Method for acquiring voltage positive sequence fundamental frequency component under distortion voltage of island micro-grid

Abstract

The invention discloses a method for acquiring a voltage positive sequence fundamental frequency component under distorted voltage of an island micro-grid, after the micro-grid is distorted, a vector of the micro-grid does not have a fixed amplitude and a fixed rotating frequency, a phase-locked loop amplifier (PLL) cannot directly monitor distorted vector voltage in real time, and the distorted vector voltage cannot be monitored in real time. The phase and frequency information of a positive sequence fundamental frequency component FPSC of the detection voltage is adopted to replace the vector information of the distorted voltage; a positive and negative sequence component separation method of a distorted voltage vector containing a negative sequence fundamental frequency component and different order harmonics is analyzed and deduced, so that required FPSC information of the positive sequence fundamental frequency component is obtained, and the influence of low order harmonics on detection voltage frequency and phase functions required by locking of a phase-locked loop is removed; and a theoretical basis is provided for phase locking and frequency locking of the power grid when the island micro-grid breaks down.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grids, and in particular relates to a method for obtaining a voltage positive sequence fundamental frequency component under a distorted voltage of an island microgrid. Background Art

[0002] At present, with the continuous occurrence of the reality of power shortage and development obstruction on islands, a new form of island power grid based on the coordinated control and efficient interaction of sources, storage and loads, namely the island microgrid system, has emerged. The island microgrid is a micro-power system applied to islands, based on the energy available on the island, and composed of various forms of power generation and supply systems, energy storage equipment and power electronic devices.

[0003] Due to the diversity of the power supply side and the power consumption side in the system, the island microgrid needs to use a large number of power electronic devices, such as transformers, converters and inverters, to uniformly manage and control the power in the distribution system. The power grid will inevitably invade various harmonics, which will cause serious distortion of the power grid voltage. In addition, the inverter will further cause the frequency jump of the power grid during the switching process between grid connection and off-grid, as well as the uncertainty of green energy and the pulse power characteristics of high-power loads, thereby greatly reducing the power quality of the power grid and greatly increasing the difficulty of grid connection control. If the island microgrid is connected to the grid, the voltage frequency and phase of the island microgrid and the main grid are not synchronized, or the voltage frequency and phase of the power consumption side and the power supply side of the island microgrid are not synchronized, the island microgrid cannot be connected to the grid normally. At the same time, the voltage asynchrony will cause a huge voltage drop on both sides of the grid, thereby forming a huge impact current, burning the circuit and destroying the entire power grid system.

[0004] Therefore, when the island microgrid is connected to the main grid or its power supply side, a synchronization technology is needed to grasp the voltage frequency difference and phase difference on both sides in real time and maintain stable synchronization of the voltage on both sides. It can quickly and accurately lock the real-time phase and frequency information of the grid voltage to be detected, which can safeguard the safe and stable operation of the power system. At the same time, its structure is relatively simple and easy to implement in engineering applications. Summary of the invention

[0005] The present invention provides a method for obtaining the voltage positive-sequence fundamental frequency component under the distorted voltage of an island microgrid, which can solve the problem of various power grid faults caused by severe voltage distortion in the island microgrid system, such as single-phase drop, three-phase drop, phase mutation, and frequency jump of the power grid voltage. The technical solution of the present invention can quickly obtain the voltage positive-sequence fundamental frequency component to achieve the purpose of phase locking and frequency locking.

[0006] To solve the above problems, the technical solution provided by the present invention is as follows:

[0007] An embodiment of the present invention provides a method for obtaining a voltage positive sequence fundamental frequency component under a distorted voltage of an island microgrid, comprising the following steps:

[0008] Step 1: After the microgrid is distorted, its vector does not have a fixed amplitude and a fixed rotation frequency. The phase-locked loop amplifier PLL cannot directly monitor the distorted vector voltage in real time. The phase and frequency information of the positive sequence fundamental frequency component FPSC of the detection voltage is used to replace the vector information of the distorted voltage.

[0009] Step 2, based on the research on two types of orthogonal signal generators, namely, the orthogonal signal generator based on inverse Park transform and the orthogonal signal generator based on adaptive filter, three positive and negative sequence decoupling methods, namely, direct decoupling method, positive and negative sequence calculation method, and complex coefficient filtering method are used to realize the positive and negative sequence separation of voltage, and then obtain the positive sequence fundamental frequency component FPSC information, and remove the influence of low-order harmonics on the detection voltage frequency and phase function required for the phase-locked loop to achieve locking.

[0010] In an optional embodiment of the present invention, the three-phase voltage of the microgrid in step 1 is affected by the converter and nonlinear electrical equipment, and a variety of harmonic currents of different orders and frequencies enter the microgrid, thereby causing the three-phase grid voltage to be distorted. The three-phase voltage is represented by a collection of multiple unbalanced and different-order harmonic components.

[0011] In an optional embodiment of the present invention, step 2 specifically includes: the direct decoupling method in step S2 includes two methods: the first is to extract the αβ AC component of the fundamental frequency in the αβ stationary coordinate system, and the second is to extract the dq DC component in the dq rotating coordinate system; voltage PNSC separation is achieved by the direct decoupling method and the complex coefficient filtering method. Among the two methods, the orthogonal signal generator module can not only generate orthogonal signals, but also remove harmonics, so the orthogonal signal generator is first analyzed, and then various positive and negative sequence decoupling methods are analyzed.

[0012] In an optional embodiment of the present invention, step 2 specifically includes: step S21, the orthogonal signal generator module adopts an IPT-QSG module and an AF-QSG module, the IPT-QSG module eliminates various harmonics mixed into the input signal through an LPF based on the dq DC components, and the AF-QSG module eliminates various harmonics mixed into the input signal through a BPF based on the αβ AC components.

[0013] In an optional embodiment of the present invention, step 2 specifically includes: step S22, based on the αβ component filtering direct decoupling method, firstly convert the distorted voltage v abc Separated into v in the stationary coordinate system α and v β Components, and then v α and v βThe FPNSC is decoupled to obtain v abc FPSC and FNSC; by using QSG module to achieve v α and v β The PNSC is decoupled, and the fundamental frequency component and its orthogonal quantity are obtained while filtering out the harmonics.

[0014] In an optional embodiment of the present invention, step 2 specifically includes: step S23, based on the dq component filtering direct decoupling method, the distorted voltage v abc Decomposed into v in the synchronously rotating coordinate system SRF d and v q , and then directly output after LPF filtering; among them, DDSRF is to set up two SRFs, a positive sequence component synchronous coordinate system PSC-SRF, and a negative sequence component synchronous coordinate system NSC-SRF. The two SRFs are used to eliminate the influence of PSC on NSC measured in NSC-SRF and the influence of NSC on voltage PSC measured in PSC-SRF, thereby realizing the decoupling between PNSCs.

[0015] In an optional embodiment of the present invention, the positive and negative sequence calculation method of step 2 includes: when the voltage is distorted, the symmetry of the three-phase voltage is destroyed, and the three-phase voltage is composed of a group of asymmetric three-phase quantities. The positive and negative sequence calculation method PNSC decomposes the asymmetric three-phase quantities into three groups of different symmetric three-phase quantities, and then uses the three groups of symmetric quantities to eliminate the mutual interference between PSC and NSC, thereby separating PSC and NSC.

[0016] In an optional embodiment of the present invention, the complex coefficient filtering method of step 2 includes: in the stationary coordinate system, the α component and the β component are phase-shifted by 90° through the complex symbol j, thereby realizing imaginary coefficient filtering, and the imaginary coefficient filter can accurately and quickly obtain the required component; modifying the frequency value ω of the imaginary part 0 and ω c , you can set the filter's passband and bandwidth frequency. When you set ω 0 When the grid voltage frequency ω' is obtained by the phase-locked loop, the system can realize frequency tracking and adaptive filtering of the grid voltage, thereby obtaining various accurate grid information.

[0017] In an optional embodiment of the present invention, step 2 specifically includes: step 24, the MCCF parallel method adopts n CCFs with different filtering frequencies connected in parallel, the input voltage passes through the CCFs with different filtering frequencies, and the harmonics of the corresponding frequencies and orders are filtered out and then separated into voltage positive and negative sequence components of the corresponding frequencies and orders, the voltage positive and negative sequence components output by the CCF are fed back to the input voltage, the input voltage is analyzed for information, the harmonics of the relevant frequencies and orders fed back are filtered out, and then the voltage information is input to each CCF again, and the frequency adaptive function of the MCCF parallel module is implemented through frequent feedback and response; by using multiple CCF parallel structures to set different selection frequencies, it is possible to filter out harmonics of different frequencies and orders from the voltage, refine the selection frequency range of the filter module, and thereby improve the filtering effect of the filter module.

[0018] Compared with the prior art, the embodiment of the present invention provides a method for obtaining the positive-sequence fundamental frequency component of the voltage under the distorted voltage of the island microgrid, which has the following beneficial effects: the present invention analyzes and summarizes the three-phase voltage vector when distortion occurs, and obtains the required positive-sequence fundamental frequency component of the voltage from the distorted voltage through three positive and negative sequence decoupling methods, namely, direct decoupling method, positive and negative sequence calculation method, and complex coefficient filtering method, providing a theoretical basis for timely phase and frequency locking of the island microgrid when a fault occurs. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A flowchart of a method for obtaining a voltage positive-sequence fundamental frequency component under a distorted voltage of an island microgrid provided in an embodiment of the present application.

[0021] Figure 2 A schematic diagram of a PNSC decoupling method for obtaining a voltage positive sequence fundamental frequency component under a distorted voltage of an island microgrid provided in an embodiment of the present application.

[0022] Figure 3 This is a structural diagram of the IPT-QSG provided in an embodiment of the present application.

[0023] Figure 4 Bode diagram of the IPT-QSG output transfer function provided in an embodiment of the present application.

[0024] Figure 5 This is a diagram of the ANC structure provided in an embodiment of the present application.

[0025] Figure 6 This is a structural diagram of the AF system when N=1 provided in an embodiment of the present application.

[0026] Figure 7 This is a structural diagram of the EPLL provided in an embodiment of the present application.

[0027] Figure 8 This is a structural diagram of the EPLL-QSG provided in an embodiment of the present application.

[0028] Fig. 9 This is a structural diagram of the AF system when N=2 provided in an embodiment of the present application.

[0029] Fig.10 This is a structural diagram of the continuous-range AF system provided in an embodiment of the present application.

[0030] Fig.11 This is a diagram of the second-order AF-QSG structure provided in an embodiment of the present application.

[0031] Fig.12 A schematic diagram of AF time domain response when the input provided in the embodiment of the present application is a sinusoidal signal.

[0032] Fig.13 This is a SOAF structure diagram provided in an embodiment of the present application.

[0033] Fig.14 The Bode diagram of SOGI-AF provided in the embodiment of the present application.

[0034] Fig.15 Schematic diagram of the direct decoupling method based on positive and negative sequence αβ component filtering provided in an embodiment of the present application.

[0035] Fig.16 Schematic diagram of the amplitude-frequency response of the PNSC using the DDSOGI direct decoupling method provided in an embodiment of the present application.

[0036] Fig.17 This is a structural diagram of the DDSRF decoupling method provided in an embodiment of the present application.

[0037] Fig.18 This is a partial structural diagram of the IPT-QSG-PNSC provided in an embodiment of the present application.

[0038] Fig.19 This is another partial structural diagram of the IPT-QSG-PNSC provided in an embodiment of the present application.

[0039] Fig. 20 This is a structural diagram of the CCF function module provided in an embodiment of the present application.

[0040] Fig.21 A schematic diagram of an implementation method of the function module of the CCF provided in an embodiment of the present application.

[0041] Fig. 22 Schematic diagram of the MCCF parallel method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0042] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0043] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a method for obtaining a voltage positive sequence fundamental frequency component under a distorted voltage of an island microgrid, comprising the following steps:

[0044] Step 1: After the microgrid is distorted, its vector does not have a fixed amplitude and a fixed rotation frequency. The phase-locked loop amplifier PLL cannot directly monitor the distorted vector voltage in real time. The phase and frequency information of the positive sequence fundamental frequency component FPSC of the detection voltage is used to replace the vector information of the distorted voltage.

[0045] Step 2, based on the research on two types of orthogonal signal generators, namely, the orthogonal signal generator based on inverse Park transform and the orthogonal signal generator based on adaptive filter, three positive and negative sequence decoupling methods, namely, direct decoupling method, positive and negative sequence calculation method, and complex coefficient filtering method are used to realize the positive and negative sequence separation of voltage, and then obtain the positive sequence fundamental frequency component FPSC information, and remove the influence of low-order harmonics on the detection voltage frequency and phase function required for the phase-locked loop to achieve locking.

[0046] The three-phase voltage of the microgrid in step 1 is affected by the converter and nonlinear electrical equipment. A variety of harmonic currents of different orders and frequencies will enter the power grid system, which will cause the three-phase grid voltage to be distorted. The three-phase voltage is represented by a collection of multiple unbalanced and different-order harmonic components, as shown in formula (2.1).

[0047]

[0048] in, The instructions are as follows:

[0049]

[0050]

[0051]

[0052] The transformer of the distributed power system is connected to the grid in a Δ-Y connection. There will be no zero-sequence current entering the grid, so the analysis will ignore Impact on the voltage vector.

[0053] Generally, the FPSC vector containing positive and negative sequence nth harmonics can be expressed as shown in formula (2.2).

[0054]

[0055] By further simplifying the formula, we can get:

[0056]

[0057] Among them, when n<0, it represents the negative sequence component (Negative Sequence Component, NSC), and when n>0, it represents the positive sequence component (Positive Sequence Component, PSC).

[0058] After Clarke transformation, equation (2.3) can be expressed as equation (2.4) in the αβ stationary coordinate system.

[0059]

[0060] in,

[0061] Then, formula (2 . 3) After Park transformation, the dq rotating coordinate system can be expressed as (2 . 6).

[0062]

[0063] in, Where θ' is the phase angle of the rotating coordinate system.

[0064] When the voltage vector rotates synchronously with the rotating coordinate system, θ' = ωt, so equation (2.6) can be converted to

[0065]

[0066] From equation (2.8), in the dq rotating coordinate system, the nth PSC is reflected as the (n-1)th harmonic on the FPSC and as the -(n+1)th harmonic on the FNSC. At this time, the amplitude and phase information of the voltage vector are as follows.

[0067]

[0068] From the above, it can be seen that the amplitude |U| and phase θ of the distorted voltage are always changing.

[0069] Step 2 specifically includes: From equations (2.4) and (2.6), it can be obtained that the direct decoupling method in step S2 includes two methods: the first is to extract the αβ AC component of the fundamental frequency in the αβ stationary coordinate system, and the second is to extract the dq DC component in the dq rotating coordinate system; voltage PNSC separation is achieved through direct decoupling method and complex coefficient filtering method. Among these two methods, the orthogonal signal generator module can not only generate orthogonal signals, but also remove harmonics. Therefore, the orthogonal signal generator is analyzed first, and then various positive and negative sequence decoupling methods are analyzed.

[0070] Step 2 specifically includes: Step S21, the orthogonal signal generator module adopts the IPT-QSG module and the AF-QSG module, the IPT-QSG module eliminates various harmonics mixed into the input signal through LPF based on the dq DC components, and the AF-QSG module eliminates various harmonics mixed into the input signal through BPF based on the αβ AC components.

[0071] 1) Orthogonal signal generator based on inverse Park transform

[0072] The quadrature signal generator (QSG) can output a signal that is orthogonal to the input sinusoidal signal. Under ideal conditions, the QSG module can not only output the orthogonal signal of the required frequency in real time, but also eliminate various harmonics mixed into the input signal.

[0073] The Park transform converts the αβ orthogonal sinusoidal signal in the αβ stationary coordinate system into the dq DC component in the dq rotating coordinate system. Conversely, the inverse Park transform converts the dq DC component in the dq rotating coordinate system into the αβ orthogonal sinusoidal signal in the αβ stationary coordinate system.

[0074] For a specific phase angle, the Park transform and inverse Park transform formulas are shown below.

[0075]

[0076] Therefore, by adding a low pass filter (LPF) to the loop control circuit containing Park transform and inverse Park transform, a set of orthogonal output signals can be obtained, which is called an inverse Park transform based orthogonal signal generator (IPT-QSG). Figure 3 This is the IPT-QSG module structure diagram.

[0077] through Figure 3 Analysis shows that when v α and v' β When it is not orthogonal, vα and v' β The voltage vector has no constant amplitude and no constant rotation frequency, so the v obtained after Park transformation is d and v q Oscillation occurs, v d and v q After filtering out the oscillation through LPF, it becomes and After the inverse Park transform, it is converted into a set of orthogonal signals v' α and v' β , at this time v α and v' α out of phase. As the PLL locks the phase of the input signal, v α Gradually becoming with v' α In phase with v' β Orthogonal.

[0078] Figure 3 The detailed derivation is as follows. The Park transformation in equation (2.11) is converted by Euler formula, and assuming that θ'=ω't, we can get:

[0079]

[0080] The above formula can be obtained by Laplace transformation:

[0081]

[0082] Assume that the cutoff frequency of LPF is ω f A first-order low-pass filter can be obtained:

[0083]

[0084] This serves as the input to the inverse Park transform.

[0085] Expressing the inverse Park transform in equation (2.12) using the Euler formula yields:

[0086]

[0087] Similarly, through Laplace transformation, we can get:

[0088]

[0089] Simplifying equations (2.14), (2.15) and (2.17) together, we can get:

[0090]

[0091] in,

[0092]

[0093] Simplifying, we get:

[0094]

[0095] By v' α and v' β Orthogonal

[0096]

[0097] By Laplace transformation, we can get

[0098]

[0099] Substituting equation (2.22) into equation (2.20) yields

[0100]

[0101] From equations (2.20) and (2.23), we can get that v' in IPT-QSG α v α The effect of the band pass filter (BPF) on v α The effect is the same as v' β v α The effect of the second-order LPF on v α The effect is the same. Among them, the center frequency ω c is the rotation frequency of the dq synchronous coordinate system, i.e. ω'; the damping coefficient is determined by k, k = 2ξ.

[0102] Assuming the cutoff frequency ω f =2π·70.7rad / s, input signal frequency ω 0 =2π·50rad / s, at this time the damping coefficient ξ=0.707. The Bode diagrams of equations (2.20) and (2.23) are as follows: Figure 4 shown.

[0103] From the figure, we can see that when ω'=ω 0 When v' α and v' β Always in quadrature and with an amplitude gain of 1.

[0104] 2) Orthogonal signal generator based on adaptive filter

[0105] Generally, the function of a filter is to eliminate various harmonics within a selected frequency range, and the coefficients in its transfer function are mostly constants. The biggest difference from other types of filters is that the adaptive filter (AF) can adaptively adjust the coefficients of the filter in real time according to signal changes through an optimization algorithm according to the needs of the system, thereby optimizing the filtering performance of the filter. Adaptive filters have been widely used in the fields of automatic control and information communication, such as interference elimination, adaptive equalization, inverse system identification and image processing.

[0106] The structure of Adaptive Noise Canceller (ANC) is as follows: Figure 5 shown.

[0107] Figure 5 In the above equation, v is the input signal, x is the input reference signal, and v' is the output signal. v is composed of the main signal n and the interference signal n 0 After x passes through AF, it produces the same 0 Similarly, v' is removed from v to generate a signal e without interference. In AF digitization, x undergoes k sampling cycles T s Take N samples and store them in a stack to generate a reference signal set X k =[x k ,x k-1 ,…,x k-N ], reference signal set X k The elements in are multiplied by the corresponding weighting coefficients and the sum is the output signal v'.

[0108] For the design of AF, the least mean square algorithm (LMS) is adopted. s When the reference signal set X is generated k =[x k ,x k-1 ,…,x k-N ] and the corresponding weighted coefficient set W k =[w k ,w k-1 ,…,w k-N ], the specific formula is as follows:

[0109] e k =v k -v' k (2.24);

[0110] w k+1 =w k +ae k x k (2.25);

[0111]

[0112] When N = 1, that is, the AF system uses LMS with only one weighting coefficient, the system structure diagram is as follows: Figure 6 shown.

[0113] Figure 6 In the figure, the virtual frame part and the coefficient are k i The discrete forward Euler integrator S(z) is equivalent to , and a = k i T s The above AF module and the traditional single-phase phase-locked loop module can be combined to form an enhanced phase-locked loop (EPLL). The EPLL structure diagram is as follows: Figure 7 shown.

[0114] In EPLL, as the x=cos(θ') obtained by the voltage-controlled oscillator gradually synchronizes with v, the interference-free signal e obtained by the AF also tends to 0. When the system is stable, the difference between the output phase θ' of EPLL and the phase θ of v is 90°, that is, θ'=θ-π / 2.

[0115] From the above, we can see that EPLL can be used as a single-phase phase-locked loop or as a QSG. The orthogonal signal generator EPLL-QSG based on the enhanced phase-locked loop, such as Figure 8 As shown, the value after integration by the integrator is the amplitude A of the input sinusoidal signal. The signal after the filter module is multiplied by the cosine value and sine value output by the voltage-controlled oscillator to generate a set of orthogonal signals, that is, the output signal v' is orthogonal to the output signal qv', and the amplitude is the same. In a stable state, when the signal e=0, the output signal v' and the input signal v are the same, that is, the phase angles of the two are the same, and the amplitudes of the signals are equal, so v and qv' are also orthogonal, and the amplitudes are equal.

[0116] like Figure 6 As shown, when N=1, that is, when the AF system adopts LMS with only one weighting coefficient, this AF has certain limitations. When the input signal v is a sinusoidal signal, the output error signal e can be zero only when the frequency and phase of the input reference signal x are equal to those of the input signal v, which greatly limits the selection range of the input reference signal x and narrows the scope of use of this AF system.

[0117] like Fig. 9 As shown in FIG. 1 , when N=2, the AF is also called Adaptive Notch Filter (ANF). In the ANF, two sin quadrature signals with a phase difference of 90° are used as reference signals in the LMS. Fig. 9 The system in the figure is a discrete system, which becomes Fig.10 The continuous system in the ANC is optimized based on the standard structure of the ANC.

[0118] like Fig.10 As shown, the input reference signal is the filter frequency ω', and the error signal is ε v , the gain coefficient is k = a / T s , where T s is the sampling period of the original discrete ANF. The following derivation and analysis Fig.10 Transfer function of AF in .

[0119] Let g(t) = kε v (t), error signal ε v Affected by the gain coefficient k, v d and v q It can be expressed as follows:

[0120]

[0121] Assume A d and A q v d and v q The integrated value, after Laplace transformation, can be obtained

[0122]

[0123] Similarly, we can get:

[0124]

[0125] The output signal v'=v' d +v' q ,Right now

[0126]

[0127] In summary, the transfer function of AF in this system is:

[0128]

[0129] The system response can be expressed by the following two second-order transfer functions:

[0130]

[0131] From the above formula, we can get that ABPF is an adaptive low-pass filter and ANF is an adaptive notch filter.

[0132] The adaptive filter AF can extract the selected frequency ω' even if it is affected by interference signals. In addition, adding an integrator after the output signal v' can be used as QSG, such as Fig.11 shown.

[0133] From equation (2.34), we can see that the transfer function of AF has two pure imaginary poles ±jω' that are symmetrical about the horizontal axis. When it is running, its function is similar to that of a resonant oscillator. When a sinusoidal signal with a frequency of ω' is input into AF, the system will act as an amplitude integrator (AI) for the input signal, so the transfer function of AF is also called a sine integrator. Its time domain response is as follows Fig.12 shown.

[0134] The AF time domain response analysis is as follows: When the input frequency is a sinusoidal signal of ω', the Laplace formula of the input signal can be obtained:

[0135]

[0136] Therefore, the time domain response of AF can be obtained by inverse Laplace transform:

[0137]

[0138] From the above two equations, it can be seen that AF plays the role of an ideal AI for sine signals, but not for cosine signals. The result has a certain steady-state error with the ideal value, but this error can be ignored under certain conditions. By multiplying AF(s) by 2, a generalized integrator (GI) can be obtained. When the frequency of GI is at the resonant frequency, its transfer function gain is infinite. Therefore, when the frequency of the input cosine signal is the resonant frequency, the GI controller can eliminate the steady-state error and obtain an ideal output signal, whose frequency is the same as the resonant frequency.

[0139] like Fig.13 As shown, this is a second-order adaptive filter (SOAF) structure that improves the operation speed while avoiding the introduction of interference errors caused by the operation.

[0140] Its transfer function is as follows:

[0141]

[0142] From the above, it can be concluded that from the perspective of the same transfer function, the above two methods have the same functional implementation. Fig.13The phase difference between the output signals v' and qv' of the SOAF shown is 90°, so the two can be used as a set of orthogonal signals, that is, this SOAF can also be used as a QSG. However, it can be seen from equations (2.42) and (2.43) that D is a BPF, Q is an LPF, and the bandwidth of D and the static gain of Q are both related to the gain coefficient k and the center frequency ω'. Therefore, when this GI is used in a frequency-varying system, the bandwidth and static gain are both variable values, which is extremely inconvenient to use. In order to improve the adaptive performance of QSG when the frequency changes, a second-order generalized integrator (SOGI) is used.

[0143] The SOGI transfer function is as follows:

[0144]

[0145] Among them, D(s) is BPF, Q(s) is LPF, and its Bode diagram is as follows Fig.14 As shown: From equations (2.45) and (2.46), it can be obtained that the bandwidth of SOGI-AF has nothing to do with the center frequency ω' and only depends on the gain coefficient k, so this AF is suitable for frequency-varying systems. When the frequency ω of the input signal v is the same as the center frequency ω', the output signals v' and qv' are not only orthogonal, but also have the same amplitude, which is a set of ideal orthogonal signals, so Fig.13 The structure in is very suitable for use as a QSG, and this system is called a second-order generalized integrator-based quadrature signal generator (SOGI-QSG).

[0146] When a sinusoidal signal is input, the time domain response of SOGI-QSG is analyzed as follows:

[0147] When the input signal v = V sin (ωt), and ω = ω', the output signals v' and qv' are obtained from equations (2.45) and (2.46):

[0148]

[0149] in,

[0150] The second-order system can be estimated to have a stable time of t s =4.6τ, in equation (2.45) and equation (2.46), τ = 2 / kω', so the estimated stabilization time of SOGI-QSG is

[0151] Step 2 specifically includes: Step S22, based on the αβ component filtering direct decoupling method, firstly convert the distorted voltage v abc Separated into v in the stationary coordinate system α and vβ Components, and then v α and v β The FPNSC is decoupled to obtain v abc FPSC and FNSC; by using QSG module to achieve v α and v β The PNSC is decoupled, and the fundamental frequency component and its orthogonal quantity are obtained while filtering out the harmonics.

[0152] From formula (2.4), when n = 1,

[0153]

[0154] In short:

[0155]

[0156] Among them, v αp 、v βp is the positive sequence αβ component, v αn 、v βn is the negative sequence αβ component.

[0157]

[0158] From equations (2.53) and (2.54), we can get the positive sequence β component v βp =qv αp With the positive sequence α component v αp The phase difference is 90°, and the negative sequence β component v βn =qv αn With the negative sequence α component v αn The phase difference is 90°, and the former lags behind the latter. Therefore, equation (2.52) can be rewritten as:

[0159]

[0160] In summary, the decoupling of PNSC on the αβ components of the stationary coordinate system can be achieved by using QSG, as Fig.15 As shown above Fig.15 In the decoupling method, the frequency signals used by the two QSGs are the PNSC frequencies themselves. From the above analysis, it can be seen that there is no coupling relationship between the PSC frequency and the NSC frequency, so the above direct decoupling method can be used for the decoupling of PNSCs with different frequencies.

[0161] The following is an analysis of the spectrum response of SOGI-QSG using the direct decoupling method based on the Decoupled DoubleSecond-order Generalized Integrator (DDSOGI) integral:

[0162] For the convenience of analysis, it is assumed that the frequency of PNSC is ω.

[0163] From equations (2.45) and (2.46), the transfer function of SOGI-QSG is:

[0164]

[0165] Depend on Fig.15 Available

[0166]

[0167] Substituting equation (2.59) into equation (2.58) yields

[0168]

[0169] Combining equations (2.60) and (2.61), we can get the transfer function of PSC:

[0170]

[0171] Similarly, we can get:

[0172] When the input frequency of NSC is ω, we have

[0173] The transfer function of NSC is

[0174] The amplitude-frequency characteristic curve of PNSC using the DDSOGI direct decoupling method is shown in the figure below. Fig.16 As shown, equations (2.62) and (2.64) are its transfer functions. Fig.16 It can be obtained that the DDSOGI direct decoupling method acts as a notch filter with a notch frequency of ω' for NSC, a BPF with a center frequency of ω' for PSC, and an LPF for harmonics. Similarly, it can be derived and analyzed that v βp ' and v βn 'To v β process.

[0175] Step 2 specifically includes: Step S23, based on the dq component filtering direct decoupling method, the distorted voltage v abc Decomposed into v in the synchronously rotating coordinate system SRF d and v q, and then directly output after LPF filtering; among them, DDSRF is to set up two SRFs, a positive sequence component synchronous coordinate system PSC-SRF, and a negative sequence component synchronous coordinate system NSC-SRF. The two SRFs are used to eliminate the influence of PSC on NSC measured in NSC-SRF and the influence of NSC on voltage PSC measured in PSC-SRF, thereby realizing the decoupling between PNSCs.

[0176] Only the influence of FNSC is eliminated and the three-phase voltage including FPNSC when voltage distortion occurs is analyzed as shown below.

[0177]

[0178] By combining equation (2.7) and equation (2.65), we can get: and The voltage vector is shown below.

[0179]

[0180] From the above, it can be concluded that FNSC causes FPSC to generate an oscillation with a frequency of -2ω and an oscillation amplitude of the negative sequence DC component V, while FPSC causes FNSC to generate an oscillation with a frequency of 2ω and an oscillation amplitude of the positive sequence DC component V. It can be deduced that in PSC-SRF The AC component on the NSC-SRF can be regarded as The DC component on is obtained by rotating the coordinate system with a rotation frequency of 2ω. Similarly, in NSC-SRF The AC component on the PSC-SRF can be regarded as The DC component on the axis is obtained by rotating the coordinate system with a rotation frequency of -2ω.

[0181] The rotation matrix is ​​as follows:

[0182]

[0183] Combining equations (2.66), (2.67) and (2.68), we can get:

[0184]

[0185]

[0186] in, is the DC component in PSC-SRF, is the DC component in NSC-SRF.

[0187] It can be further deduced that

[0188] Based on the above derivation process, the DDSRF decoupling method is proposed, and its structure is as follows: Fig.17 As shown, where LPF(s) = ω f / s+ω f ,ω f is the cut-off frequency.

[0189] In PSC-SRF The response generated can be Multiply by the rotation matrix with frequency 2ω After conversion, is composed of PSC-SRF minus The response generated in PSC-SRF is filtered out by LPF to remove harmonics and interference; conversely, we can get The above method can realize the decoupling of PNSC under distorted voltage. The LPF in the DDSRF decoupling method can be replaced by LPF of other orders without affecting its working principle.

[0190] The following will take the first-order LPF as an example to analyze the DDSRF decoupling method:

[0191]

[0192] make,

[0193]

[0194] Combining equations (2.72) and (2.73), we can obtain:

[0195]

[0196] Substituting equation (2.70) into equation (2.75) yields:

[0197]

[0198] Coefficient Matrix Simplified, we can get

[0199]

[0200] make,

[0201]

[0202] From this, we can get that formula (2.76) can be simplified to

[0203]

[0204] It can be obtained that the PSC output transfer function of the DDSRF decoupling method is

[0205]

[0206] in,

[0207]

[0208] H 11 =H 22 =H(s 3 +2ω f s 2 +4ω' 2 s+4ω f ω' 2 ) (2.82.a);

[0209] H 12 =-H 21 =-H(2ω f ω's) (2.82.b);

[0210] Similarly, the NSC output transfer function of the DDSRF decoupling method can be obtained as follows:

[0211]

[0212] By transferring the transfer function of the DDSRF decoupling method in SRF to the stationary coordinate system, we can obtain

[0213]

[0214] in,

[0215] H a =H 11 (s+jω')+H 11 (s-jω') (2.85.a);

[0216] H b =H 12 (s+jω')-H 12 (s-jω') (2.85.b);

[0217] H c =H 11 (s+jω')-H 11 (s-jω') (2.85.c);

[0218] H d =H 12 (s+jω')+H 12 (s-jω') (2.85.d);

[0219] Simplified, we can get

[0220]

[0221] Among them, ω f is the cut-off frequency, and ω' is the detected frequency.

[0222] In the DDSRF decoupling method, For example, By v α and v β After eliminating the interference of NSC on PSC, the sum is obtained. From the transfer function, v α The harmonic interference is eliminated by using a BPF with a center frequency of ω'. β The interference of harmonics is eliminated by using LPF with a cutoff frequency of ω'.

[0223] The positive and negative sequence calculation method of step 2 includes: when the voltage is distorted, the symmetry of the three-phase voltage is destroyed, and the three-phase voltage is composed of a group of asymmetric three-phase quantities. The positive and negative sequence calculation method PNSC decomposes the asymmetric three-phase quantities into three groups of different symmetrical three-phase quantities, and then uses the three groups of symmetrical quantities to eliminate the mutual interference between PSC and NSC, thereby separating PSC and NSC.

[0224] From formula (2.1), when the voltage is distorted, the three-phase voltage is expressed as

[0225]

[0226] in,

[0227]

[0228] In formula (2.88), a is a 120° phase shift signal, that is,

[0229] Ignoring the influence of ZSC and converting the three-phase voltage to the αβ stationary coordinate system, we can get

[0230]

[0231] Substituting equation (2.88) into equation (2.89) yields

[0232]

[0233] Then perform the inverse Clarke transformation to obtain

[0234]

[0235] in,

[0236]

[0237] In formula (2.92), q is the 90° phase shift signal, that is, After q operation, the orthogonal signal of the input signal can be obtained.

[0238] From the above, we can get that in the αβ stationary coordinate system, v αβ Separate the positive and negative sequence components v αβ+ and v αβ- , just get v αβ Therefore, the QSG in the above embodiment can be used to decouple the PNSC.

[0239] The following is an example of the QSG positive and negative sequence calculation method based on the inverse Park transform (IPT-QSG-PNSC). Fig.18 and Fig.19 shown.

[0240] The three-phase voltage is first converted into the component v in the αβ stationary coordinate system by Clarke transformation α and v β , and then use IPT-QSG to filter and obtain its orthogonal signal, and finally separate PNSC through the PNSC module to obtain the required voltages PSC and NSC.

[0241] From equation (2.91.a), the transfer function of the PSC obtained by IPT-QSG-PNSC is:

[0242]

[0243] Where D(s) and Q(s) are the transfer functions of IPT-QSG-PNSC, ω f is the cutoff frequency of LPF, ω' is the reference signal frequency, and k = ω f / ω'.

[0244] Similarly, the transfer function of NSC is

[0245]

[0246] From equations (2.86) and (2.93), it can be seen that the DDSRF decoupling method and IPT-QSG-PNSC are two equivalent methods. Both of them realize the same function in different coordinate systems and can be used to separate the voltage PSC and NSC. When k = 2ω f / ω', from the perspective of the same transfer function, the DDSRF decoupling method and the IPT-QSG-PNSC method are identical.

[0247] The complex coefficient filtering method of step 2 includes: in the stationary coordinate system, the α component and the β component are phase-shifted by 90° through the complex symbol j, thereby realizing the imaginary coefficient filtering. The imaginary coefficient filter can accurately and quickly obtain the required component; modify the frequency value ω of the imaginary part 0 and ω c , you can set the filter's passband and bandwidth frequency. When you set ω 0 When the grid voltage frequency ω' is obtained by the phase-locked loop, the system can realize frequency tracking and adaptive filtering of the grid voltage, thereby obtaining various accurate grid information.

[0248] Complex Coefficient Filters (CCF) is a method of using filters with complex coefficients to separate PNSC from voltage. This method does not require synchronous rotating coordinate system conversion or symmetrical component method, and can extract FPSC, FNSC and other harmonic components of the voltage from the distorted voltage. When the voltage is distorted, harmonics of different orders will be added to the voltage. In order to extract the fundamental frequency component in a timely and effective manner, the system usually adds a complex filter to filter out the harmonics.

[0249] The following will analyze the structure of CCF by taking the elimination of FNSC as an example. Fig. 20 This is the CCF function module structure diagram. Fig.21 This is the implementation method of the CCF function module. Fig. 20 and Fig.21 In, v αβ is the input voltage, is the output voltage of n times PSC, is the output voltage times NSC.

[0250] Depend on Fig. 20 and Fig.21 It can be obtained that the α component and the β component in the stationary coordinate system achieve a 90° phase shift through the complex symbol j, thereby realizing the imaginary coefficient filtering. The required components can be obtained accurately and quickly through the imaginary coefficient filter. Modify the frequency value ω of the imaginary part 0 and ω c , you can set the filter's passband and bandwidth frequency. When you set ω 0 When the grid voltage frequency ω' is obtained by the phase-locked loop, the system can realize frequency tracking and adaptive filtering of the grid voltage, thereby obtaining various accurate grid information.

[0251] The following is the derivation process of the CCF transfer function, from Fig. 20 We can get:

[0252]

[0253] When 0=ω', that is, ω 0 When the grid voltage frequency is obtained by the phase-locked loop, Fig.21 The method can be converted as follows:

[0254]

[0255]

[0256] Combining equation (2.97) and equation (2.98) we can get:

[0257]

[0258] Similarly, we can get:

[0259]

[0260] Combining the two equations, we can simplify them to get:

[0261]

[0262] Similarly, we can get:

[0263]

[0264] By comparing equation (2.86) with equation (2.101), it can be obtained that from the perspective of the same transfer function, the DDSRF decoupling method and the CCF method have the same functions in achieving the same results.

[0265] Step 2 specifically includes: Step 24, the MCCF parallel method adopts n CCFs with different filtering frequencies connected in parallel, and the input voltage passes through the CCFs with different filtering frequencies, and the harmonics of the corresponding frequencies and orders are filtered out, and then separated into voltage positive and negative sequence components of the corresponding frequencies and orders. The voltage positive and negative sequence components output by the CCF are fed back to the input voltage, and the input voltage is analyzed, and the harmonics of the relevant frequencies and orders fed back are filtered out, and then the voltage information is input to each CCF again, and the frequency adaptive function of the MCCF parallel module is implemented through frequent feedback and response; by using multiple CCF parallel structures to set different selection frequencies, it is possible to filter out harmonics of different frequencies and orders from the voltage, refine the selection frequency range of the filter module, and thereby improve the filtering effect of the filter module.

[0266] The parallel filter module method will be used to solve the problem of low-order harmonics being difficult to filter out, and the multiple complex coefficient filter (MCCF) parallel method will be proposed using the complex coefficient filter method as an example. The MCCF parallel structure diagram is shown in Fig. 22 shown.

[0267] In summary, although the present invention has been disclosed as above in terms of preferred embodiments, the above preferred embodiments are not intended to limit the present invention. A person skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be based on the scope defined in the claims.

Claims

1. A method for obtaining voltage positive sequence fundamental frequency component under distorted voltage of an island microgrid, characterized in that: The following steps are involved: Step 1: After the microgrid is distorted, its vector does not have a fixed amplitude and a fixed rotation frequency. The phase-locked loop amplifier PLL cannot directly monitor the distorted vector voltage in real time. The phase and frequency information of the positive sequence fundamental frequency component FPSC of the detection voltage is used to replace the vector information of the distorted voltage. Step 2, based on the research on two types of orthogonal signal generators, namely, the orthogonal signal generator based on inverse Park transform and the orthogonal signal generator based on adaptive filter, three positive and negative sequence decoupling methods, namely, direct decoupling method, positive and negative sequence calculation method, and complex coefficient filtering method are used to realize the positive and negative sequence separation of voltage, and then obtain the positive sequence fundamental frequency component FPSC information, and remove the influence of low-order harmonics on the detection voltage frequency and phase function required for the phase-locked loop to achieve locking.

2. The method for obtaining the voltage positive sequence fundamental frequency component under the distorted voltage of an island microgrid according to claim 1 is characterized in that: The three-phase voltage of the microgrid in step 1 is affected by the converter and nonlinear electrical equipment. A variety of harmonic currents of different orders and frequencies will enter the microgrid, which will cause the three-phase grid voltage to be distorted. The three-phase voltage is represented by a collection of multiple unbalanced and different-order harmonic components.

3. The method for obtaining the voltage positive sequence fundamental frequency component under the distorted voltage of an island microgrid according to claim 2 is characterized in that: Step 2 specifically includes: the direct decoupling method in step S2 includes two methods: the first is to extract the αβ AC component of the fundamental frequency in the αβ stationary coordinate system, and the second is to extract the dq DC component in the dq rotating coordinate system; the voltage PNSC separation is achieved by the direct decoupling method and the complex coefficient filtering method. Among the two methods, the orthogonal signal generator module can not only generate orthogonal signals, but also remove harmonics, so the orthogonal signal generator is analyzed first, and then various positive and negative sequence decoupling methods are analyzed.

4. The method for obtaining the voltage positive sequence fundamental frequency component under the distorted voltage of an island microgrid according to claim 3 is characterized in that: Step 2 specifically includes: Step S21, the orthogonal signal generator module adopts the IPT-QSG module and the AF-QSG module, the IPT-QSG module eliminates various harmonics mixed into the input signal through LPF based on the dq DC components, and the AF-QSG module eliminates various harmonics mixed into the input signal through BPF based on the αβ AC components.

5. The method for obtaining the voltage positive sequence fundamental frequency component under the distorted voltage of an island microgrid according to claim 4 is characterized in that: Step 2 specifically includes: Step S22, based on the αβ component filtering direct decoupling method, firstly convert the distorted voltage v abc Separated into v in the stationary coordinate system α and v β Components, and then v α and v β The FPNSC is decoupled to obtain v abc FPSC and FNSC; by using QSG module to achieve v α and v β The PNSC is decoupled, and the fundamental frequency component and its orthogonal quantity are obtained while filtering out the harmonics.

6. The method for obtaining voltage positive sequence fundamental frequency component under distorted voltage of an island microgrid according to claim 4 is characterized in that: Step 2 specifically includes: Step S23, based on the dq component filtering direct decoupling method, the distorted voltage v abc Decomposed into v in the synchronously rotating coordinate system SRF d and v q , and then directly output after LPF filtering; among them, DDSRF is to set up two SRFs, a positive sequence component synchronous coordinate system PSC-SRF, and a negative sequence component synchronous coordinate system NSC-SRF. The two SRFs are used to eliminate the influence of PSC on NSC measured in NSC-SRF and the influence of NSC on voltage PSC measured in PSC-SRF, thereby realizing the decoupling between PNSCs.

7. The method for obtaining voltage positive sequence fundamental frequency component under distorted voltage of an island microgrid according to claim 1 is characterized in that: The positive and negative sequence calculation method of step 2 includes: when the voltage is distorted, the symmetry of the three-phase voltage is destroyed, and the three-phase voltage is composed of a group of asymmetric three-phase quantities. The positive and negative sequence calculation method PNSC decomposes the asymmetric three-phase quantities into three groups of different symmetrical three-phase quantities, and then uses the three groups of symmetrical quantities to eliminate the mutual interference between PSC and NSC, thereby separating PSC and NSC.

8. The method for obtaining voltage positive sequence fundamental frequency component under distorted voltage of an island microgrid according to claim 1, characterized in that: The complex coefficient filtering method of step 2 includes: in the stationary coordinate system, the α component and the β component are phase-shifted by 90° through the complex symbol j, thereby realizing the imaginary coefficient filtering. The imaginary coefficient filter can accurately and quickly obtain the required components; modify the frequency values ​​ω0 and ω c , the passband and bandwidth frequency of the filter can be set. When ω0 is set to the grid voltage frequency ω' obtained by the phase-locked loop, the system can realize frequency tracking and adaptive filtering of the grid voltage, thereby obtaining various accurate grid information.

9. The method for obtaining voltage positive sequence fundamental frequency component under distorted voltage of an island microgrid according to claim 1, characterized in that: Step 2 specifically includes: Step 24, the MCCF parallel method adopts n CCFs with different filtering frequencies connected in parallel, and the input voltage passes through the CCFs with different filtering frequencies, and the harmonics of the corresponding frequencies and orders are filtered out, and then separated into voltage positive and negative sequence components of the corresponding frequencies and orders. The voltage positive and negative sequence components output by the CCF are fed back to the input voltage, and the input voltage is analyzed, and the harmonics of the relevant frequencies and orders fed back are filtered out, and then the voltage information is input to each CCF again, and the frequency adaptive function of the MCCF parallel module is implemented through frequent feedback and response; by using multiple CCF parallel structures to set different selection frequencies, it is possible to filter out harmonics of different frequencies and orders from the voltage, refine the selection frequency range of the filter module, and thereby improve the filtering effect of the filter module.

Images