An active damping method based on resonant current differentiation and parameter sweep adaptation

Abstract

The present invention discloses an active damping method based on resonant current differentiation and parameter scanning adaptation. An active damping control link is introduced into the existing grid-connected converter output current control system. The active damping control link specifically includes extracting the harmonic components in the grid-connected converter output current, then passing it through a variable gain differentiation link, then through a low-pass filtering link, and finally dividing it by the virtual resistance value in the active damping control and adding it to the reference value of the grid-connected converter output current to achieve active damping control. A differential gain adaptive adjustment method is introduced into the active damping control link to adaptively adjust the gain of the differentiation link based on the grid-connected current error and the overcurrent and DC overvoltage protection signals during the device startup process. By integrating the resonant current differentiation and variable gain differentiation methods, the present invention enables active damping to address both the requirements of grid-connected converter system stability control and grid current quality in weak power grids.

Description

Technical Field

[0001] The present invention relates to the technical field of power electronics, and in particular to an active damping method based on resonant current differentiation and parameter scanning adaptation. Background Art

[0002] Grid-connected converters are increasingly being used in renewable energy grid-connected power generation, transmission, and power quality management. However, my country's renewable energy landscape is characterized by large-scale centralized access and long-distance transmission. Due to the long transmission lines, which typically require multiple boost transformers, the equivalent grid impedance is high, resulting in low short-circuit resistance and, consequently, weak grid strength.

[0003] In recent years, destabilizing oscillations caused by the interaction between grid-connected converters and weak grids have become frequent both domestically and internationally. These phenomena affect a wide range of grid-connected converter applications, including wind power, photovoltaic grid-connected systems, and high-speed railway traction power supply systems. These phenomena have become increasingly prominent with the increasing number of grid-connected converters, and have become a significant factor affecting the safe operation of power grids. The "Guidelines for Power System Safety and Stability," implemented on July 1, 2020, require that measures be implemented to mitigate the risk of subsynchronous or supersynchronous oscillations in renewable energy stations and transmission projects.

[0004] To improve the adaptability of grid-connected converters to weak grid conditions, the first existing solution is to optimize the converter's control parameters to suppress the destabilizing oscillations that can occur when the converter is connected to a weak grid. However, this solution requires upgrading the existing grid-connected converter control system, which is complex to implement. Furthermore, the required downtime for adjustments will affect normal system operation and increase workload.

[0005] The second existing solution is to add a separate converter specifically for stabilizing the grid-connected converter system in weak power grids, or to incorporate an active damping algorithm into an existing converter. Through active damping control, the converter behaves as a "virtual" damping resistor connected in parallel to the PCC point at harmonic frequencies other than the fundamental frequency, thereby suppressing potential destabilizing oscillations in multi-converter grid-connected systems. This converter, which incorporates an active damping algorithm into the control system to suppress destabilizing oscillations in grid-connected converter systems in weak power grids, is generally referred to as an active damper. It is worth noting that, compared to the first solution, this solution does not require individual modifications to existing grid-connected converter control systems, is flexible and simple, and has promising engineering application prospects.

[0006] However, the technical idea of the second solution mentioned above is to "directly divide the harmonics in the PCC point voltage by the virtual resistance value and then add it to the reference value of the grid-connected converter output current". Although the physical meaning of this approach is clear, it will introduce the grid voltage background harmonic disturbance into the converter output current reference value, thereby weakening the system's ability to suppress the grid voltage background harmonic disturbance, making it difficult to take into account the grid current quality, which is one of the most important performance indicators of the system. It is worth noting that in addition to being characterized by a large grid equivalent impedance, a weak grid usually contains rich background harmonics. Therefore, the second solution of the prior art is difficult to take into account both the stable control of the grid-connected converter system under a weak grid and the ability to suppress the grid voltage background harmonic disturbance (i.e., the grid current quality). Summary of the Invention

[0007] The purpose of the present invention is to overcome the defects of the prior art and provide an active damping method based on resonant current differential and parameter scanning adaptation. By integrating the two approaches of "resonant current differential" and "variable gain differential", active damping can take into account the requirements of both stable control of the grid-connected converter system and grid current quality under weak power grids.

[0008] The objective of the present invention is achieved as follows: an active damping method based on resonant current differentiation and parameter scanning adaptation introduces an active damping control link into the original grid-connected converter output current control system, wherein the active damping control link specifically includes extracting the harmonic components in the output current of the grid-connected converter, then passing through a variable gain differentiation link, then passing through a low-pass filtering link, and finally dividing by the virtual resistance value in the active damping control and superimposing it on the reference value of the grid-connected converter output current to achieve active damping control; a differential gain adaptive adjustment method is introduced into the active damping control link to adaptively adjust the gain of the differential link according to the grid-connected current error in steady state and the overcurrent and DC overvoltage protection signals during the device startup process.

[0009] As a further limitation of the present invention, the differential gain adaptive adjustment method includes two parts: one is the adaptive adjustment of the differential gain when the steady-state output current exhibits unstable oscillation after the device is started; the other is the adjustment of the current state of the parameter scan when overcurrent or DC overvoltage protection occurs during the startup process.

[0010] As a further limitation of the present invention, the adaptive adjustment of the differential gain when the steady-state output current exhibits unstable oscillation after the device is started specifically includes: first, performing a self-increment operation on the time counting variable, and then judging whether the value of the time counting variable is greater than or equal to the parameter scan adaptive adjustment time segment; if not, the integer variable and the parameter scan flag variable are not changed; if so, the average value of the absolute value of the error within a power frequency cycle after that moment is calculated, and the time counting variable is cleared to zero; and then further judging whether the average value of the absolute value of the error within a power frequency cycle is greater than the differential gain adjustment threshold; if not, the differential gain meets the conditions for stable operation of the system; otherwise, it means that the differential gain at this time cannot meet the conditions for stable operation of the system; and the gain of the differential link is adjusted by changing the integer variable and the parameter scan flag variable.

[0011] As a further limitation of the present invention, the adjustment of the current state of the parameter scan when overcurrent or DC overvoltage protection occurs during the startup process specifically includes: first determining whether there is an overcurrent or DC overvoltage signal during the startup process of the device. If not, it means that the differential gain at this time may be able to meet the conditions for stable operation of the system, and there is no need to change the integer variable and the parameter scan flag variable. On the contrary, it means that the differential gain at this time cannot meet the conditions for stable operation of the system, and the gain of the differential link needs to be adjusted. By changing the integer variable and the parameter scan flag variable, the current state of the parameter scan is adjusted.

[0012] As a further limitation of the present invention, determining the differential gain adjustment threshold specifically includes: performing full-load simulation under an ideal power grid without an active damping control link, solving the average value of the absolute value of the error within one power frequency cycle in steady state, and selecting a number greater than the average value according to a certain margin as the differential gain adjustment threshold.

[0013] As a further limitation of the present invention, the determination of the parameter scanning adaptive adjustment time segment specifically includes: performing simulation under the working condition that the gain of the differential link is a constant value, finding all the grid impedance intervals that enable the grid-connected converter to operate stably to reach the maximum, and the differential gain K corresponding to the sum of the various grid impedance intervals completely covering the possible variation range of the grid impedance v1 ~K vn ;

[0014] Using the differential gain K v1 ~K vn As well as the upper and lower limits of the grid impedance range in which the grid-connected converter can operate stably under the corresponding differential gain, simulation is carried out under the condition that the output current reference value of the grid-connected converter suddenly changes from 0A to full load, and the adjustment time of the system in the corresponding transient process is obtained. The maximum value of the system adjustment time in the transient process under all the above conditions is recorded as the parameter scanning adaptive adjustment time segment with a certain margin.

[0015] As a further limitation of the present invention, the integer variable and the parameter scan flag variable are key information in the differential gain array that reflects the current state of the parameter scan, and are read and stored in a non-volatile memory.

[0016] The present invention adopts the above technical solution, and compared with the prior art, the beneficial effects are as follows: the feedback signal of the active damping control in the present invention is derived from the harmonics in the output current of the grid-connected converter. The physical meaning of performing differential operation on the harmonics in the output current of the grid-connected converter is to convert the harmonic current into harmonic voltage, that is, indirectly estimate the harmonic voltage through the harmonic current. Compared with the prior art solution that directly uses the harmonic voltage as the feedback signal of the active damping control, this method of indirectly estimating the harmonic voltage from the harmonic current avoids introducing the background harmonic disturbance of the grid voltage into the reference value of the converter output current; the present invention adopts the "variable gain differential" link, which can achieve stable control of the grid-connected converter system when the grid impedance varies over a wide range under weak grid conditions. While achieving stable control of the grid-connected converter system under weak grid conditions, the system's ability to suppress background harmonic disturbance of the grid voltage, that is, the quality of the grid current, will not be sacrificed; the key control parameters in the present invention can be determined by simulation methods, and the accuracy requirements are low, the parameter design is simple, and it is conducive to engineering applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is based on the active damping control principle of resonant current differentiation and parameter scanning adaptation.

[0018] Figure 2 This is the output current control principle of the grid-connected converter when there is no active damping control link.

[0019] Figure 3 The differential gain K is not introduced v The adaptive adjustment strategy is based on the active damping control principle of resonant current differential.

[0020] Figure 4 It is the grid impedance range in which the converter can operate stably under different differential gains in the embodiment.

[0021] Figure 5 This is the initialization process for the differential gain array and parameter sweep adaptation result variables.

[0022] Figure 6(a) shows the differential gain K of the present invention. v Flowchart of the subroutine of the first part of the adaptive adjustment strategy.

[0023] FIG6(b) is the differential gain K of the present invention v Flowchart of the subroutine in the second part of the adaptive adjustment strategy.

[0024] FIG7( a ) is a simulation result of the active damper control method provided by the present invention under ideal grid conditions with p=0 and the self-adding mode as the parameter scanning starting state.

[0025] FIG7( b ) is a simulation result of the active damper control method provided by the present invention when the grid impedance is 2 mH and p=0 and the self-adding mode are used as the parameter scanning starting state.

[0026] FIG7( c ) is a simulation result of the active damper control method provided by the present invention when the grid impedance is 5 mH and p=0 and the self-adding mode are used as the parameter scanning starting state.

[0027] FIG7( d ) is a simulation result of the active damper control method provided by the present invention under ideal grid conditions with p=1 and the self-adding mode as the parameter scanning starting state.

[0028] FIG8( a ) shows the steady-state simulation results of the existing active damping control method that directly uses “harmonics in the PCC point voltage” as the feedback quantity under ideal grid conditions and conditions containing grid voltage background harmonics.

[0029] FIG8( b ) is a steady-state simulation result of the existing active damping control method that directly uses “harmonics in the PCC point voltage” as the feedback quantity under the condition of a grid impedance of 2 mH and grid voltage background harmonics.

[0030] FIG8( c ) is a steady-state simulation result of the active damper control method provided by the present invention under an ideal power grid and a working condition containing background harmonics of the power grid voltage.

[0031] FIG8( d ) is a steady-state simulation result of the active damper control method provided by the present invention under the condition that the grid impedance is 2 mH and the grid voltage contains background harmonics. DETAILED DESCRIPTION

[0032] As shown in Figure 1, an active damping method based on resonant current differentiation and parameter scanning adaptation is introduced into the original grid-connected converter output current control system. The active damping control link specifically includes extracting the harmonic components in the grid-connected converter output current, then passing it through a variable gain differentiation link, then through a low-pass filtering link, and finally dividing it by the virtual resistance value in the active damping control and adding it to the reference value of the grid-connected converter output current to achieve active damping control; the active damping control link introduces a differential gain adaptive adjustment method to adaptively adjust the gain of the differential link according to the grid-connected current error in steady state and the overcurrent and DC overvoltage protection signals during the device startup process.

[0033] The method of the present invention simply requires the introduction of an "active damping control link" into the existing grid-connected converter output current control system, without changing the existing control parameters. To fully illustrate the specific scheme and implementation effects of the present invention, a 22kvar static VAR generator is used as an example. The active damper method based on resonant current differentiation and parameter sweep adaptation provided by the present invention is specifically implemented according to the following steps:

[0034] Step 1) Refer to Figure 1 , in the figure is the reference value of the converter output current, i L is the converter output current, u g is the grid voltage, u PCC is the voltage at the grid connection point of the converter, G i (s) is the current regulator, G d (s) is the delay introduced by digital control and PWM comparison link, G p (s) is the operational admittance of the filter inductor at the output of the converter, G LPF (s) is a second-order low-pass filter, G g (s) is the grid impedance, R v is the virtual damping resistance value in active damping control, K v is the gain of the differential link, G BPF (s) is a bandpass filter, and the active damping control link is based on the grid current error E rr And the gain K of the differential link of the device's overcurrent and DC overvoltage protection signals v Make adaptive adjustments.

[0035] Step 2) Remove Figure 1 For the active damping control link in Figure 2 , without active damping control link and under ideal grid conditions (i.e., grid impedance is zero), full load simulation is performed (usually, the converter control parameters are designed to ensure that the grid-connected converter can operate stably under ideal grid conditions), and the absolute value of the error within one power frequency cycle in steady state |E rr The average value of | is selected, and a number slightly larger than the average value (which can be selected with a margin of 5% to 10%) is selected as the differential gain adjustment threshold δ used in the following steps. For example, in the embodiment, under the ideal power grid condition, |E is within one power frequency cycle at full load. rr The average value of | is 8.4A. Therefore, the differential gain adjustment threshold δ can be taken as 8.82A based on a 5% margin.

[0036] Step 3) Based on the active damping control of the resonant current differential, the grid impedance range that the system can adapt to is different under different differential link gains, and there is a rule that as the differential gain increases, the upper and lower limits of the grid impedance range that the system can adapt to will increase. Therefore, in order to make the system able to adapt to the wide range of grid impedance changes under weak grid conditions, refer to Figure 3 , without introducing the differential gain K v Adaptive adjustment strategy, that is, the gain K of the differential link v To simulate the working condition of constant value, first find out the grid impedance range [0mH, L g1_up ] reaches its maximum when the corresponding differential link gain K v1 , and then find the grid impedance range [L g2_down , L g2_up ] reaches its maximum when the corresponding differential link gain K v2 , and continue in this way to find the grid impedance range [L gn_down , L gn_up ] reaches its maximum when the corresponding differential link gain K vn , when L gn_up Greater than or equal to the maximum grid impedance value L that the converter is required to adapt to during design g_max The process ends when v1 ~K vn The sum of the grid impedance ranges in which the converter can operate stably completely covers the possible variation range of the grid impedance [0mH, L g_max ], in the above process, L g2_down The value should be slightly smaller than L g1_up (Can be selected with a margin of 5% to 10%), similarly, L g3_down The value should be slightly smaller than L g2_up , and so on, L gn_down The value should be slightly smaller than L g(n-1)_up For example, in the embodiment, the maximum grid impedance value required for the converter to operate stably is 5mH. Based on a 5% margin, the grid impedance range [0mH, L g1_up ] reaches the maximum [0mH, 0.66mH] corresponding to the differential link gain K v1 =7×10 -5 , the grid impedance range in which the grid-connected converter can operate stably [L g2_down , L g2_up ] reaches the maximum [0.627mH, 3.42mH], corresponding to the differential link gain K v2 =1.2×10 -3, the grid impedance range in which the grid-connected converter can operate stably [L g3_down , L g3_up ] reaches the maximum [3.25mH, 6.46mH], the corresponding differential link gain K v3 =6×10 -3 Since the upper limit of the interval 6.46mH is greater than the maximum grid impedance value 5mH required for the converter to operate stably during design, the process ends. Figure 4 In this embodiment, the designed K v1 , K v2 and K v3 The sum of the grid impedance ranges in which the converter can operate stably under the three differential gains can completely cover the grid impedance variation range [0mH, 5mH] required during the design.

[0037] Step 4) In the present invention, one of the criteria for determining whether the differential gain needs to be adjusted is whether the system is in a stable state under the current differential gain. This solution is achieved by determining whether the average value of the absolute value of the error within one power frequency cycle in the steady state is greater than the differential gain adjustment threshold δ designed in step 2). To determine whether the system has entered a steady state, it is necessary to find out the adjustment time of the system during the transition from transient state to steady state. For this purpose, refer to Figure 3 , without introducing the "differential gain K v When the adaptive adjustment strategy is “set”, let K v Take the values obtained in step 3) above respectively, and take the upper and lower limits of the grid impedance range in which the grid-connected converter can operate stably under the corresponding differential gain. Simulate the working condition in which the output current reference value of the grid-connected converter suddenly changes from 0A to full load, and obtain the adjustment time of the system in the corresponding transient process. The maximum value of the system adjustment time in the transient process under all the above working conditions is recorded as T r_max , then a time slightly larger than this value (which can be selected with a margin of 5% to 10%) is taken as the parameter scanning adaptive adjustment time segment λ used in the following steps. In the above steps, the grid impedance is respectively taken as the upper and lower limits of the grid impedance range in which the grid-connected converter can operate stably under each differential gain. The reason is that although the system is stable at this time, the stability margin is small, and the adjustment time of the system in the transient process is long; for example, using the differential gain K designed in step 3) v1 ~K v3 And the upper and lower limits of the grid impedance range in which the grid-connected converter can operate stably under each gain. In the embodiment, in K v =7×10 -5 , when the grid impedance is 0mH, the system adjustment time is 1.1 seconds. v =7×10 -5 , when the grid impedance is 0.66mH, the system adjustment time is 1.5 seconds.v =1.2×10 -3 , when the grid impedance is 0.627mH, the system adjustment time is 0.3 seconds. v =1.2×10 -3 , when the grid impedance is 3.42mH, the system adjustment time is 1.6 seconds. v =6×10 -3 , when the grid impedance is 3.25mH, the system adjustment time is 0.36 seconds. v =6×10 -3 , when the grid impedance is 6.46mH, the system adjustment time is 1.25 seconds; the maximum value of the system adjustment time in the transient process under all the above working conditions is 1.6 seconds. Therefore, based on a 5% margin, the "parameter scanning adaptive adjustment time segment" λ can be taken as 1.68 seconds.

[0038] Step 5) Refer to Figure 5 , define a floating point differential gain array of length n in the system control program, where each member is assigned the n differential gain K designed in step 3) above. v1 ~K vn Then, the initial values of the integer variable p (as the subscript of the variable in the differential gain array to be extracted in the following steps) and the parameter scan flag variable (as the flag indicating whether the current variable p is in the increment or decrement mode in the following steps) are read from the non-volatile memory. It should be noted that the integer variable p and the parameter scan flag variable are key information reflecting the current state of the parameter scan. In order to ensure that the starting point of the parameter scan is still the state of the last parameter scan after the system is powered off and restarted, they must be updated to the non-volatile memory when their values change. In addition, the initial value of the integer variable p when it is first used in the non-volatile memory can be set to any number between 0 and (n-1), and the initial value of the parameter scan flag variable when it is first used in the non-volatile memory can be set to any one of the self-incrementing and self-decrementing modes. For example, in the embodiment, the differential gain array length n=3, "1" is used to represent the self-incrementing mode, and "0" is used to represent the self-decrementing mode. At this time, the initial value of the integer variable p when it is first used in the non-volatile memory can be set to 0 (that is, the starting point of the parameter scan is the first variable of the differential gain array), and the initial value of the parameter scan flag variable when it is first used in the non-volatile memory can be set to 1 (that is, the starting state of the parameter scan is the self-incrementing mode).

[0039] Step 6) Using the overcurrent and DC overvoltage signals of the control system, the differential gain designed in step 2) to adjust the threshold δ, the differential gain K designed in step 3) v1 ~K vn, the parameter scan adaptive adjustment time segment λ designed in step 4), the differential gain array, integer variable p, and parameter scan flag variable in step 5) are used to design the differential gain K based on parameter scan adaption v Adaptive adjustment strategy. Referring to Fig. 6, the "differential gain K v Adaptive adjustment strategy" consists of two subroutines given in Fig. 6(a) and Fig. 6(b), and each subroutine is executed once after the system sampling time arrives. Referring to Fig. 6(a), after each system sampling time arrives, first perform an increment operation on the time count variable (the increment amount each time is the system sampling time); then judge whether the value of the time count variable is greater than or equal to the parameter scan adaptive adjustment time segment λ designed in step 4). If the condition is not met, it means that the parameter scan adaptive adjustment time has not arrived, so the variable p and the parameter scan flag variable are not changed, and the variable K v [p] in the differential gain array is still used as the gain of the differential link at this time. If the condition is met, calculate the average value of the absolute value of the error within one power frequency cycle after this moment, and clear the time count variable. Then further judge whether the average value of the absolute value of the error within one power frequency cycle is greater than the differential gain adjustment threshold δ designed in step 2). If the condition is not met, it means that the differential gain at this time can meet the condition for the stable operation of the system, so the variable p and the parameter scan flag variable do not need to be changed, and the variable K v [p] in the differential gain array is still used as the gain of the differential link at this time. If the condition is met, it means that the differential gain at this time cannot meet the condition for the stable operation of the system, and the gain of the differential link needs to be adjusted. Subsequently, judge whether the parameter scan flag at this time is in the increment-by-one mode. If it is in the increment-by-one mode, execute p = p + 1, and write the adjusted variable p into the non-volatile memory. If it is not in the increment-by-one mode (i.e., it is in the decrement-by-one mode at this time), execute p = p - 1, and write the adjusted variable p into the non-volatile memory; since the value of p should be between 0 and (n - 1), therefore, if p = n - 1, it means that the parameter has been scanned to the last variable in the differential gain array at this time, and the subsequent parameter scan flag should be changed to the decrement-by-one mode, and the adjusted parameter scan flag is written into the non-volatile memory, and then the variable K v [p] in the differential gain array is used as the differentially adjusted gain. If p = 0, it means that the parameter has been scanned to the first variable in the differential gain array at this time, and the subsequent parameter scan flag should be changed to the increment-by-one mode, and the adjusted parameter scan flag is written into the non-volatile memory, and then the variable K v [p] in the differential gain array is used as the differentially adjusted gain. If 0 < p < (n - 1), it means that the parameter has not been scanned to the first or last variable in the differential gain array at this time, so there is no need to process the parameter scan flag, and the variable Kv [p]As the differential gain after adaptive adjustment, the subroutine of this part ends;

[0040] There are two manifestations when the grid-connected converter is in an unstable state. One is that the device can start, but the steady-state output current shows unstable oscillation. In this case, the subroutine given by the above 6(a) realizes the adaptive adjustment of the differential gain K v ; The other is that the device cannot start, and overcurrent or DC overvoltage protection occurs during the startup process. Therefore, in this working condition, it is also necessary to adjust the current state of parameter scanning. Referring to Figure 6(b), after each system sampling time arrives, first judge whether there is an overcurrent or DC overvoltage signal during the startup process of the device. If the condition is not established, it means that the differential gain at this time may be able to meet the conditions for the stable operation of the system, so there is no need to change the variable p and the parameter scanning flag variable. If the condition is established, it means that the differential gain at this time cannot meet the conditions for the stable operation of the system, and the gain of the differential link needs to be adjusted. Subsequently, judge whether the parameter scanning flag at this time is in the self-increment mode. If it is in the self-increment mode, execute p = p + 1 and write the adjusted variable p into the non-volatile memory. If it is not in the self-increment mode (that is, the self-decrement mode at this time), execute p = p - 1 and write the adjusted variable p into the non-volatile memory; Since the value of p should be between 0 and (n - 1), therefore, if p = n - 1, it means that the parameter has been scanned to the last variable in the differential gain array at this time, and the subsequent parameter scanning flag should be changed to the self-decrement mode and the adjusted parameter scanning flag should be written into the non-volatile memory. If p = 0, it means that the parameter has been scanned to the first variable in the differential gain array at this time, and the subsequent parameter scanning flag should be changed to the self-increment mode and the adjusted parameter scanning flag should be written into the non-volatile memory. If 0 < p < (n - 1), it means that the parameter has not been scanned to the first or last variable in the differential gain array at this time, and there is no need to process the parameter scanning flag. The subroutine of this part ends.

[0041] To more intuitively show the effectiveness of the adaptive active damping control method based on resonant current differential and parameter scanning described in the present invention, referring to Figure 7(a), using the active damping control method described in the present invention, under ideal grid conditions with p = 0 (that is, the variable K v [0] in the differential gain array is the differential gain) and the self-increment mode as the starting state of parameter scanning, the simulation results show that since K v [0] corresponding differential gain 7×10 -5 can ensure the stable operation of the system under ideal grid conditions, so the value of the differential gain after the "parameter scanning adaptive" adjustment of the system is 7×10 -5, the converter output current is normal; Referring to FIG7 (b), the active damping control method of the present invention is adopted, and the grid impedance is 2mH under the working condition with p = 0 (ie, the variable K in the differential gain array v [0] is the differential gain) and the self-adding mode is the starting state of the parameter sweep. The simulation results show that due to K v [1] The corresponding differential gain is 1.2×10 -3 It can ensure the stable operation of the system under the condition of grid impedance of 2mH. Therefore, after the system is adjusted by "parameter scanning and adaptive", the value of the differential gain is adjusted from the initial value of 7×10 -5 Adjusted to 1.2×10 -3 , before the differential gain is adjusted, the output current of the converter oscillates greatly and the system is unstable. After the differential gain is adaptively adjusted, the output current of the converter is normal. Referring to FIG7(c), the active damping control method of the present invention is used, and the grid impedance is 5mH and p=0 (that is, the variable K in the differential gain array) is set to zero. v [0] is the differential gain) and the self-adding mode is the starting state of the parameter sweep. The simulation results show that due to K v [2] The corresponding differential gain is 6×10 -3 It can ensure the system to operate stably under the condition of grid impedance of 5mH. Therefore, after the system is adjusted by "parameter scanning and adaptive", the differential gain is first adjusted by the initial value of 7×10 -5 Adjusted to 1.2×10 -3 , and then by 1.2×10 -3 Adjusted to 6×10 -3 , when the differential gain is 7×10 -5 and 1.2×10 -3 The output current of the current transformer oscillates greatly, and the system becomes unstable. The final differential gain is adaptively adjusted to 6×10 -3 , the converter output current is normal; Referring to FIG7 (d), the active damping control method of the present invention is adopted, under ideal grid conditions, p = 1 (ie, the variable K in the differential gain array v [1] is the differential gain) and the self-adding mode is the starting state of the parameter sweep. The simulation results show that due to K v [0] The corresponding differential gain is 7×10 -5 It can ensure the stable operation of the system under ideal grid conditions. Therefore, after the system is adjusted by "parameter scanning and adaptive", the differential gain is first adjusted by the initial value of 1.2×10 -3 Adjusted to 6×10 -3 , and then by 6×10 -3 Adjusted to 1.2×10 -3 , and finally by 1.2×10 -3 Adjusted to 7×10 -5, when the differential gain is 1.2×10 -3 and 6×10 -3 The output current of the current transformer oscillates greatly, and the system becomes unstable. The final differential gain is adaptively adjusted to 7×10 -5 , the converter output current is normal;

[0042] In order to more intuitively further demonstrate the beneficial effects of the adaptive active damping control method based on resonant current differentiation and parameter scanning of the present invention compared with the existing active damping control method that directly uses "harmonics in PCC point voltage" as feedback, referring to Figure 8(a), under ideal grid conditions, when the grid voltage contains background harmonics, although the output current of the converter can be stable under the existing active damping control, the distortion is more obvious, that is, the system has a low ability to suppress the background harmonic disturbance of the grid voltage. At this time, the THD of the current is 13.8%. For comparison, referring to Figure 8(c), under the same working conditions, the output current of the converter can be stabilized under the existing active damping control. The THD of the current after the active damping control described in the present invention is 10.3%. In addition, referring to Figure 8(b), when the grid impedance is 2mH, the THD of the converter output current under the existing active damping control is 4%. For comparison, referring to Figure 8(d), the THD of the current after the active damping control described in the present invention is 1.4% under the same operating conditions. It can be seen that under the same operating conditions, the active damping control method described in the present invention can not only achieve stable control of the grid-connected converter system under weak grid conditions, but also does not sacrifice the system's ability to suppress grid voltage background harmonic disturbances, that is, the quality of the grid-connected current.

[0043] The present invention is not limited to the above-mentioned embodiments. On the basis of the technical solutions disclosed in the present invention, those skilled in the art can make some substitutions and modifications to some of the technical features therein according to the disclosed technical content without creative labor, and these substitutions and modifications are all within the protection scope of the present invention.

Claims

1. An active damping method based on resonant current differentiation and parameter sweep adaptation introduces an active damping control link into the original grid-connected converter output current control system, characterized in that: The active damping control link specifically includes extracting harmonic components from the output current of the grid-connected converter, passing them through a variable gain differential link, and then through a low-pass filtering link. Finally, the resulting components are divided by the virtual resistance value in the active damping control and added to the reference value of the grid-connected converter output current to achieve active damping control. The active damping control link introduces a differential gain adaptive adjustment method to adaptively adjust the gain of the differential link based on the grid-connected current error and the overcurrent and DC overvoltage protection signals during the device startup process. The differential gain adaptive adjustment method includes two methods: one is the adaptive adjustment of the differential gain when the steady-state output current exhibits unstable oscillation after the device is started; Second, when overcurrent or DC overvoltage protection occurs during startup, the current state of the parameter scan is adjusted; The adaptive adjustment of the differential gain when the steady-state output current exhibits unstable oscillation after the device is started specifically includes: first, performing a self-addition operation on the time counting variable, and then judging whether the value of the time counting variable is greater than or equal to the parameter scanning adaptive adjustment time segment; if not, the integer variable and the parameter scanning flag variable are not changed; if so, the average value of the absolute value of the error in the next power frequency cycle is calculated, and the time counting variable is cleared to zero; and then further judging whether the average value of the absolute value of the error in the power frequency cycle is greater than the differential gain adjustment threshold; if not, the differential gain meets the conditions for stable operation of the system; otherwise, it means that the differential gain at this time cannot meet the conditions for stable operation of the system, and the gain of the differential link is adjusted by changing the integer variable and the parameter scanning flag variable; The adjustment of the current state of the parameter scan when overcurrent or DC overvoltage protection occurs during the startup process specifically includes: first determining whether there is an overcurrent or DC overvoltage signal during the startup process of the device. If not, it means that the differential gain at this time may be able to meet the conditions for stable operation of the system, and there is no need to change the integer variable and the parameter scan flag variable. On the contrary, it means that the differential gain at this time cannot meet the conditions for stable operation of the system, and the gain of the differential link needs to be adjusted. By changing the integer variable and the parameter scan flag variable, the current state of the parameter scan is adjusted.

2. The active damping method based on resonant current differentiation and parameter sweep adaptation according to claim 1, characterized in that: Determining the differential gain adjustment threshold specifically includes: performing a full-load simulation under an ideal power grid without an active damping control link, solving the average value of the absolute value of the error within a power frequency cycle in a steady state, and selecting a number greater than the average value according to a certain margin as the differential gain adjustment threshold.

3. The active damping method based on resonant current differentiation and parameter sweep adaptation according to claim 1, characterized in that: The determination of the parameter scanning adaptive adjustment time segment specifically includes: performing simulation under the working condition that the gain of the differential link is a constant value, finding all the grid impedance intervals that enable the grid-connected converter to operate stably to reach the maximum, and the differential gain K corresponding to the sum of each grid impedance interval completely covering the possible variation range of the grid impedance v1 ~K vn ; Using the differential gain K v1 ~K vn As well as the upper and lower limits of the grid impedance range in which the grid-connected converter can operate stably under the corresponding differential gain, simulation is carried out under the condition that the output current reference value of the grid-connected converter suddenly changes from 0A to full load, and the adjustment time of the system in the corresponding transient process is obtained. The maximum value of the system adjustment time in the transient process under all the above conditions is recorded as the parameter scanning adaptive adjustment time segment with a certain margin.

4. The active damping method based on resonant current differentiation and parameter sweep adaptation according to claim 1, characterized in that: The integer variable and parameter scan flag variable are key information in the differential gain array that reflects the current state of the parameter scan, and are read and stored through a non-volatile memory.

Images