A control parameter identification method for new energy units based on finite oscillation data

Abstract

The present application provides a control parameter identification method for a new energy generator set based on finite oscillation data, comprising obtaining n sets of actual subsynchronous / supersynchronous oscillation measurement data, extracting key features of each oscillation mode; establishing an equivalent RLC circuit of a new energy generator set grid-connected system, calculating the impedance value of the new energy generator set at n oscillation frequencies; establishing a theoretical impedance model of the new energy generator set, and respectively dividing the n oscillation frequencies f os Substituting the impedance of the new energy generator set with the main circuit parameters into the equations, we obtain n impedance expressions for the new energy generator set containing the control parameters to be identified. We perform a participation analysis based on the dominant oscillation mode to determine the number m of control parameters to be identified. We then use the impedances calculated based on the equivalent RLC circuit and the theoretical impedance model to construct a set of n simultaneous equations. By comparing the number n of equations with the number m of control parameters to be identified, we solve for the control parameters to be identified. This application uses finite oscillation measurement data to accurately identify the control parameters of new energy generator sets, offering ease of operation and high practicality.

Description

Technical Field

[0001] The present application relates to the field of renewable energy power generation technology, and in particular to a method for identifying control parameters of a renewable energy unit based on finite oscillation data. Background Art

[0002] Subsynchronous and supersynchronous oscillations are a common problem in my country's large-scale renewable energy systems, seriously impacting the safe and stable operation of the power grid. As renewable energy installed capacity continues to increase, more regions will face the risk of subsynchronous and supersynchronous oscillations. However, for a large number of older or imported renewable energy units in renewable energy bases, key control parameters and impedance characteristics of their "gray box" models are difficult to obtain due to commercial confidentiality and discontinued models, making traditional modeling methods inapplicable.

[0003] To accurately assess the system's oscillation risk, it's imperative to obtain accurate models for analysis and research. Existing literature often uses time-domain response or frequency-domain data methods to identify parameters in "gray-box" models. However, these methods have limitations: time-domain response methods can only identify a single control link at a time and require the injection of specific disturbances, making identification more complex; frequency-domain data methods require precise broadband impedance measurements, but impedance measurement accuracy is difficult to guarantee in the field. Existing research lacks direct use of measured oscillation data for "gray-box" parameter identification. Summary of the Invention

[0004] In view of the defects in the prior art, the purpose of this application is to provide a new energy unit control parameter identification method based on finite oscillation data to accurately identify the control parameters of the new energy unit.

[0005] In one aspect of the present application, a method for identifying control parameters of a new energy generator set based on finite oscillation data is provided, the method comprising:

[0006] Obtain n sets of actual subsynchronous / supersynchronous oscillation measurement data and extract the key features of each oscillation mode;

[0007] Based on the key characteristics of the oscillation mode, the equivalent RLC circuit of the new energy unit grid-connected system is established, and the impedance value of the new energy unit at n oscillation frequencies is calculated;

[0008] According to the known control structure, the theoretical impedance model of the new energy unit is established, and the n oscillation frequencies f os and main circuit parameters, and obtain the impedance expressions of n new energy units containing the control parameters to be identified;

[0009] Perform participation analysis based on the dominant oscillation mode to determine the number m of control parameters to be identified;

[0010] The impedance calculated based on the equivalent RLC circuit and the theoretical impedance model is combined into n equations. The control parameters to be identified are solved by judging the size between the number of equations n and the number of control parameters to be identified m.

[0011] Furthermore, the measurement data includes the voltage and current time domain waveforms of the ports of the new energy generator set, and the oscillation frequencies of the oscillation components of each group are different;

[0012] The key features of the oscillation mode include: oscillation voltage and current phasors, oscillation frequency and attenuation factor.

[0013] Furthermore, n sets of actual subsynchronous / supersynchronous oscillation measurement data are obtained to extract the key features of each oscillation mode, including:

[0014] Obtain measurement data of n groups of sub- / super-synchronous oscillations;

[0015] Through the Hilbert transform method based on variational mode decomposition (VMD), VMD decomposes the time domain data into a series of intrinsic mode functions (IMFs) with finite bandwidth, and then fits the IMF components to obtain the oscillating voltage and current phasors, oscillation frequency and attenuation factor.

[0016] Furthermore, based on the key characteristics of the oscillation mode, an equivalent RLC circuit of the grid-connected system of the new energy unit is established, and the impedance value of the new energy unit at n oscillation frequencies is calculated, specifically including:

[0017] From n sets of oscillation measurement data, obtain the key features of n sets of dominant oscillation modes and establish the second-order equivalent RLC circuit of the system at n oscillation frequencies;

[0018] According to the obtained oscillation measurement data and the key characteristics of the oscillation mode, based on the expression of the equivalent RLC second-order circuit, the impedance value Zg of the new energy unit at n oscillation frequencies is calculated.

[0019] Furthermore, the expression based on the equivalent RLC second-order circuit is:

[0020]

[0021] Where: Z g is the complex domain expression of the equivalent impedance of the grid-side new energy unit; R g 、R s are the equivalent resistances of the grid side and the new energy unit respectively; L g 、C g 、X g are the grid-side equivalent inductance, capacitance and reactance respectively; Ls 、C s is the equivalent inductance and capacitance of the new energy unit; U s , I s and are the voltage, current amplitude and phase angle of the new energy unit side in the oscillation data; f os is the oscillation frequency; j is the complex imaginary unit.

[0022] Furthermore, the RLC second-order circuit includes:

[0023] Regarding the sub- / super-synchronous oscillation phenomenon generated by direct-drive wind turbines connected to a weak grid, the grid side exhibits a resistance-inductance characteristic, while the direct-drive wind turbine exhibits a resistance-capacitance characteristic, forming an RLC resonant circuit at the super-synchronous resonant frequency.

[0024] Regarding the subsynchronous resonance phenomenon generated by the series compensation transmission system of the doubly fed wind turbine, the grid side presents resistance-capacitance characteristics, and the doubly fed wind turbine presents resistance-inductance characteristics, forming an RLC resonant circuit at the subsynchronous resonant frequency.

[0025] Furthermore, based on the known control structure, the theoretical impedance model of the new energy unit is established, and the n oscillation frequencies f os Substituting the main circuit parameters into the equation, we can obtain the impedance expressions of n new energy units containing the control parameters to be identified, including:

[0026] Obtain the control structure of the new energy unit and establish the AC port positive sequence impedance model Z of the new energy unit th (jω), ω is the angular frequency;

[0027] Among them, Z th The derivation process of (jω) is as follows:

[0028] Establish the impedance model Z of the AC port of the new energy unit in the dq coordinate system dq (s), through the relationship between dq impedance and sequence impedance, the positive sequence impedance model Z of the AC port of the new energy unit is obtained. th (jω), the calculation process is as follows:

[0029]

[0030] where Z pn (s) is the two-dimensional improved sequence impedance of the AC port of the new energy unit; Z pp (s), Z pn (s), Z np (s), Z nn (s) are Z pn (s); ω0 is the power frequency; s represents the Laplace domain.

[0031] Substitute n oscillation frequencies and main circuit parameters into the AC port positive sequence impedance model Z th (jω), we can get the impedance expressions of n new energy units containing the control parameters to be identified.

[0032] Furthermore, based on the dominant oscillation mode, participation analysis is performed to determine the number m of control parameters to be identified, specifically including:

[0033] The participation factor characterizes the contribution of the state variable to the mode, based on which participation analysis is performed;

[0034] The participation expression is as follows:

[0035]

[0036] Where: k w Characterizes the participation of the i-th mode in the participation factor with respect to the k-th state variable in all state variables; u ki 、v ik are the corresponding values ​​of the i-th mode in the left and right eigenvectors of the state matrix A with respect to the k-th state variable; N is the total number of oscillation modes;

[0037] By solving the participation of each state variable under the dominant oscillation mode, the state variable with the largest participation is recorded as the dominant state variable under the dominant oscillation mode. The corresponding control link contains m control parameters to be identified; the remaining control parameters select typical values.

[0038] Furthermore, the impedance calculated based on the equivalent RLC circuit and the theoretical impedance model is used to simultaneously establish n equations. By judging the size between the number of equations n and the number of control parameters to be identified m, the control parameters to be identified are solved, specifically including:

[0039] If the number of equations n is greater than or equal to the number of parameters to be identified m, it is an overdetermined system of equations. You can choose any m linearly independent equations from the n equations and solve them to get the unique solution of the control parameters to be identified.

[0040] If the number of equations n is less than the number of parameters to be identified m, it is an indeterminate system of equations and constraints need to be added to solve the equations.

[0041] Furthermore, the constraint condition is: supplementing typical values ​​of mn non-dominant parameters and solving n control parameters to be identified by setting up n equations in parallel;

[0042] The division of dominant and non-dominant control parameters is determined by parameter sensitivity, which is defined as follows:

[0043]

[0044] Where:i is the i-th eigenvalue; u i 、v i are the left and right eigenvectors corresponding to the eigenvalue respectively; α represents a parameter.

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

[0046] This application provides a practical method for obtaining key control parameters in the "gray box" model of a new energy unit. This method is based on a new energy unit control parameter identification method using finite oscillation data, and can accurately identify the control parameters of the new energy unit. It has a simple mechanism, intuitive calculation, and the advantages of easy operation and strong practicality. It can extract key morphological features only through the unit port oscillation data, and then realize the identification of the dominant control parameters, which is easy to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0048] Figure 1 This is a flow chart of a method for identifying control parameters of a new energy generator set based on finite oscillation data according to an embodiment of the present application.

[0049] Figure 2 This is a detailed flow chart of a method for identifying control parameters of a new energy generator set based on finite oscillation data according to an embodiment of the present application.

[0050] Figure 3 Schematic diagram of a first- and second-order equivalent RLC circuit at the oscillation frequency of an embodiment of the present application.

[0051] Figure 4 FIG. 1 is another schematic diagram of a second-order equivalent RLC circuit at the oscillation frequency according to an embodiment of the present application.

[0052] Figure 5 This is a typical topology of a doubly-fed wind turbine generator system with series compensation in a preferred embodiment of the present application. DETAILED DESCRIPTION

[0053] The present application is described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but are not intended to limit the present application in any form. It should be noted that those skilled in the art may make several variations and improvements without departing from the scope of the present application. These all fall within the scope of protection of the present application.

[0054] This application provides a new energy unit control parameter identification method based on finite oscillation data, which realizes accurate identification of new energy unit control parameters through finite oscillation measurement data, and has the advantages of simple operation and strong practicality.

[0055] Reference Figure 1 As shown in FIG. 1 , a method for identifying control parameters of a new energy generator set based on finite oscillation data is provided in accordance with an embodiment of the present application. The method includes:

[0056] S1. Obtain n sets of actual subsynchronous / supersynchronous oscillation measurement data and extract key features of each oscillation mode.

[0057] S2. Based on the key characteristics of the oscillation mode, an equivalent RLC circuit of the grid-connected system of the new energy unit is established, and the impedance value of the new energy unit at n oscillation frequencies is calculated.

[0058] S3. According to the known control structure, the theoretical impedance model of the new energy unit is established, and n oscillation frequencies f are respectively os Substituting the main circuit parameters into the equation, we can obtain the impedance expressions of n new energy units containing the control parameters to be identified.

[0059] S4. Perform participation analysis based on the dominant oscillation mode to determine the number m of control parameters to be identified.

[0060] S5. The impedance calculated based on the equivalent RLC circuit and the theoretical impedance model is combined into n sets of equations. The control parameters to be identified are solved by judging the size between the number of equations n and the number of control parameters to be identified m.

[0061] This application can effectively predict and suppress sub / supersynchronous oscillations and improve the overall stability of the system by accurately identifying the impedance characteristics and control parameters of new energy units at specific oscillation frequencies; based on the identified control parameters, the control strategy of the new energy units can be adjusted and optimized in a targeted manner to adapt to the dynamic changes of the power grid and improve the operating efficiency and reliability of the units.

[0062] Specifically, n groups of sub / supersynchronous oscillation measurement data are obtained from actual operation, and these data are preprocessed to extract the key features of each oscillation mode, such as oscillation frequency, amplitude, phase, etc.; then, based on the extracted key features of the oscillation mode, an equivalent RLC circuit model of the new energy unit grid-connected system is established, and then the model is used to calculate the impedance value of the new energy unit at n oscillation frequencies, and according to the known new energy unit control structure, its theoretical impedance model is established; the n oscillation frequencies and main circuit parameters are substituted into the theoretical impedance model to obtain n new energy unit impedance expressions containing the control parameters to be identified; participation analysis is performed based on the dominant oscillation mode to determine the number m of control parameters to be identified; the impedance value calculated based on the equivalent RLC circuit is combined with the impedance expression in the theoretical impedance model to form n equation groups; finally, the relationship between the number n of the equation group and the number m of control parameters to be identified is judged, and the control parameters to be identified are obtained by solving the equation group or optimizing the algorithm.

[0063] In the above embodiment, the new energy generator set is a "gray box" model, that is, the control structure is known but the control parameters are unknown. Specific objects of the new energy generator set include: permanent magnet direct-drive wind turbine set, doubly-fed wind turbine set, and photovoltaic power generation unit.

[0064] This application provides a practical method for obtaining key control parameters in the "gray box" model of a new energy unit. This method can accurately identify the control parameters of the new energy unit only through limited oscillation measurement data, and has the advantages of simple operation and strong practicality.

[0065] In the above embodiment, in S1, the measurement data includes the voltage and current time domain waveforms of the new energy unit port, and the oscillation frequencies of each group of oscillation components are different; the key features of the oscillation mode include: oscillation voltage and current phasors, oscillation frequency and attenuation factor.

[0066] Furthermore, in S1, it specifically includes:

[0067] Obtaining measurement data of n groups of sub- / super-synchronous oscillations;

[0068] Through the Hilbert transform method based on variational mode decomposition (VMD), VMD decomposes the time domain data into a series of intrinsic mode functions (IMFs) with finite bandwidth, and then fits the IMF components to obtain the oscillating voltage and current phasors, oscillation frequency and attenuation factor.

[0069] Among them, VMD is an adaptive, non-recursive signal processing method that can effectively process nonlinear and non-stationary signals and is suitable for oscillation analysis in complex power systems.

[0070] Specifically, time-domain waveform data of voltage and current are collected from sensors at the ports of new energy units (such as wind turbines and solar inverters), and the preprocessed time-domain data are decomposed into a series of intrinsic mode functions (IMFs) with finite bandwidth by applying the variational mode decomposition (VMD) method; each IMF component is subjected to Hilbert transform to obtain its instantaneous frequency, amplitude and other information; finally, key characteristic parameters such as oscillating voltage and current phasors, oscillation frequency and attenuation factor are extracted from the IMF components through data fitting or parameter estimation methods.

[0071] like Figure 3 As shown in the figure, in this specific application example, the new energy unit adopts a 2MW doubly-fed wind turbine unit, and the research scenario is a doubly-fed wind turbine unit connected to the grid through series compensation, with a series compensation degree of 8.46%.

[0072] When the output reaches 2MW, key oscillation features are extracted from the recorded oscillating voltage and current data. The oscillation modal key feature extraction method is based on the Hilbert transform of variational mode decomposition (VMD). VMD decomposes the time domain data into a series of intrinsic mode functions (IMFs) with finite bandwidth. The IMF components (corresponding to each oscillation mode) are then fitted to obtain the oscillating voltage and current phasors, oscillation frequency, and attenuation factor. Two sets of oscillation data were recorded. Data one corresponds to an oscillating current of 8091.3∠0.6390°A, an oscillating voltage of 42.1∠0.3556°V, an oscillating frequency of 6.57Hz, and an attenuation factor of 2.453; data two corresponds to an oscillating current of 8455.6∠0.5505°A, an oscillating voltage of 44.3∠0.2451°V, an oscillating frequency of 6.43Hz, and an attenuation factor of 1.419.

[0073] This application provides stability analysis information for the system by extracting key characteristic parameters such as the oscillating voltage and current phasors, oscillation frequency and attenuation factor. The VMD method can adaptively decompose complex signals into multiple clear IMF components, effectively distinguish oscillation components of different frequencies, and improve the accuracy of oscillation detection. Compared with traditional methods such as Fourier transform or wavelet transform, VMD has more advantages in processing nonlinear and non-stationary signals, and can better adapt to complex oscillation phenomena in power systems.

[0074] In some possible embodiments, S2 specifically includes:

[0075] From n sets of oscillation measurement data, the key features of n sets of dominant oscillation modes are obtained, and the second-order equivalent RLC circuits of the system at n oscillation frequencies are established.

[0076] According to the obtained oscillation measurement data and the key characteristics of the oscillation mode, based on the expression of the equivalent RLC second-order circuit, the impedance value Zs of the new energy unit at n oscillation frequencies is calculated.

[0077] Among them, the RLC second-order circuit includes:

[0078] Reference Figure 3 As shown in the figure, for the sub / super synchronous oscillation phenomenon generated by the direct-drive wind turbine generator system connected to the weak grid, the grid side presents a resistance-inductance characteristic, and the direct-drive wind turbine generator system presents a resistance-capacitance characteristic, forming an RLC resonant circuit at the super synchronous resonant frequency.

[0079] Specifically, it includes the first new energy unit side and the first grid side, wherein the direct-drive wind turbine Δu on the first new energy unit side os One end is grounded, and the other end is connected to one end of the resistor Rs, the other end of the resistor Rs is connected to one end of the capacitor Cs, the other end of the capacitor Cs is connected to one end of the resistor Rg on the grid side, the other end of the resistor Rg is connected to the inductor Lg, and the other end of the inductor Lg is grounded.

[0080] Among them, Z s The complex domain expression for the grid impedance; Δi s represents the current in the dominant oscillation mode; Δu s Indicates the grid connection point voltage under the dominant oscillation mode; Δu os represents the equivalent voltage source of the dominant oscillation mode;

[0081] Reference Figure 4 As shown in the figure, for the subsynchronous resonance phenomenon generated by the series compensation transmission system of the doubly fed wind turbine, the grid side presents a resistance-capacitance characteristic, and the doubly fed wind turbine presents a resistance-inductance characteristic, forming an RLC resonant circuit at the subsynchronous resonance frequency.

[0082] Specifically, it includes a second energy unit side and a second power grid side, wherein one end of the doubly fed wind turbine on the second energy unit side is grounded, and the other end is connected to one end of the resistor Rs, the other end of the resistor Rs is connected to one end of the inductor Ls, the other end of the inductor Ls is connected to one end of the resistor Rg, the other end of the resistor Rg is connected to one end of the capacitor Cg, and the other end of the capacitor Cg is grounded.

[0083] In the above embodiment, the expression based on the equivalent RLC second-order circuit is:

[0084]

[0085] Where: Z g is the complex domain expression of the equivalent impedance of the grid-side new energy unit; R g 、R s are the equivalent resistances of the grid side and the new energy unit respectively; Lg 、C g 、X g are the grid-side equivalent inductance, capacitance and reactance respectively; L s 、C s is the equivalent inductance and capacitance of the new energy unit; U s , I s and are the voltage, current amplitude and phase angle of the new energy unit side in the oscillation data; f os is the oscillation frequency; j is the complex imaginary unit.

[0086] In some possible embodiments, S3 specifically includes:

[0087] Obtain the control structure of the new energy unit and establish the AC port positive sequence impedance model Z of the new energy unit th (jω), ω is the angular frequency.

[0088] Z th The derivation process of (jω) is as follows:

[0089] First, establish the impedance model Z of the AC port of the new energy unit in the dq coordinate system dq (s), and then through the relationship between dq impedance and sequence impedance, the positive sequence impedance model Z of the AC port of the new energy unit is further obtained. th (jω), the calculation process is as follows:

[0090]

[0091] where Z pn (s) is the two-dimensional improved sequence impedance of the AC port of the new energy unit; Z pp (s), Z pn (s), Z np (s), Z nn (s) are Z pn (s); ω0 is the power frequency; s represents the Laplace domain.

[0092] Substitute n oscillation frequencies and main circuit parameters into the AC port positive sequence impedance model Z th (jω), we can get the impedance expressions of n new energy units containing the control parameters to be identified.

[0093] In some possible embodiments, S4 specifically includes:

[0094] The participation factor characterizes the contribution of the state variable to the mode, based on which participation analysis is performed. The participation expression is as follows:

[0095]

[0096] Where: k w Characterizes the participation of the i-th mode in the participation factor with respect to the k-th state variable in all state variables; u ki 、v ik are the corresponding values ​​of the i-th mode in the left and right eigenvectors of the state matrix A with respect to the k-th state variable; N is the total number of oscillation modes.

[0097] By solving the participation of each state variable under the dominant oscillation mode, the state variable with the largest participation is recorded as the dominant state variable under the dominant oscillation mode. The corresponding control link contains m control parameters to be identified; the remaining control parameters select typical values.

[0098] In some possible embodiments, S5 specifically includes:

[0099] S51. If the number of equation groups n is greater than or equal to the number of parameters to be identified m, it is an overdetermined equation group. You can select any m linearly independent equation groups from the n equation groups and solve the equation groups to obtain the unique solution of the control parameters to be identified.

[0100] S52. If the number n of the equations is less than the number m of parameters to be identified, it is an indeterminate equation system and constraints need to be added to solve the equations.

[0101] In the above embodiment, the constraints are: supplementing the typical values ​​of mn non-dominant parameters and solving n equations in parallel to obtain n control parameters to be identified;

[0102] The division of dominant and non-dominant control parameters is determined by parameter sensitivity, which is defined as follows:

[0103]

[0104] Where: i is the i-th eigenvalue; u i 、v i are the left and right eigenvectors corresponding to the eigenvalue respectively; α represents a parameter.

[0105] The greater the parameter sensitivity, the more significant the impact of its change on system stability. Therefore, the control parameter with a significantly higher sensitivity than other variables is the dominant parameter.

[0106] like Figure 5 As shown, it is a typical topology of a doubly-fed wind turbine generator system with series compensation. Combined with a specific application example, the technical solution provided by this application is further described in detail as follows.

[0107] Specifically, it includes: one end of the wind turbine is connected to the transmission system, the other end of the transmission system is connected to the doubly fed generator, a capacitor Cw is connected in parallel between the machine-side converter and the grid-side converter, the machine-side converter is connected to the machine-side converter control, and the machine-side converter control includes a power outer loop and a current inner loop; the grid-side converter is connected to the grid-side converter control, and the grid-side converter control includes a current outer loop and a voltage outer loop, the output of the grid-side converter is connected to one end of the inductor Lw, one output of the doubly fed generator is connected to the machine-side converter, and the other output is connected to the other end of the inductor Lw, the other end of the inductor Lw is connected to the input end of the transmission line composed of inductors, resistors, and capacitors, and the output end of the transmission line is linked to the AC power grid.

[0108] In this specific application example, the AC grid voltage is 690V / 50Hz, and the new energy station passes through the line impedance Z L and series compensation X C Then connected to the grid.

[0109] like Figure 5 As shown in the figure, in this specific application example, the new energy unit adopts a 2MW doubly-fed wind turbine unit, and the research scenario is a doubly-fed wind turbine unit connected to the grid through series compensation, with a series compensation degree of 8.46%.

[0110] When the output reaches 2MW, key oscillation features are extracted from the recorded oscillating voltage and current data. The oscillation modal key feature extraction method is based on the Hilbert transform of variational mode decomposition (VMD). VMD decomposes the time domain data into a series of intrinsic mode functions (IMFs) with finite bandwidth. The IMF components (corresponding to each oscillation mode) are then fitted to obtain the oscillating voltage and current phasors, oscillation frequency, and attenuation factor. Two sets of oscillation data were recorded. Data one corresponds to an oscillating current of 8091.3∠0.6390°A, an oscillating voltage of 42.1∠0.3556°V, an oscillating frequency of 6.57Hz, and an attenuation factor of 2.453; data two corresponds to an oscillating current of 8455.6∠0.5505°A, an oscillating voltage of 44.3∠0.2451°V, an oscillating frequency of 6.43Hz, and an attenuation factor of 1.419.

[0111] according to Figure 4 As shown, the grid side presents resistance-capacitance characteristics, and the doubly fed wind turbine presents resistance-inductance characteristics, forming an RLC resonant circuit at the subsynchronous resonant frequency. Based on the equivalent RLC second-order circuit, the following expression is obtained:

[0112]

[0113] Where: Z gis the complex domain expression of the equivalent impedance of the grid-side new energy unit; R g 、R s are the equivalent resistances of the grid side and the new energy unit respectively; L g 、C g 、X g are the grid-side equivalent inductance, capacitance and reactance respectively; L s 、C s is the equivalent inductance and capacitance of the new energy unit; U s , I s and are the voltage, current amplitude and phase angle of the new energy unit side in the oscillation data; f os is the oscillation frequency; j is the complex imaginary unit.

[0114] From this, we can calculate that for data 1: R g =0.005Ω, X g =-0.0015Ω, R s =-0.0048Ω, X g =0.0015Ω, the AC port impedance at the oscillation frequency of the doubly fed wind turbine is Z g =(-0.0048+j 0.0015)Ω; For data 2: R g =0.00499Ω, X g =-0.0016Ω, R s =-0.0049Ω, X g =0.0016Ω, the AC port impedance at the oscillation frequency of the doubly fed wind turbine is Z g =(-0.0049+j 0.0016)Ω.

[0115] Obtain the control structure of the new energy unit and establish the AC port positive sequence impedance model Z of the new energy unit th (jw), the oscillation frequency f os After substituting the main circuit parameters, an impedance expression of the new energy unit containing the control parameters to be identified can be obtained.

[0116] The participation analysis of the series-compensated grid-connected system of the doubly-fed wind turbine generator system shows that the participation of the rotor-side current inner loop is 48.2%, far exceeding the contribution of other control links to the dominant mode. Therefore, the control parameters corresponding to the rotor-side current inner loop are selected: the proportional coefficient Kp and the integral coefficient Ki as the two control parameters to be identified. The remaining control parameters are selected by general design methods to obtain the typical values ​​of the control parameters. The general design method of the doubly-fed wind turbine generator system can be obtained according to the literature [1] Deng Fujin. Design and implementation of dual PWM converter for doubly-fed wind turbine generator [D]. Shanghai Jiaotong University, 2008.

[0117] When data 1 and data 2 are used as the oscillation data sources, the number of equation groups is equal to the number of parameters to be identified. At this time, solving the equation group can obtain the unique solution of the control parameters to be identified.

[0118] In this embodiment, the theoretical impedance Z th (jω) and the Z calculated above g The two sets of equations are combined, and Kp is obtained by solving the equation (Zg=Zth). p =0.0212, K i =3.9744, actual K p =0.02, K i =4, high recognition accuracy.

[0119] When only Data 1 is used as the oscillation data source, the number of equations is less than or equal to the number of parameters to be identified, resulting in an indeterminate system of equations. Constraints must be added to solve the equations. This requires adding a typical value for a non-dominant control parameter to solve for the dominant control parameter to be identified.

[0120] The division of dominant and non-dominant control parameters is determined by parameter sensitivity, which is defined as follows:

[0121]

[0122] Where: i is the i-th eigenvalue; u i 、v i are the left and right eigenvectors corresponding to the eigenvalue respectively; α represents a parameter.

[0123] The larger the absolute value of the real part of the parameter sensitivity, the more significant the impact of its change on system stability. Therefore, the control parameter whose absolute value of the real part of the parameter sensitivity is significantly higher than that of other variables is the dominant parameter.

[0124] In this embodiment, the inner ring proportional coefficient K of the q-axis rotor side current is obtained. p is the dominant control parameter, and its parameter sensitivity is 2.5595±j 0.1708; and the integral coefficient K i The parameter sensitivity is only 0.0309 ± j 0.0782, which has little effect on the dominant mode. Therefore, in this embodiment, K i Set it as the non-dominant control parameter. Substitute the theoretical impedance Z of the main circuit parameters and the typical values ​​of the non-dominant control parameters into th (jω) and the Z calculated above g Combined, solve for K p =0.0204, actual K p =0.02, high recognition accuracy.

[0125] This application realizes accurate identification of control parameters of new energy units only through limited oscillation measurement data, and has the advantages of simple operation and strong practicality.

[0126] The above describes the specific embodiments of the present application. It should be understood that the present application is not limited to the specific embodiments described above, and those skilled in the art may make various modifications or variations within the scope of the claims, which do not affect the substantive content of the present application. The above preferred features may be used in any combination as long as they do not conflict with each other.

Claims

1. A method for identifying control parameters of a new energy unit based on finite oscillation data, characterized in that: The method comprises: Obtain n sets of actual subsynchronous / supersynchronous oscillation measurement data and extract the key features of each oscillation mode; Based on the key characteristics of the oscillation mode, the equivalent RLC circuit of the grid-connected system of the new energy unit is established, and the impedance value of the new energy unit at n oscillation frequencies is calculated; According to the known control structure, the theoretical impedance model of the new energy unit is established, and the n oscillation frequencies f os and main circuit parameters, and obtain the impedance expressions of n new energy units containing the control parameters to be identified; Perform participation analysis based on the dominant oscillation mode to determine the number m of control parameters to be identified; The impedance calculated based on the equivalent RLC circuit and the theoretical impedance model is used to form a set of n equations. The control parameters to be identified are solved by judging the size between the number of equations n and the number of control parameters to be identified m. According to the known control structure, the theoretical impedance model of the new energy unit is established, and n oscillation frequencies f are respectively os Substituting the main circuit parameters into the equation, we can obtain the impedance expressions of n new energy units containing the control parameters to be identified, including: Obtain the control structure of the new energy unit and establish the AC port positive sequence impedance model Z of the new energy unit th (jω), ω is the angular frequency; Among them, Z th The derivation process of (jω) is as follows: Establish the impedance model Z of the AC port of the new energy unit in the dq coordinate system dq (s), through the relationship between dq impedance and sequence impedance, the positive sequence impedance model Z of the AC port of the new energy unit is obtained. th (jω), the calculation process is as follows: Among them, Z pn (s) is the two-dimensional improved sequence impedance of the AC port of the new energy unit; Z pp (s), Z pn (s), Z np (s), Z nn (s) are Z pn (s) is the matrix element; ω0 is the power frequency; s represents the Laplace domain; Substitute n oscillation frequencies and main circuit parameters into the AC port positive sequence impedance model Z th (jω), we can get the impedance expressions of n new energy units containing the control parameters to be identified.

2. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 1, characterized in that: The measured data includes the voltage and current time domain waveforms of the ports of the new energy generator set, and the oscillation frequencies of the oscillation components of each group are different; The key features of the oscillation mode include: oscillation voltage and current phasors, oscillation frequency and attenuation factor.

3. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 2, characterized in that: Obtain n sets of actual subsynchronous / supersynchronous oscillation measurement data and extract the key features of each oscillation mode, including: Obtaining measurement data of n groups of sub- / super-synchronous oscillations; Through the Hilbert transform method based on variational mode decomposition (VMD), VMD decomposes the time domain data into a series of intrinsic mode functions (IMFs) with finite bandwidth, and then fits the IMF components to obtain the oscillating voltage and current phasors, oscillation frequency and attenuation factor.

4. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 1, characterized in that: Based on the key characteristics of the oscillation mode, an equivalent RLC circuit of the grid-connected system of the new energy unit is established, and the impedance value of the new energy unit at n oscillation frequencies is calculated, including: From n sets of oscillation measurement data, obtain the key features of n sets of dominant oscillation modes and establish the second-order equivalent RLC circuit of the system at n oscillation frequencies; According to the obtained oscillation measurement data and the key characteristics of the oscillation mode, based on the expression of the equivalent RLC second-order circuit, the impedance value Zg of the new energy unit at n oscillation frequencies is calculated.

5. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 4 is characterized in that: The expression based on the equivalent RLC second-order circuit is: Where: Z g is the complex domain expression of the equivalent impedance of the grid-side new energy unit; R g 、R s are the equivalent resistances of the grid side and the new energy unit respectively; L g 、C g 、X g are the grid-side equivalent inductance, capacitance and reactance respectively; L s 、C s is the equivalent inductance and capacitance of the new energy unit; U s , I s and are the voltage, current amplitude and phase angle of the new energy unit side in the oscillation data; f os is the oscillation frequency; j is the complex imaginary unit.

6. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 4, characterized in that: The RLC second-order circuit includes: Regarding the sub- / super-synchronous oscillation phenomenon generated by direct-drive wind turbines connected to a weak grid, the grid side exhibits a resistance-inductance characteristic, while the direct-drive wind turbine exhibits a resistance-capacitance characteristic, forming an RLC resonant circuit at the super-synchronous resonant frequency. Regarding the subsynchronous resonance phenomenon generated by the series compensation transmission system of the doubly fed wind turbine, the grid side presents resistance-capacitance characteristics, and the doubly fed wind turbine presents resistance-inductance characteristics, forming an RLC resonant circuit at the subsynchronous resonant frequency.

7. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 1, characterized in that: Based on the dominant oscillation mode, participation analysis is performed to determine the number m of control parameters to be identified. Specifically, the participation factor represents the contribution of the state variable to the mode, based on which participation analysis is performed; The participation expression is as follows: Where: k w Characterizes the participation of the i-th mode in the participation factor with respect to the k-th state variable in all state variables; u ki 、v ik are the corresponding values ​​of the i-th mode in the left and right eigenvectors of the state matrix A with respect to the k-th state variable; N is the total number of oscillation modes; By solving the participation of each state variable under the dominant oscillation mode, the state variable with the largest participation is recorded as the dominant state variable under the dominant oscillation mode. The corresponding control link contains m control parameters to be identified; the remaining control parameters select typical values.

8. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 1, characterized in that: The impedance calculated based on the equivalent RLC circuit and the theoretical impedance model is used to simultaneously establish n equations. By judging the size between the number of equations n and the number of control parameters to be identified m, the control parameters to be identified are solved, specifically including: If the number of equations n is greater than or equal to the number of parameters to be identified m, it is an overdetermined system of equations. Select m linearly independent equations from the n equations and solve them to obtain the unique solution of the control parameters to be identified. If the number of equations n is less than the number of parameters to be identified m, it is an indeterminate system of equations and constraints need to be added to solve the equations.

9. The method for identifying control parameters of a new energy generator set based on finite oscillation data according to claim 8, characterized in that: The constraints are: supplementing typical values ​​of mn non-dominant parameters and solving n control parameters to be identified by setting up n equations in parallel; The division of dominant and non-dominant control parameters is determined by parameter sensitivity, which is defined as follows: Where: i is the i-th eigenvalue; u i 、v i are the left and right eigenvectors corresponding to the eigenvalue respectively; α represents a parameter.

Images