cycle takes 1, and the inverter outputs the normal voltage. In this way, while protecting the safety of the power device, a controllable voltage regulation is maintained, and the key regulation time is reserved for the current PI control.

[0058] The minimum control law strategy with reactive power droop - current PI control in parallel, the current PI control and the reactive power droop loop operate in parallel, and the minimum control law is taken to achieve the autonomous and fast switching between the internal potential control of the virtual synchronous machine and the current limiting mode. Specifically, the output U o1 * of the reactive power loop is obtained by superimposing the calculation result ΔU1 of the reactive power droop algorithm and the preset internal potential U o ; the output U o2 * of the current PI control is obtained by superimposing the calculation result ΔU2 of the current PI controller and the preset internal potential U o ; then the output U o1 * of the reactive power loop and the output U o2 * of the current PI controller take the min() function.

[0059] Under normal operating conditions, the output of the current PI controller automatically saturates to the upper limit, and the system adopts reactive power droop control. Under the action of the min() function, the output of the reactive power loop is used as the final internal potential reference U*, maintaining the grid voltage regulation function; the energy storage inverter needs to provide high-quality sinusoidal reactive current support during grid low-voltage faults; during grid low-voltage faults, the system automatically enters the constant current control mode. At this time, the current PI controller exits the positive saturation state because the detected actual current exceeds the set value, and the current PI controller is negatively saturated. The output after the negative saturation of the current PI controller is superimposed with the preset internal potential U o After that, the output voltage reference value U of the current loop is significantly reduced o2 *. Under the action of the min() function, the output of the current loop is used as the final internal potential reference U*, and the system automatically switches to the constant current control mode to ensure that the output sinusoidal reactive current meets the grid-forming requirements.

[0060] Regardless of the operating conditions, the output of the reactive power loop and the output of the current PI loop always pass through the min() function operation to obtain the minimum control law, denoted as U * , and this minimum control law is converted into a symmetric three-phase voltage U a ,U b ,U c , and receives the pulse width modulation ratio d to control the actual internal potential U of the virtual synchronous machine a (1 - d),U b (1 - d),U c (1 - d), where the required dq rotating coordinate system angle θ e is obtained by calculating the swing equation in the virtual synchronous machine control strategy.

[0061] This minimum control law control mechanism realizes the steady-state current sinusoidal control under low-voltage faults by dynamically reducing the amplitude of the internal potential U of the virtual synchronous machine a (1 - d),U b (1 - d),U c (1 - d). Among them, the current reference value I of the current PI controller ref will be elaborated in detail later.

[0062] Under the condition of a sudden drop in the grid voltage, the setting of the saturation lower limit of the current PI controller is very important. The saturation lower limit of the current PI controller determines the maximum value of the sinusoidal current output by the inverter under the worst condition, that is, the grid short-circuit fault. To cope with the worst grid voltage short-circuit fault, according to the difference between the inverter output current I feedback and the current reference value I ref , the minimum internal potential limit is carried out according to the PI regulation result of the difference, and the saturation lower limit of the current PI controller is fixed as:

[0063] U *lowlimit = -U0 + kω0LI SC

[0064] where U * lowlimit is the saturation lower limit of the current PI regulator; ω0 is the rated angular frequency of the power grid, with a value of 314.159 rad / s; L is the inductor on the inverter side, with a value of 1 mH; I SC is the amplitude of the short-circuit current allowed for the inverter. In this small-power verification platform, the value is 10 A; U0 is the preset internal electromotive force, with a typical value of 311 V. In this experiment, the value is 10 V; k is the adjustment coefficient. According to the current PI control saturation lower limit required for the sinusoidal output current of the inverter under short-circuit conditions, the value of k is fine-tuned. In this experiment, when the voltage drops extremely, when the saturation lower limit of the current PI controller is -0.75*Uo, the output current of the inverter can be limited below 1.2 times the rated current, that is, k is about 0.796.

[0065] The calculation formula for the magnitude of the inverter output current is as follows:

[0066]

[0067] The calculation formulas for the active power P and the reactive power Q are as follows:

[0068] P = 1.5(u α i α + u β i β )

[0069] Q = 1.5(u α i β - u β i α )

[0070] During low-voltage ride-through, the current reference value I ref of the current PI regulator is calculated as follows:

[0071]

[0072] In this experiment, a small-power verification platform is used. The rated current of the inverter is 10 A. Substituting into the calculation, the reference value I qref of the reactive current injected by the inverter into the power grid needs to meet the following conditions:

[0073]

[0074] When the grid-side uses the control method of grid-voltage orientation, the active current I d injected by the inverter into the power grid can be obtained by coordinate transformation of the current components in the three-phase natural coordinate system.

[0075]

[0076] The reactive power droop of the present invention is similar to the excitation regulation of a synchronous generator. By simulating the voltage and reactive power relationship of a synchronous generator, the output reactive power is regulated to maintain the system voltage stability. The design formula of the reactive power loop is as follows:

[0077] ΔU1 = K q (Q * -Q)

[0078] In the formula, Q * is the desired reactive power, which is set to 0 under normal conditions; Q is the actual reactive power; K q is the reactive power droop coefficient, which is adjusted according to the actual situation.

[0079] The virtual synchronous machine control strategy 3 simulates the governor speed regulation process of a synchronous generator by constructing the swing equation of the virtual synchronous machine with a typical algorithm of a grid-forming inverter. By simulating the speed and active power droop relationship of a synchronous generator, the output active power is regulated to maintain the system frequency stability.

[0080] The primary frequency regulation formula is:

[0081] P m = P ref + K p (ω0 - ω)

[0082] In the formula, ω is the actual angular frequency of the power grid; ω0 is 314.159 rad / s; K p is the active power droop coefficient, which is taken as 100.0 in the experiment; P ref is the active power command value, which is taken as 50.0 W in the experiment; P m is the virtual mechanical power.

[0083] To simulate the rotor characteristics of a synchronous generator, the virtual inertia link in the present invention is shown as follows. Based on the output frequency deviation Δω, a reference frequency ω0 is superimposed, so that the system can automatically adjust the power when the frequency deviates. Finally, the angular frequency is integrated to obtain the dq rotating coordinate system angle θ e .

[0084] The virtual synchronous machine swing equation is designed as follows:

[0085]

[0086] In the formula, J simulates the rotor moment of inertia of a synchronous generator, which is taken as 0.5 Kg·m 2 ; D P simulates the damping coefficient of a synchronous generator, which is taken as 10.0 N·s / m; θ e is the dq rotating coordinate system angle value.

[0087] Figure 5 The flowchart of the software control algorithm used in the present invention is given. First, the instantaneous values of the three-phase currents at the sampling connection point are sampled to calculate the active power P, reactive power Q, and the peak value I of the phase current peak . At the moment of voltage flicker, the system immediately starts the peak current prediction hysteresis control. By dynamically adjusting the duty ratio of the output voltage, a key response time window is provided for the subsequent current PI control link. Under the normal operating state of the power grid, the output of the current PI controller automatically reaches the positive saturation limit value. After being superimposed with the preset internal potential reference U o , its output value is always greater than the output value of the reactive power droop loop. At this time, the output of the reactive power loop is automatically selected as the internal potential reference benchmark through the min() function to maintain the normal voltage regulation function of the system. When a power grid short-circuit fault occurs, the system enters the fault protection mode: the short-circuit current exceeds the set threshold of the PI controller, causing the output of the controller to enter the negative saturation state. After being superimposed with Uo, the output value of the current loop is significantly reduced. At this time, the output of the current loop is automatically selected as the internal potential reference benchmark through the min() function to weaken the amplitude of the internal potential of the virtual synchronous machine. While ensuring the provision of necessary reactive power support, the sinusoidal waveform characteristic of the output current is strictly maintained to achieve safe and reliable fault ride-through.

[0088] Verified by experiments, the control strategy of the present invention exhibits excellent fault ride-through performance. Figure 6 The operating waveforms of the small-power energy storage grid-forming inverter under rated conditions are shown. It can be seen that the system stably maintains the output of a rated current of 10A, and the waveform quality is excellent. Figure 7 The experimental waveforms shown reveal the dynamic response characteristics of the system under extreme power grid faults: when the simulated three-phase load suddenly changes from no-load to complete short-circuit, the peak current prediction control is immediately started, and the inrush current is smoothly limited within 1.5 times the rated value at the initial stage of the fault. Subsequently, the adjustment amount of the current PI regulator output -0.75U o is superimposed with the reference internal potential U o , and the output voltage U* is dynamically reduced to below 10V to achieve the rapid convergence of the current. The experimental results show that the system can still maintain excellent current sinusoidality after entering the steady state, and the current peak value is strictly controlled within the safe range of 1.2 times the rated value. This series of response characteristics fully prove that through the synergistic effect of the peak current prediction hysteresis control and the minimum internal potential limit in the present invention, not only the rapid suppression of the fault current is achieved, but also the steady-state operation quality is guaranteed, and the operation reliability of the new energy power generation system under power grid fault conditions is significantly improved. In addition, the peak current prediction hysteresis control technology described in the present invention has obvious advantages compared with the traditional pulse-by-pulse and hysteresis-type hard turn-off technologies, as shown in Table 1.

[0089]

[0090] Table 1

[0091] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above specific embodiments, and the above specific embodiments and the descriptions in the specification are only for further illustrating the principles and preparation effects of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope claimed by the present invention. The scope claimed by the present invention is defined by the claims and their equivalents.

Claims

1. A grid-forming control method for an energy storage inverter considering low-voltage faults, characterized in that, Including: A peak current prediction hysteresis control strategy that predicts the peak values of the three-phase inverter currents based on the inverter output voltage, the three-phase grid-connected currents, and the three-phase grid-connected voltages, calculates the duty cycle correction amount according to the difference between the maximum value among the predicted values of the three-phase inverter currents and the set maximum current value, and adjusts the duty cycle of the inverter output voltage according to the duty cycle correction amount; A minimum control law strategy with reactive power droop - current PI control in parallel that performs droop control on the reactive power injected by the inverter into the grid, accumulates a preset internal electromotive force on the basis of the droop control result to obtain a first voltage reference value, performs a minimum internal electromotive force limit on the output of the current PI regulator of the inverter, accumulates a preset internal electromotive force on the basis of the minimum internal electromotive force limit result to obtain a second voltage reference value, and takes the minimum value of the first voltage reference value and the second voltage reference value as the internal electromotive force reference; and, A virtual synchronous machine control strategy that calculates the coordinate rotation angle through the virtual synchronous machine swing equation, and the control command of the inverter is obtained according to the inverter output voltage duty cycle, the internal electromotive force reference, and the coordinate rotation angle.

2. The grid-forming control method of an energy storage type inverter considering low voltage faults according to claim 1, characterized in that In the peak current prediction hysteresis control strategy, the peak values of the three-phase currents of the inverter are predicted based on the output voltage of the inverter, the three-phase currents at the grid connection point, and the three-phase voltages at the grid connection point. Specifically: where, i n (k), i n (k + 1) are the predicted peak values of the n-phase current of the inverter at time k and time k + 1, n = a, b, c, T s is the sampling period, L is the inductor on the inverter side, v(k) is the output voltage of the inverter at time k, and u(k) is the grid connection point voltage at time k.

3. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 2, wherein, In the peak current prediction hysteresis control strategy, the duty cycle correction amount is calculated based on the difference between the maximum value of the predicted peak values of the three-phase currents of the inverter and the set maximum current value. Specifically: where d′ is the duty cycle correction amount, and i max (k + 1) is the maximum value of the predicted peak values of the three-phase currents of the inverter at the (k + 1)-th moment, I max_ref is the set maximum current value, and I max_ref is 1.2 times the rated current.

4. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 3, characterized in that, In the peak current prediction hysteresis control strategy, the duty cycle of the inverter output voltage 1 - d is adjusted according to the duty cycle correction amount, where d = max(0, d′), and the value range of d is [0, 1].

5. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 4, wherein, In the minimum control law strategy with reactive power droop-current PI control in parallel, droop control is performed on the reactive power injected by the inverter into the grid. Specifically: ΔU1 = K q (Q * - Q), where ΔU1 is the result of droop control, K q is the reactive power droop coefficient, Q * is the desired reactive power, and Q is the reactive power injected by the inverter into the grid.

6. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 5, characterized in that, In the minimum control law strategy with parallel reactive power droop-current PI control, in the current PI regulator of the inverter, the calculation method of the current reference value is as follows: Where, I ref is the current reference value, I qref is the reference value of the reactive current injected by the inverter into the power grid, and I d is the active current injected by the inverter into the power grid.

7. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 6, characterized in that, In the minimum control law strategy of reactive power droop-current PI control in parallel, the output of the current PI regulator of the inverter is limited by the minimum internal potential, specifically: according to the lower limit of the current PI control saturation required for the sinusoidal output current of the inverter under the grid short-circuit condition, for U * lowlimit =-U0 + kω0LI SC adjust the value of k in, where U * lowlimit is the lower limit of the current PI regulation saturation, U0 is the preset internal potential, ω0 is the rated angular frequency of the grid, and I SC is the amplitude of the short-circuit current allowed by the inverter.

8. The grid-forming control method of the energy storage type inverter considering low voltage faults according to claim 7, wherein, In the virtual synchronous machine control strategy, the swing equation of the virtual synchronous motor calculates the rotation angle of the coordinate system by constructing a virtual inertia link that superimposes the rated angular frequency of the power grid on the frequency deviation. where θ e is the angle of the dq rotating coordinate system, D P is the damping coefficient simulating the synchronous generator, J is the rotor inertia of the simulated synchronous generator, P is the active power injected by the inverter into the power grid, and P m is the virtual mechanical power.

9. An electronic device, including a memory and a processor, with a computer program stored on the memory and running on the processor, and when the processor runs the computer program, it executes the steps of the grid-forming control method for the energy storage type inverter according to any one of claims 1 to 8.

10. A computer-readable storage medium, with a computer program stored thereon, and when the computer program runs, it executes the steps of the grid-forming control method for the energy storage type inverter according to any one of claims 1 to 8.

Images