Grid-connected inverter voltage sag detection method and system based on fast amplitude estimation

Abstract

The present invention provides a method and system for detecting voltage drop of a grid-connected inverter based on rapid amplitude estimation, and relates to the field of electrical engineering. The detection method includes voltage sampling, first-order low-pass filtering, coordinate operation, calculation of the positive-sequence amplitude and negative-sequence amplitude of the voltage, sliding average filtering, and fault type judgment. In response to the problem of slow detection speed of traditional voltage drop detection methods, the method provides a rapid amplitude estimation scheme. The scheme does not require the participation of a phase-locked loop, and calculates the positive-sequence amplitude and negative-sequence amplitude of the grid-connected voltage based on real-time sampled voltage data, thereby improving the voltage drop detection speed of the grid-connected inverter. The scheme is simple and easy to implement, which is conducive to the low-voltage ride-through operation of the grid-connected inverter.

Description

Technical Field

[0001] The present invention relates to the field of electrical engineering, and in particular to a method and system for detecting voltage drop of a grid-connected inverter based on fast amplitude estimation. Background Art

[0002] With the rapid development of renewable energy power generation systems, grid-connected inverters have become widely used as the primary power conversion interface in power systems. In the event of a grid short-circuit fault, grid-connected inverters, due to their limited overload capacity, can cause widespread disconnection, further escalating the scale of the incident. To address this, many countries have developed detailed renewable energy grid specifications requiring grid-connected inverters to have low voltage ride-through (LVRT) capabilities. The ability of grid-connected inverters to quickly and accurately detect the type and depth of grid faults is not only fundamental to their LVRT capabilities but also a prerequisite for their reactive power compensation capabilities.

[0003] Typical grid-connected inverter fault detection methods mostly rely on analyzing voltage data characteristics and employing phase-locked loop (PLL) solutions. Existing voltage amplitude detection methods take into account factors such as harmonics and phase angle mutations, but are limited by the amount of voltage data and the PLL control bandwidth. This results in voltage amplitude detection requiring at least half a grid cycle, and still fails to address the slow speed of grid-connected inverter fault detection. For example:

[0004] The article entitled "Research on Voltage Effective Value Detection Method" (Wei Zhixuan, Wang Yongqiang, Gao Xiuli, Liu Hui, Wei Shutian. Research on Voltage Effective Value Detection Method [J]. Electrical Engineering, 2008(10):35+38.) proposed an effective value calculation method, which samples the voltage data of a power grid cycle and performs root mean square calculation to obtain the voltage effective value. It has high detection accuracy and anti-interference ability, but there is a transition time of a power grid cycle, poor real-time performance, and a long detection time.

[0005] The patent, titled "Low Voltage Ride Through and Islanding Effect Detection Method and Device for Photovoltaic Grid-Connected Inverters" (Li Xianyun, Zhou Yu, Wang Shuzheng. Low Voltage Ride Through and Islanding Effect Detection Method and Device for Photovoltaic Grid-Connected Inverters [P]. Jiangsu Province: CN104678326B, September 12, 2017), calculates the voltage amplitude using the maximum or minimum points of a sinusoidal voltage cycle. This method can obtain single-phase voltage amplitudes in the event of an asymmetric grid fault, with a detection time of one-quarter of the grid cycle. However, this method struggles to accurately detect voltage amplitudes when harmonics are present, significantly impacting detection accuracy.

[0006] The paper titled "A Detection of Voltage Flicker Signals Based on Short-Time Fourier Transform" (Hu Diangang, Ma Xiping, Zhao Fengzhan, et al. A Detection of Voltage Flicker Signals Based on Short-Time Fourier Transform [J]. Power Grid & Clean Energy, 2020, 36(3):8.) uses the base spectrum sequence in the short-time Fourier transform amplitude matrix of the voltage signal to obtain the power frequency amplitude of the voltage signal. This method is not only suitable for steady-state voltage amplitude detection, but also has high detection accuracy under short-term voltage variations. However, the detection speed is affected by the voltage history data, and the calculation time is relatively long.

[0007] A paper titled "Modeling and Tuning of an Improved Delayed-Signal-Cancellation PLL For Microgrid Application" (Rasheduzzaman M, Kimball J. Modeling and Tuning of an Improved Delayed-Signal-Cancellation PLL For Microgrid Application [J]. IEEE Transactions on Energy Conversion, 2018: 1-1.) proposes an improved delayed signal cancellation phase-locked loop (PLL). This approach improves the speed and accuracy of voltage amplitude detection by using a positive sequence detector and a notch filter. However, due to the stability requirements of grid-connected inverters, the control bandwidth of the PLL is typically set relatively small, limiting the speed of voltage amplitude detection.

[0008] The above analysis shows that voltage sag detection methods for grid-connected inverters have received extensive attention and research, particularly methods for improving the accuracy of voltage amplitude detection under various operating conditions. However, research on methods for improving the speed of voltage amplitude detection has been relatively limited. While some studies have improved the speed of voltage amplitude detection, this has reduced the accuracy of voltage amplitude detection and may even cause the grid-connected inverter to fail to meet low voltage ride-through standards.

[0009] In summary, the existing voltage drop detection method still has the following problems:

[0010] 1. When obtaining the voltage amplitude by analyzing the voltage data characteristics, many methods require voltage data of at least half a grid cycle to ensure the accuracy of voltage amplitude detection, but this method will also extend the voltage amplitude detection time.

[0011] 2. When using a phase-locked loop solution to detect voltage amplitude, it is limited by the control bandwidth. Although adding units such as pre-filters and notch filters can reduce the impact of frequency fluctuations and harmonics on the voltage amplitude detection accuracy, it also reduces the voltage amplitude detection speed. Summary of the Invention

[0012] This paper aims to improve voltage amplitude detection speed while ensuring accuracy. It proposes a method for detecting voltage sags in grid-connected inverters based on rapid amplitude estimation. Compared to traditional voltage amplitude detection methods, this method does not require a phase-locked loop (PLL) and instead obtains voltage amplitudes based on real-time voltage data, eliminating the impact of the amount of voltage data on voltage amplitude detection time.

[0013] The object of the present invention is achieved as follows. The present invention provides a method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation, comprising the following steps:

[0014] Step 1: Let the grid fundamental frequency be f g , the sampling frequency of the grid-connected inverter is f s , the number of sampling times in a power grid fundamental wave is N, N = f s / f g ; Sample the grid-connected voltage of the grid-connected inverter N+M times at equal time intervals to obtain N+M grid-connected voltage sampling values, and form a sequence E of the N+M grid-connected voltage sampling values, E={u k1 ,u k2 ,...,u kN ,u k(N+1) ,...,u k(N+M)}, where k is the phase sequence, k = a, b, c, M is the number of multi-sampling times, M and N are both positive integers, M < N;

[0015] From u k1 Start and push back one grid-connected voltage sampling value each time, intercept a sequence of N grid-connected voltage sampling values from sequence E for M+1 times, and form M+1 sampling arrays X m , m is the sampling array X m The serial number, m=0,1,...,M, in each sampling array X m contains N grid-connected voltage sampling values, and any grid-connected voltage sampling value is recorded as the grid-connected voltage u imk , i=1,2,...,N;

[0016] Step 2: Low-pass filter the sample array X m Each grid voltage u imk Perform first-order low-pass filtering to obtain the grid-connected voltage after filtering out low-order harmonics, and record it as the filtered voltage u′ imk ;

[0017] Step 3: filter the voltage u′ imk Expressed as u′ ima ,u′ imb ,u′ imc, and then through Clarke transformation, we get the α-axis filtered voltage u in the two-phase stationary coordinate system imα and β-axis filtered voltage u imβ ;

[0018] Step 4: α-axis filtered voltage u obtained in step 3 imα and β-axis filtered voltage u imβ , the positive sequence amplitude of the filtered voltage e is obtained through the amplitude calculation formula im1 and the negative sequence amplitude of the filtered voltage e im2 ;

[0019] Step 5: Calculate the positive sequence amplitude of the filtered voltage e im1 and the negative sequence amplitude of the filtered voltage e im2 Perform sliding average filtering to obtain the grid voltage positive sequence amplitude e m1 and the negative sequence amplitude of the grid voltage e m2 ;

[0020] Step 6: Note that the rated value of the grid voltage is E rate , by detecting the voltage drop state of the grid-connected inverter, the fault type is judged.

[0021] Preferably, the M+1 sampling arrays X in step 1 m The formation process is as follows:

[0022] Introduce array X, X={x1,x2,...,x i ,...,x N}, where x i is any element in array X and is recorded as element x i , set the initial value of all elements in array X to 0;

[0023] Sampling and forming the obtained N+M grid-connected voltage sampling values into a sequence E;

[0024] The first time {u k1 ,u k2 ,...,u kN} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E0;

[0025] The second time we extract {u k2 ,...,u kN ,u k(N+1)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E1;

[0026] The third time {u k3 ,...,u k(N+1) ,u k(N+2)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E2;

[0027] Similarly, the M+1th interception of {u k(M+1) ,u k(M+2) ,...,u k(N+M)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E M ;

[0028] That is, through the above truncation, we can obtain M+1 sequences E m , the M+1 sequences E m The grid-connected voltage sampling values in the array are filled into the array X according to the original arrangement order, so that M+1 sampling arrays X are obtained. m .

[0029] Preferably, the transfer function of the low-pass filter in step 2 is:

[0030]

[0031] Where, T filter is the time constant of the low-pass filter, and s is the Laplace operator.

[0032] Preferably, the coordinate conversion formula of the Clarke transformation in step 3 is:

[0033]

[0034] Preferably, the amplitude calculation formula in step 4 is:

[0035]

[0036] Where ω is the voltage angular frequency, u′ imα is the α-axis filtered voltage u imα The first derivative of u′ imβ is the β-axis filtered voltage u imβ The first derivative of .

[0037] Preferably, the expression of the sliding average filter in step 5 is:

[0038]

[0039] Preferably, the specific method of determining the fault type by detecting the voltage drop state of the grid-connected inverter in step 6 is as follows:

[0040] When e m1 ≥0.9×E rate When , the grid voltage amplitude corresponding to the mth recursive sampling moment is within the normal range, and there is no fault in the grid;

[0041] When e m1 <0.9×E rate and e m2 =0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops symmetrically;

[0042] When e m1 <0.9×E rate and e m2 ≠0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops asymmetrically.

[0043] The present invention also provides a grid-connected inverter voltage drop detection system based on rapid amplitude estimation, comprising a microprocessor and a memory connected to each other, wherein the memory includes a computer-readable storage medium, and the microprocessor is programmed or configured to execute the steps of the grid-connected inverter voltage drop detection method based on rapid amplitude estimation.

[0044] The present invention also provides a computer-readable storage medium storing a computer program programmed or configured to execute the method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] 1. No phase-locked loop is required, and the voltage amplitude detection speed is not limited by the amount of voltage data. The voltage amplitude detection time is reduced by real-time calculation of the positive and negative sequence amplitudes of the grid-connected voltage.

[0047] 2. The voltage detection accuracy of the present invention is high and it can accurately detect the positive sequence amplitude and negative sequence amplitude of the grid-connected voltage.

[0048] 3. The control method of the present invention only performs mathematical processing on the real-time sampled voltage data, and the method is simple and easy to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A block diagram of the detection method of the present invention;

[0050] Figure 2 is a flow chart of the detection method of the present invention;

[0051] Figure 3 A topological diagram of a grid-connected inverter used in the simulation of the present invention;

[0052] Figure 4 The results of the positive and negative sequence voltage amplitude detection of the grid-connected inverter using phase-locked loop technology after the three-phase grid voltage drops to 0.2pu.

[0053] Figure 5The detection results of the positive-sequence amplitude and negative-sequence amplitude of the grid-connected voltage after the three-phase grid voltage drops to 0.2 pu in the grid-connected inverter adopting the voltage drop detection technology of the present invention are as follows;

[0054] Figure 6 The results of the positive and negative sequence amplitude detection of the grid voltage after the single-phase grid voltage drops to 0.2 pu in a grid-connected inverter using phase-locked loop technology are shown in the figure.

[0055] Figure 7 The present invention provides detection results of the positive-sequence amplitude and negative-sequence amplitude of the grid-connected voltage after the single-phase grid voltage drops to 0.2 pu in a grid-connected inverter that adopts the voltage drop detection technology of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is further described below with reference to the accompanying drawings.

[0057] Figure 1 is a block diagram of the detection method of the present invention, Figure 2 It is the implementation flow chart of the detection method of the present invention, Figure 1 and Figure 2 It can be seen that the grid-connected inverter voltage drop detection method based on fast amplitude estimation of the present invention includes voltage sampling, first-order low-pass filtering, coordinate operation, calculation of voltage positive sequence amplitude and negative sequence amplitude, sliding average filtering and fault type judgment.

[0058] Specifically, the following steps are included:

[0059] Step 1: Let the grid fundamental frequency be f g , the sampling frequency of the grid-connected inverter is f s , the number of sampling times in a power grid fundamental wave is N, N = f s / f g ; Sample the grid-connected voltage of the grid-connected inverter N+M times at equal time intervals to obtain N+M grid-connected voltage sampling values, and form a sequence E of the N+M grid-connected voltage sampling values, E={u k1 ,u k2 ,...,u kN ,u k(N+1) ,...,u k(N+M)}, where k is the phase sequence, k = a, b, c, M is the number of multi-sampling times, M and N are both positive integers, M < N;

[0060] From u k1 Start and push back one grid-connected voltage sampling value each time, intercept a sequence of N grid-connected voltage sampling values from sequence E for M+1 times, and form M+1 sampling arrays X m , m is the sampling array X m The serial number, m=0,1,...,M, in each sampling array Xm contains N grid-connected voltage sampling values, and any grid-connected voltage sampling value is recorded as the grid-connected voltage u imk , i=1,2,...,N;

[0061] In this embodiment, the M+1 sampling arrays X m The formation process is as follows:

[0062] Introduce array X, X={x1,x2,...,x i ,...,x N}, where x i is any element in array X and is recorded as element x i , set the initial value of all elements in array X to 0;

[0063] Sampling and forming the obtained N+M grid-connected voltage sampling values into a sequence E;

[0064] The first time {u k1 ,u k2 ,...,u kN} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E0;

[0065] The second time we extract {u k2 ,...,u kN ,u k(N+1)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E1;

[0066] The third time {u k3 ,...,u k(N+1) ,u k(N+2)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E2;

[0067] Similarly, the M+1th interception of {u k(M+1) ,u k(M+2) ,...,u k(N+M)} segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E M ;

[0068] That is, through the above truncation, we can obtain M+1 sequences E m ,m=0,1,...,M,M+1 sequences E M The grid-connected voltage sampling values in the array are filled into the array X according to the original arrangement order, so that M+1 sampling arrays X are obtained. m .

[0069] Step 2: Low-pass filter the sample array X m Each grid voltage u imk Perform first-order low-pass filtering to obtain the grid-connected voltage after filtering out low-order harmonics, and record it as the filtered voltage u′ imk .

[0070] In this embodiment, the transfer function of the low-pass filter is:

[0071]

[0072] Where, T filter is the time constant of the low-pass filter, and s is the Laplace operator.

[0073] Step 3: filter the voltage u′ imk Expressed as u′ ima ,u′ imb ,u′ imc , and then through Clarke transformation, we get the α-axis filtered voltage u in the two-phase stationary coordinate system imα and β-axis filtered voltage u imβ .

[0074] In this embodiment, the coordinate conversion formula of the Clarke transformation is:

[0075]

[0076] Step 4: According to the α-axis filtered voltage u obtained in step 4 imα and β-axis filtered voltage u imβ , the positive sequence amplitude of the filtered voltage e is obtained through the amplitude calculation formula im1 and the negative sequence amplitude of the filtered voltage e im2 .

[0077] In this embodiment, the amplitude calculation formula is:

[0078]

[0079] Where ω is the voltage angular frequency, u′ imα is the α-axis filtered voltage u imα The first derivative of u′ imβ is the β-axis filtered voltage u imβ The first derivative of .

[0080] Step 5: Calculate the positive sequence amplitude of the filtered voltage e im1 and the negative sequence amplitude of the filtered voltage e im2 Perform sliding average filtering to obtain the grid voltage positive sequence amplitude e m1 and the negative sequence amplitude of the grid voltage e m2 .

[0081] In this embodiment, the expression of the sliding average filter is:

[0082]

[0083] Step 6: Note that the rated value of the grid voltage is E rate , by detecting the voltage drop state of the grid-connected inverter, the fault type is judged.

[0084] In this embodiment, the specific method is as follows:

[0085] When e m1 ≥0.9×E rate When , the grid voltage amplitude corresponding to the mth recursive sampling moment is within the normal range, and there is no fault in the grid;

[0086] When e m1 <0.9×E rate and e m2 =0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops symmetrically;

[0087] When e m1 <0.9×E rate and e m2 ≠0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops asymmetrically.

[0088] The present invention also provides a grid-connected inverter voltage drop detection system based on rapid amplitude estimation, comprising a microprocessor and a memory connected to each other, wherein the memory includes a computer-readable storage medium, and the microprocessor is programmed or configured to execute the steps of the grid-connected inverter voltage drop detection method based on rapid amplitude estimation.

[0089] The present invention also provides a computer-readable storage medium storing a computer program programmed or configured to execute the method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation.

[0090] In order to prove the technical achievements of this scheme, MATLAB / Simulink simulation was carried out.

[0091] Figure 3 is the topology diagram of the grid-connected inverter in the simulation. Figure 3 It can be seen that the topology includes the DC side power supply U dc , three-phase full-bridge inverter circuit, filter inductor L f , filter capacitor C f , grid-connected equivalent inductance L g and three-phase grid E g , the three-phase full-bridge inverter circuit is connected in series with the DC side power supply Udc and filter inductor L f Between, filter capacitor C f Connected in parallel to the filter inductor L f and grid-connected equivalent inductance L g Between, the grid equivalent inductance L g Connect to three-phase grid E g .

[0092] The specific parameters are set as follows: the base frequency of the grid-connected voltage is 50Hz, the sampling frequency of the grid-connected inverter is 16kHz, and the rated value of the grid-connected voltage is E rate It is 311V, N=320, M=29.

[0093] By establishing a simulation model of the system, two voltage fault types are set: three-phase voltage symmetrically dropping to 0.2pu and single-phase voltage dropping to 0.2pu. The phase-locked loop and the voltage drop detection method of the present invention are used for simulation respectively, and the voltage drop detection method is plotted. Figure 4 、 Figure 5 、 Figure 6 and Figure 7 .

[0094] Attachment Figure 4 、 5 The horizontal axis of Figures 6 and 7 is time, and the vertical axis is the positive sequence amplitude and negative sequence amplitude of the grid voltage. Due to the limitation of the control bandwidth, the phase-locked loop needs 15 milliseconds to accurately detect the positive sequence amplitude and negative sequence amplitude of the grid voltage. Figure 4 、 6 As shown. The voltage drop detection method of the present invention does not require the participation of a phase-locked loop, and the positive sequence amplitude and negative sequence amplitude of the grid-connected voltage are obtained based on real-time voltage data. The detection time is 3 milliseconds, as shown. Figure 5 、 7 The actual simulation results are consistent with the analysis.

Claims

1. A method for detecting voltage sag of a grid-connected inverter based on fast amplitude estimation, characterized in that: The following steps are involved: Step 1: Let the grid fundamental frequency be f g , the sampling frequency of the grid-connected inverter is f s , the number of sampling times in a power grid fundamental wave is N, N = f s / f g ; Sample the grid-connected voltage of the grid-connected inverter N+M times at equal time intervals to obtain N+M grid-connected voltage sampling values, and form a sequence E of the N+M grid-connected voltage sampling values, E={u k1 ,u k2 ,...,u kN ,u k(N+1) ,...,u k(N+M) }, where k is the phase sequence, k = a, b, c, M is the number of multi-sampling times, M and N are both positive integers, M < N; From u k1 Start and push back one grid-connected voltage sampling value each time, intercept a sequence of N grid-connected voltage sampling values from sequence E for M+1 times, and form M+1 sampling arrays X m , m is the sampling array X m The serial number, m = 0, 1, ..., M, in each sampling array X m contains N grid-connected voltage sampling values, and any grid-connected voltage sampling value is recorded as the grid-connected voltage u imk , i=1, 2, ..., N; Step 2: Low-pass filter the sample array X m Each grid voltage u imk Perform first-order low-pass filtering to obtain the grid-connected voltage after filtering out low-order harmonics, and record it as the filtered voltage u′ imk ; Step 3: filter the voltage u′ imk Expressed as u′ ima , u′ imb , u′ imc , and then through Clarke transformation, we get the α-axis filtered voltage u in the two-phase stationary coordinate system imα and β-axis filtered voltage u imβ ; Step 4: α-axis filtered voltage u obtained in step 3 imα and β-axis filtered voltage u imβ , the positive sequence amplitude of the filtered voltage e is obtained through the amplitude calculation formula im1 and the negative sequence amplitude of the filtered voltage e im2 ; Step 5: Calculate the positive sequence amplitude of the filtered voltage e im1 and the negative sequence amplitude of the filtered voltage e im2 Perform sliding average filtering to obtain the grid voltage positive sequence amplitude e m1 and the negative sequence amplitude of the grid voltage e m2 ; Step 6: Note that the rated value of the grid voltage is E rate , by detecting the voltage drop state of the grid-connected inverter, the fault type is judged.

2. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The M+1 sampling array X described in step 1 m The formation process is as follows: Introduce array X, X={x1,x2,...,x i ,...,x N }, where x i is any element in array X and is recorded as element x i , set the initial value of all elements in array X to 0; Sampling and forming the obtained N+M grid-connected voltage sampling values into a sequence E; The first time {u k1 ,u k2 ,...,u kN } segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E0; The second time we extract {u k2 ,...,u kN ,u k(N+1) } segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E1; The third time {u k3 ,...,u k(N+1) ,u k(N+2) } segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E2; Similarly, the M+1th interception of {u k(M+1) ,u k(M+2) ,...,u k(N+M) } segment and form a new sequence consisting of N grid-connected voltage sampling values and record it as sequence E M ; That is, through the above truncation, we can obtain M+1 sequences E m , the M+1 sequences E m The grid-connected voltage sampling values in the array are filled into the array X according to the original arrangement order, so that M+1 sampling arrays X are obtained. m .

3. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The transfer function of the low-pass filter in step 2 is: Where, T filter is the time constant of the low-pass filter, and s is the Laplace operator.

4. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The coordinate conversion formula of the Clarke transformation in step 3 is:

5. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The amplitude calculation formula in step 4 is: Where ω is the voltage angular frequency, u′ imα is the α-axis filtered voltage u imβ The first derivative of u′ imβ is the β-axis filtered voltage u imβ The first derivative of .

6. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The expression of the sliding average filter in step 5 is:

7. The method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to claim 1, wherein: The specific method of determining the fault type by detecting the voltage drop state of the grid-connected inverter in step 6 is as follows: When e m1 ≥0.9×E rate When , the grid voltage amplitude corresponding to the mth recursive sampling moment is within the normal range, and there is no fault in the grid; When e m1 <0.9×E rate and e m2 =0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops symmetrically; When e m1 <0.9×E rate and e m2 ≠0, the grid voltage amplitude corresponding to the mth recursive sampling moment drops asymmetrically.

8. A grid-connected inverter voltage drop detection system based on fast amplitude estimation, comprising a microprocessor and a memory connected to each other, characterized in that: The memory includes a computer-readable storage medium, and the microprocessor is programmed or configured to execute the steps of the method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program programmed or configured to execute the method for detecting voltage sag of a grid-connected inverter based on rapid amplitude estimation according to any one of claims 1 to 7.

Images