Power grid frequency estimation method and system based on flexible dynamic frequency-locked loop

Abstract

The invention discloses a power grid frequency estimation method and system based on a flexible dynamic frequency-locked loop, and the method comprises the steps: 1, obtaining a voltage under a dq coordinate system after the voltage of a power grid is subjected to dq conversion, and obtaining an observation value of a dq-axis voltage after the voltage under the dq coordinate system passes through a front low-pass filter; 2, defining an auxiliary variable, and establishing a mathematical relationship between a frequency error and an imaginary part of the auxiliary variable; 3, designing a frequency observation loop, establishing a closed-loop transfer function of observation frequency, and decoupling the prefilter and the frequency observation loop; directly filtering the frequency signal by using an in-loop filter to obtain an observation frequency; 4, the observation frequency is subjected to differential operation and then added with the adjustable initial phase angle to obtain a phase angle generated by FLL, the phase angle generated by FLL is fed back to act on dq conversion of the power grid voltage, and closed-loop control is achieved. According to the method, the calculation amount can be reduced while the filtering performance is improved, and the power grid frequency is estimated.

Description

Technical Field

[0001] The invention relates to the field of power converter control, and in particular to a power grid frequency estimation method based on a flexible dynamic frequency locked loop. Background Art

[0002] As the penetration rate of renewable energy increases, the application scale of grid-connected converters is increasing. In the process of renewable energy access to the grid, it is necessary to accurately estimate the dynamic characteristics of the grid, such as changes in voltage, frequency and phase. The observation of the dynamic characteristics of the grid directly affects the operating stability and response speed of the grid-connected converter.

[0003] Whether it is the frequency phase-locked loop SOGI-FLL or the reduced-order generalized integrator frequency-locked loop ROGI-FLL, they are designed based on the stationary coordinate system. In the stationary coordinate system, the voltage is an AC quantity, which makes the establishment of the small signal model more complicated, and the result of observing the voltage is not as simple and clear as the DC quantity. The frequency-locked loop technology in the synchronous coordinate system, namely the SRF-FLL technology, has established a standard second-order model that can accurately describe the dynamic process of the frequency-locked loop FLL and simplify the parameter design. However, the prefilter parameters in SRF-FLL determine the filtering frequency, and the damping factor is affected by the prefilter parameters and the frequency observation gain at the same time. Once the prefilter parameters are determined, it is necessary to trade off between the damping factor and the adjustment time to select the frequency observation gain. On this basis, the improved PI-SRF-FLL does not need to make a trade-off to select parameters. It only needs to select the frequency observation gain according to the required adjustment time, without considering the damping factor constraint, and select the prefilter parameters according to the filtering requirements. However, it still has certain limitations when facing different dynamic performance requirements, and its ability to suppress harmonics is limited. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in a synchronous coordinate system, how to realize a frequency-locked loop that can decouple the dynamic interaction between the pre-filter and the frequency observation loop, so that the frequency response of the frequency-locked loop can be freely designed according to the requirements of frequency dynamics, thereby improving the accuracy of power grid frequency estimation.

[0005] In order to solve the above technical problems, the present invention provides a method for estimating power grid frequency based on a flexible dynamic frequency locked loop, comprising the following steps:

[0006] Step 1. Input grid voltage u αβ After dq transformation, the voltage u in the dq coordinate system is obtained dq , voltage u in dq coordinate system dq After the pre-low-pass filter, the observed values ​​of the dq axis voltage are obtained ;

[0007] Step 2. Using the observed value of voltage , the error between the observed voltage and the actual voltage of the dq axis dq Define auxiliary variables to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between

[0008] Step 3. Design a frequency observation loop. In the frequency observation loop, use the frequency error ω obtained in step 2 e To the auxiliary variable imaginary part x aI The mathematical relationship between them is used to establish a closed-loop transfer function of the observation frequency. The PI link and the pre-filter are used to offset the coefficients to achieve the decoupling of the pre-filter and the frequency observation loop. The frequency signal is directly filtered using the in-loop filter to obtain the observation frequency. ;

[0009] Step 4. Observation frequency After differential operation, the phase angle generated by FLL is obtained by adding the adjustable initial phase angle θ0 , the phase angle generated by the FLL Feedback acts on the grid voltage u αβ The dq conversion can realize closed-loop control.

[0010] In the above-mentioned method for estimating the frequency of a power grid based on a flexible dynamic frequency-locked loop, in step 1, in a synchronous coordinate system, the dq axis voltage Expressed as

[0011] (1)

[0012] Among them, the observation phase angle , is the FLL observation frequency, is the phase angle error between the observed voltage and the input voltage, is the dq transformation operator, U is the grid voltage amplitude, θ is the grid voltage phase angle, The initial phase angle is adjustable;

[0013] The pre-low-pass filter is expressed by equation (2):

[0014] (2)

[0015] k is the cutoff frequency, where the superscript “^” represents the observed value, is the error between the observed voltage and the actual voltage of the dq axis.

[0016] In the above-mentioned method for estimating the power grid frequency based on a flexible dynamic frequency-locked loop, in step 2, an auxiliary variable is defined to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between

[0017] Define auxiliary variable x a for

[0018] (6)

[0019] where x aR is the real part of the auxiliary variable, x aI is the imaginary part of the auxiliary variable;

[0020] At this time, the auxiliary variable x a Derivative, get the auxiliary variable x a The dynamic equation of

[0021] (7)

[0022] ω e The frequency error ω is the error between the observed frequency and the actual frequency of the power grid. e Divide it into real and imaginary parts and write it in matrix form:

[0023] [ x a x I ] = 1 s 2 + 2 ks + k 2 + ω e 2 [ − ω e s + k ] U 2 ω e (8)

[0024] The dynamic frequency error is modeled as a linear transfer function using equation (8):

[0025] (9)

[0026] is the Laplace operator in the transfer function.

[0027] In the above-mentioned method for estimating power grid frequency based on a flexible dynamic frequency-locked loop, in step 3, the transfer function of the PI link is: , the transfer function of the prefilter is , the PI link is used to realize the decoupling of the pre-filter and the frequency observation loop.

[0028] In the aforementioned power grid frequency estimation method based on a flexible dynamic frequency locking loop, the frequency observation loop is:

[0029] (10)

[0030] (11)

[0031] (12)

[0032] Among them, x aI is the imaginary part of the auxiliary variable defined, Im() means taking the imaginary part, "*" means complex conjugate, d is the frequency observation gain, e d is the d-axis voltage error, e q is the q-axis voltage error, is the d-axis observed voltage, is the q-axis observed voltage, G(s) is a customizable transfer function, x G is the intermediate variable of the filter in the input frequency observation loop; the proportional coefficient in the PI link is taken as 1, and the integral coefficient is also taken as the cutoff frequency k of the pre-filter, so as to cancel the coefficients. is the observation frequency of the flexible dynamic FLL.

[0033] The above-mentioned power grid frequency estimation method based on a flexible dynamic frequency-locked loop converts the observed frequency into the complex frequency domain:

[0034] (13)

[0035] Combining equation (9) we can get the small signal block diagram of the flexible dynamic FLL, as Figure 2 As shown, the closed-loop transfer function of the observation frequency is for

[0036] (14)

[0037] Indicates the grid frequency.

[0038] In the above-mentioned method for estimating the power grid frequency based on a flexible dynamic frequency-locked loop, in step 3, in the process of designing the custom transfer function G(s), the flexible dynamic FLL is first designed as a typical first-order and second-order system, and the expression is as follows:

[0039] (15)

[0040] (16)

[0041] In the formula, is a typical first-order system. is a typical second-order system; T is the adjustment time of a typical first-order system, ζ is the damping ratio of a typical second-order system, ω n is the natural oscillation frequency. Using equation (14), the corresponding custom transfer function G(s) is:

[0042] (17)

[0043] (18)

[0044] The custom transfer function in the frequency observation loop corresponding to a typical first-order system, Customize the transfer function in the frequency observation loop corresponding to a typical second-order system;

[0045] If the custom transfer function G(s) is designed as a constant, the overall frequency response of the flexible dynamic FLL is a first-order system; if the custom transfer function G(s) is designed as a first-order system, the overall frequency response of the flexible dynamic FLL is a second-order system.

[0046] In the above-mentioned method for estimating the frequency of a power grid based on a flexible dynamic frequency-locked loop, in step 3, when designing the custom transfer function G(s), when there are harmonics in the system, the custom transfer function G(s) is designed as a HOLPF. The typical HOLPF is expressed as

[0047] (19)

[0048] The HOLPF consists of n integrators, x i represents the output state of the i-th integrator, e i represents the output x of the i+1th integrator i+1 The difference with x1 is also the input of the i-th integrator, Represents the state quantity x i The derivative of is the coefficient;

[0049] Where i = 1, 2, ..., n, n is the order of HOLPF, and the output variable of the entire HOLPF is The transfer function is

[0050] (20)

[0051] make (twenty one)

[0052] The transfer function of HOLPF is obtained as

[0053] (twenty two)

[0054] The flexible dynamic FLL transfer function at this time is obtained as

[0055] (twenty three)

[0056] make , can be obtained

[0057] (twenty four)

[0058] At this time, the flexible dynamic FLL open-loop transfer function can be regarded as a cascade of an n-order HOLPF and an integral link, with coefficients b1, b2, ..., b nAccording to the design of filtering capability, the frequency observation gain d is adjusted by considering the phase margin, or the flexible dynamic FLL is regarded as a (n+1)-order HOLPF as a whole using equation (24), so that the overall design parameters b1, b2, …, b n and frequency observation gain d.

[0059] The above-mentioned method for estimating the power grid frequency based on a flexible dynamic frequency-locked loop, when n is an even number, the HOLPF can be written as a cascade of n / 2 second-order LPFs,

[0060] (25)

[0061] When n is an odd number, an additional first-order low-pass filter LPF is cascaded so that the HOLPF is calculated according to formula (25).

[0062] A computer system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0063] Beneficial effects achieved by the present invention: The grid frequency estimation method provided by the present invention utilizes a frequency-locked loop (FLL) whose dynamic characteristics can be flexibly designed in a synchronous coordinate system, decouples the dynamic interaction between the pre-filter and the frequency observation loop, and enables the frequency response of the FLL to be freely designed according to the frequency dynamics required. At the same time, unlike the traditional structure, the flexible dynamic FLL can place a high-order filter (HOLPF) in the frequency observation loop to achieve direct filtering of the frequency error signal. The filter only needs to be designed for the frequency signal, which can reduce the amount of calculation while improving the filtering performance, and realize the estimation of the grid frequency, such as fast frequency estimation or frequency protection for active frequency support of photovoltaic stations, etc., which improves the operating stability and response speed of the grid-connected converter. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a block diagram of a flexible dynamic FLL in Embodiment 1 of the present invention;

[0065] Figure 2 is a flexible dynamic FLL small signal block diagram in Embodiment 1 of the present invention;

[0066] Figure 3 is a structural block diagram of a high-order filter in Embodiment 3 of the present invention;

[0067] Figure 4 It is a block diagram of the RoCoF detection method in Example 3 of the present invention. DETAILED DESCRIPTION

[0068] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0069] Example 1

[0070] like Figure 1 As shown, this embodiment provides a method for estimating power grid frequency based on a flexible dynamic frequency locked loop, comprising the following steps:

[0071] Step 1. Input grid voltage u αβ After dq transformation, the voltage u in the dq coordinate system is obtained dq , voltage u in dq coordinate system dq After the pre-low-pass filter, the observed values ​​of the dq axis voltage are obtained ;

[0072] Step 2. Using the observed value of voltage , the error between the observed voltage and the actual voltage of the dq axis dq Define auxiliary variables to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between

[0073] Step 3. Design a frequency observation loop. In the frequency observation loop, use the frequency error ω obtained in step 2 e To the auxiliary variable imaginary part x aI The mathematical relationship between them is used to establish a closed-loop transfer function of the observation frequency. The PI link and the pre-filter are used to offset the coefficients to achieve the decoupling of the pre-filter and the frequency observation loop. The frequency signal is directly filtered using the in-loop filter to obtain the observation frequency. ;

[0074] Step 4. Observation frequency After differential operation, the phase angle generated by FLL is obtained by adding the adjustable initial phase angle θ0 , the phase angle generated by the FLL Feedback acts on the grid voltage u αβ The dq conversion can realize closed-loop control.

[0075] In step 1, in the synchronous coordinate system, the dq axis voltage Expressed as

[0076] (1)

[0077] Among them, the observation phase angle , is the FLL observation frequency, is the phase angle error between the observed voltage and the input voltage, is the dq transformation operator, U is the grid voltage amplitude, θ is the grid voltage phase angle, The initial phase angle is adjustable;

[0078] like Figure 1As shown, the pre-low-pass filter is expressed by equation (2):

[0079] (2)

[0080] k is the cutoff frequency, where the superscript “^” represents the observed value, is the error between the observed voltage and the actual voltage of the dq axis;

[0081] Similar to conventional FLL, due to the lack of closed-loop feedback of the grid phase angle, the phase angle generated dynamically and flexibly There is a static error between the actual grid phase angle , calculated by the following formula

[0082] (3)

[0083] is the d-axis observed voltage, is the q-axis observed voltage;

[0084] At this time, the accurate observation value of the grid phase angle for

[0085] (4)

[0086] Accurate grid voltage observation Expressed as

[0087] (5)

[0088] in is the voltage observation amplitude. From formula (5), we can know that the grid voltage observation value Able to complete the orientation of the grid voltage and always ensure .

[0089] In step 2, define auxiliary variables to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between

[0090] Define auxiliary variable x a for

[0091] (6)

[0092] where x aR is the real part of the auxiliary variable, x aI is the imaginary part of the auxiliary variable;

[0093] At this time, the auxiliary variable x a Derivative, get the auxiliary variable x a The dynamic equation of

[0094] (7)

[0095] ω e The frequency error ω is the error between the observed frequency and the actual frequency of the power grid. e Divide it into real and imaginary parts and write it in matrix form:

[0096] [ x a x I ] = 1 s 2 + 2 ks + k 2 + ω e 2 [ − ω e s + k ] U 2 ω e (8)

[0097] In general, the error ,at this time , the dynamic frequency error is modeled as a linear transfer function by equation (8)

[0098] (9)

[0099] is the Laplace operator in the transfer function;

[0100] Formula (9) shows that the imaginary part of the auxiliary variable x aI Contains the frequency error, so the imaginary part of the auxiliary variable x can be used aI Derived observation frequency The transfer function of .

[0101] In step 3, the transfer function of the PI link is , the transfer function of the prefilter is , the PI link is used to realize the decoupling of the pre-filter and the frequency observation loop;

[0102] Depend on Figure 1 It can be seen that the frequency observation loop is:

[0103] (10)

[0104] (11)

[0105] (12)

[0106] Among them, x aI is the imaginary part of the auxiliary variable defined, Im() means taking the imaginary part, "*" means complex conjugate, d is the frequency observation gain, e d is the d-axis voltage error, e q is the q-axis voltage error, is the d-axis observed voltage, is the q-axis observed voltage, G(s) is a customizable transfer function used to adjust dynamic performance and achieve flexible dynamic customized design, x Gis the intermediate variable of the filter in the input frequency observation loop; the proportional coefficient in the PI link is taken as 1, and the integral coefficient is also taken as the cutoff frequency k of the pre-filter, so as to cancel the coefficients. is the observation frequency of the flexible dynamic FLL.

[0107] Converting the observed frequency into the complex frequency domain gives:

[0108] (13)

[0109] Combining equation (9) we can get the small signal block diagram of the flexible dynamic FLL, as Figure 2 As shown, the closed-loop transfer function of the observation frequency is for

[0110] (14)

[0111] Indicates the grid frequency;

[0112] Formula (14) describes the closed-loop dynamic characteristics of the frequency observation. The dynamic characteristics of the flexible dynamic FLL are determined by the frequency observation gain d and the custom transfer function G(s). The desired dynamic response can be obtained by properly adjusting the custom transfer function G(s) and the frequency observation gain d. It can be obtained that the transfer function of the flexible dynamic FLL will be one order higher than the freely designed transfer function G(s). There is no pre-filter parameter cutoff frequency k in formula (14), which means that the flexible dynamic FLL realizes the decoupling of the pre-filter and the frequency observation loop, so that the two parts can be designed separately.

[0113] In step 4, the estimated frequency is used to calculate the phase angle for the dq conversion in step 1 to achieve closed-loop control.

[0114] Example 2

[0115] In step 3, when designing the custom transfer function G(s), the flexible dynamic FLL is first designed as a typical first-order and second-order system. The expression is as follows:

[0116] (15)

[0117] (16)

[0118] In the formula, is a typical first-order system. is a typical second-order system; T is the adjustment time of a typical first-order system, ζ is the damping ratio of a typical second-order system, ω n is the natural oscillation frequency. Using equation (14), the corresponding custom transfer function G(s) is:

[0119] (17)

[0120] (18)

[0121] The custom transfer function in the frequency observation loop corresponding to a typical first-order system, Customize the transfer function in the frequency observation loop corresponding to a typical second-order system;

[0122] By comparison, if the custom transfer function G(s) is designed as a constant, the overall frequency response of the flexible dynamic FLL can be made a first-order system; if the custom transfer function G(s) is designed as a first-order system, the frequency response can be made a second-order system.

[0123] Example 3

[0124] In step 3, when designing the custom transfer function G(s), when there are harmonics in the system, in order to improve the anti-harmonic disturbance capability of FLL, the custom transfer function G(s) can be designed as HOLPF. Figure 3 is the structural diagram of HOLPF, such as Figure 3 As shown, the typical HOLPF is expressed as

[0125] (19)

[0126] The HOLPF consists of n integrators, x i represents the output state of the i-th integrator, e i represents the output x of the i+1th integrator i+1 The difference with x1 is also the input of the i-th integrator, Represents the state quantity x i The derivative of is the coefficient;

[0127] Where i = 1, 2, ..., n, n is the order of HOLPF, and the output variable of the entire HOLPF is The transfer function is

[0128] (20)

[0129] make

[0130] (twenty one)

[0131] The transfer function of HOLPF is obtained as

[0132] (twenty two)

[0133] Figure 4The flexible dynamic FLL equivalent diagram when the custom transfer function G(s) is designed as HOLPF. From the figure, the flexible dynamic FLL transfer function at this time is obtained as

[0134] (twenty three)

[0135] make , can be obtained

[0136] (twenty four)

[0137] At this time, the flexible dynamic FLL open-loop transfer function can be regarded as a cascade of an n-order HOLPF and an integral link, with coefficients b1, b2, ..., b n According to the design of filtering capability, the frequency observation gain d is adjusted by considering the phase margin, or the flexible dynamic FLL is regarded as a (n+1)-order HOLPF as a whole using equation (24), so that the overall design parameters b1, b2, …, b n and frequency observation gain d.

[0138] Usually when n is an even number, HOLPF can be written as a cascade of n / 2 second-order LPFs.

[0139] (25)

[0140] When n is an odd number, an additional first-order low-pass filter LPF (Low Pass Filter) can be cascaded so that the HOLPF can be calculated according to the above formula. According to the time domain analysis of a typical second-order system, when the damping ratio ζ=0.707, the overshoot of the system is moderate and the adjustment time is short, so it is suitable to fix the damping ratio of all second-order LPFs to 0.707.

[0141] In addition, the input of the flexible dynamic frequency-locked loop in the last integral link is the observed value of the rate of change of frequency RoCoF (Rate of Change of Frequency), such as Figure 4 shown.

[0142] A computer system comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0143] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0144] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0145] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0146] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0147] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for estimating power grid frequency based on a flexible dynamic frequency-locked loop. It is characterized in that The following steps are involved: Step 1. Input grid voltage u αβ After dq transformation, the voltage u in the dq coordinate system is obtained dq , voltage u in dq coordinate system dq After the pre-low-pass filter, the observed values ​​of the dq axis voltage are obtained ; Step 2. Using the observed value of voltage , the error between the observed voltage and the actual voltage of the dq axis dq Define auxiliary variables to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between Step 3. Design a frequency observation loop. In the frequency observation loop, use the frequency error ω obtained in step 2 e To the auxiliary variable imaginary part x aI The mathematical relationship between them is used to establish a closed-loop transfer function of the observation frequency; the PI link and the pre-filter are used to cancel the coefficients to achieve the decoupling of the pre-filter and the frequency observation loop; the frequency signal is directly filtered using the in-loop filter to obtain the observation frequency ; Step 4. Observation frequency After differential operation, the adjustable initial phase angle θ 0 Add to get the phase angle generated by FLL , the phase angle generated by the FLL Feedback acts on the grid voltage u αβ The dq conversion can realize closed-loop control.

2. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 1, It is characterized in that In step 1, in the synchronous coordinate system, the dq axis voltage Expressed as (1) Among them, the observation phase angle , is the FLL observation frequency, is the phase angle error between the observed voltage and the input voltage, is the dq transformation operator, U is the grid voltage amplitude, θ is the grid voltage phase angle, The initial phase angle is adjustable; The pre-low-pass filter is expressed by equation (2): (2) k is the cutoff frequency, where the superscript "^" represents the observed value, is the error between the observed voltage and the actual voltage of the dq axis.

3. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 2, It is characterized in that In step 2, define auxiliary variables to establish the frequency error ω e To the auxiliary variable imaginary part x aI The mathematical relationship between Define auxiliary variable x a for (6) where x aR is the real part of the auxiliary variable, x aI is the imaginary part of the auxiliary variable; For auxiliary variable x a Derivative, get the auxiliary variable x a The dynamic equation of (7) ω e The frequency error ω is the error between the observed frequency and the actual frequency of the power grid. e Divide it into real and imaginary parts and write it in matrix form: (8) The dynamic frequency error is modeled as a linear transfer function using equation (8): (9) is the Laplace operator in the transfer function.

4. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 3, It is characterized in that In step 3, the transfer function of the PI link is , the transfer function of the prefilter is , the PI link is used to realize the decoupling of the pre-filter and the frequency observation loop.

5. The method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 3, It is characterized in that The frequency observation loop is: (10) (11) (12) Among them, x aI is the imaginary part of the auxiliary variable defined, Im() means taking the imaginary part, "*" means complex conjugate, d is the frequency observation gain, e d is the d-axis voltage error, e q is the q-axis voltage error, is the d-axis observed voltage, is the q-axis observed voltage, G(s) is a customizable transfer function, x G is the intermediate variable of the filter in the input frequency observation loop; the proportional coefficient in the PI link is taken as 1, and the integral coefficient is also taken as the cutoff frequency k of the pre-filter, so as to cancel the coefficients. is the observation frequency of the flexible dynamic FLL.

6. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 5, It is characterized in that Converting the observed frequency into the complex frequency domain gives: (13) Combining equation (9) to obtain the small signal block diagram of the flexible dynamic FLL, as shown in Figure 2, the closed-loop transfer function of the observation frequency is for (14) Indicates the grid frequency.

7. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 6, It is characterized in that In step 3, when designing the custom transfer function G(s), the flexible dynamic FLL is first designed as a typical first-order and second-order system, as shown in the following expression: (15) (16) In the formula, is a typical first-order system. is a typical second-order system; T is the adjustment time of a typical first-order system, ζ is the damping ratio of a typical second-order system, ω n is the natural oscillation frequency. Using equation (14), the corresponding custom transfer function G(s) is: (17) (18) The custom transfer function in the frequency observation loop corresponding to a typical first-order system, Customize the transfer function in the frequency observation loop corresponding to a typical second-order system; If the custom transfer function G(s) is designed as a constant, the overall frequency response of the flexible dynamic FLL is a first-order system; if the custom transfer function G(s) is designed as a first-order system, the overall frequency response of the flexible dynamic FLL is a second-order system.

8. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 7, It is characterized in that In step 3, when designing the custom transfer function G(s), when there are harmonics in the system, the custom transfer function G(s) is designed as HOLPF, which is expressed as (19) The HOLPF consists of n integrators, x i represents the output state quantity of the i-th integrator, e i represents the output x of the i+1th integrator i+1 With x 1 The difference between , which is also the input of the i-th integrator, Represents the state quantity x i The derivative of is the coefficient; Where i = 1, 2, ..., n, n is the order of HOLPF, and the output variable of the entire HOLPF is The transfer function is (20) make (twenty one) The transfer function of HOLPF is obtained as (22) The flexible dynamic FLL transfer function at this time is obtained as (23) make , can be obtained (24) At this time, the flexible dynamic FLL open-loop transfer function can be regarded as a cascade of an n-order HOLPF and an integral link, and the coefficient b 1 , b 2 ,… ,b n According to the design of filtering capability, the frequency observation gain d is adjusted by considering the phase margin, or the flexible dynamic FLL is regarded as a (n+1)-order HOLPF as a whole using equation (24), so that the overall design parameter b 1 , b 2 , …, b n and frequency observation gain d.

9. A method for estimating power grid frequency based on a flexible dynamic frequency locked loop according to claim 8, It is characterized in that When n is an even number, HOLPF can be written as a cascade of n / 2 second-order LPFs. (25) When n is an odd number, an additional first-order low-pass filter LPF is cascaded so that the HOLPF is calculated according to formula (25).

10. A computer system, It is characterized in that The method comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 9.

Images