Analysis model and suppression method for parallel oscillation of follow-to-network converter

Abstract

The invention provides an analysis model and a suppression method for parallel oscillation of a following-networking converter, and belongs to the technical field of power system control. The method comprises the following steps: constructing a small signal model of the following network construction converter parallel system; on the basis of the small signal model, constructing a multi-dimensional frequency domain impedance tensor of the follow-up network construction converter parallel system; performing frequency domain sampling and matrix decomposition on the multi-dimensional frequency domain impedance tensor, identifying an unstable coupled oscillation mode in the following network construction converter parallel system, and determining a dominant oscillation frequency and a dominant converter set of the unstable coupled oscillation mode; and injecting preset virtual impedance into a local control loop of each networking converter, and executing control loop parameter self-adaptive adjustment on the networking converter into which the virtual impedance is injected. According to the method, the virtual impedance is directionally injected into the key converter, and the self-adaptive parameter adjustment is combined, so that the precise suppression of the specific oscillation mode is realized.

Description

Technical Field

[0001] This invention relates to the field of power system control technology, specifically to a parallel oscillation analysis model and suppression method for grid converters. Background Technology

[0002] With the continuous increase in the proportion of renewable energy power generation and the rapid development of microgrid technology, grid-connected converters with autonomous grid-building capabilities are widely used in power systems and are gradually becoming core equipment for building new power systems. In practical engineering, to meet capacity and reliability requirements, multiple grid-connected converters often need to operate in parallel, sharing a common AC bus to form a centralized power supply or distributed collaborative grid-connected system. This parallel architecture, on the one hand, meets the ever-increasing load power supply demand through capacity aggregation, and on the other hand, improves the system's power supply reliability through equipment redundancy. When a single converter fails, the remaining units can continue to maintain stable system operation.

[0003] Especially in scenarios where intermittent renewable energy sources such as wind and solar power are massively integrated into the grid, grid-connected converter systems not only need to perform basic energy conversion functions, but also undertake the critical tasks of providing voltage and frequency support, inertia response, and damping characteristics for weak power grids. With the continuous increase in system capacity, ensuring stable and coordinated operation among multiple grid-connected converters operating in parallel, and avoiding oscillations caused by control interactions, has become a core challenge that urgently needs to be addressed in the field of new energy power generation grid connection technology. Summary of the Invention

[0004] The purpose of this invention is to provide a parallel oscillation analysis model and suppression method for grid-connected converters, which is used to analyze and actively suppress multi-mode oscillation problems that occur in the operation of grid-connected converter systems, so as to ensure system stability.

[0005] To achieve the above objectives, this invention provides a parallel oscillation analysis model and suppression method for grid-connected converters, comprising: constructing a small-signal model of the parallel grid-connected converter system based on the control structure of each grid-connected converter and the grid parameters of the parallel grid-connected converter system; constructing a multi-dimensional frequency domain impedance tensor of the parallel grid-connected converter system based on the small-signal model; performing frequency domain sampling and matrix decomposition on the multi-dimensional frequency domain impedance tensor to identify unstable coupled oscillation modes in the parallel grid-connected converter system, and determining the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation modes; injecting a preset virtual impedance into the local control loop of each grid-connected converter in the dominant converter set to suppress the corresponding unstable coupled oscillation mode; and performing adaptive adjustment of control loop parameters for the grid-connected converters with injected virtual impedance based on the unstable coupled oscillation mode to suppress oscillations in the parallel grid-connected converter system.

[0006] Optionally, a small-signal model of the parallel system of grid-connected converters is constructed, including: extracting the control structure parameters and grid operating parameters of each grid-connected converter; the control structure parameters include the voltage outer loop proportional coefficient and the current inner loop proportional-integral coefficient; the grid operating parameters include the equivalent grid impedance characteristics at the point of common coupling; based on the extracted control structure parameters and grid operating parameters, a linearized state equation of the parallel system of grid-connected converters is established in the linearized neighborhood of the rated operating point using the state-space averaging method.

[0007] Optionally, the linearized state equation is established in a synchronous rotating coordinate system; the linearized state equation is subjected to a Laplace transform to obtain a transfer function matrix characterizing the multi-input multi-output characteristics of the grid-connected converter parallel system. The transfer function matrix is ​​used to characterize the output impedance characteristics and input admittance characteristics of the grid-connected converter parallel system.

[0008] Optionally, constructing the multidimensional frequency domain impedance tensor of the parallel grid converter system includes: performing a frequency domain scan on the transfer function matrix obtained from the small-signal model within a preset frequency band; calculating the output impedance matrix and the input admittance matrix at each frequency point of the frequency domain scan; and combining the output impedance matrix and the input admittance matrix in order of frequency points to form a multidimensional frequency domain impedance tensor to characterize the global impedance characteristics of the parallel grid converter system.

[0009] Optionally, the step of performing frequency domain sampling and matrix decomposition on the multidimensional frequency domain impedance tensor includes: performing frequency domain sampling on the multidimensional frequency domain impedance tensor at a set sampling interval within a preset frequency band; performing singular value decomposition on the impedance matrix at each sampling frequency point to obtain the corresponding singular value spectrum and singular vector; and identifying the unstable coupled oscillation mode based on the trajectory characteristics of the singular value spectrum according to the generalized Nyquist stability criterion.

[0010] Optionally, determining the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation mode includes: extracting frequency points in the singular value spectrum whose singular value amplitude is greater than a first preset threshold and whose phase margin is lower than a second preset threshold as candidate frequency points; sorting the candidate frequency points in descending order of singular value amplitude; and determining the center frequency corresponding to the candidate frequency point with the largest singular value amplitude after sorting as the dominant oscillation frequency.

[0011] Optionally, determining the dominant oscillation frequency and the dominant converter set of the unstable coupled oscillation mode further includes: obtaining the left singular vector and right singular vector corresponding to the dominant oscillation frequency; calculating the participation factor of each grid converter to the unstable coupled oscillation mode, wherein the participation factor is obtained by taking the modulus of the product of the corresponding elements of the left singular vector and the right singular vector; arranging the participation factors in descending order, selecting the grid converters arranged at the top of the preset proportion; and forming the dominant converter set from the selected grid converters.

[0012] Optionally, injecting a preset virtual impedance into the local control loop of each grid converter includes: obtaining the dominant oscillation frequency of the unstable coupled oscillation mode; configuring the transfer function of the band-stop filter type virtual impedance based on the dominant oscillation frequency; and injecting the configured virtual impedance into the current loop control loop of each grid converter in the set of dominant converters in the form of voltage feedforward.

[0013] Optionally, the adaptive adjustment of the execution control loop parameters includes: periodically collecting oscillation data of the parallel system with the grid-connected converter to update the damping ratio change of the unstable coupled oscillation mode; using the updated damping ratio change as feedback, adjusting the proportional-integral parameters of the inner current loop of the grid-connected converter with the injected preset virtual impedance through a fuzzy controller; and locking the control parameters at this time as the optimal parameter set when the oscillation energy decay rate of the parallel system with the grid-connected converter meets the preset stability criterion.

[0014] On the other hand, the present invention provides a parallel oscillation analysis model and suppression system for a grid-connected converter, which is used to implement the parallel oscillation analysis model and suppression method for a grid-connected converter. The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the parallel oscillation analysis model and suppression method for the grid-connected converter.

[0015] The aforementioned technical solution achieves multi-dimensional analysis of the multi-mode oscillation characteristics of the entire grid-connected converter parallel system by constructing a multi-dimensional frequency-domain impedance tensor. This allows for accurate identification of the dominant frequencies of unstable coupled oscillation modes and the key converter sets. By directionally injecting virtual impedance into the key converters and combining it with adaptive parameter adjustment, precise suppression of specific oscillation modes is achieved, effectively avoiding interference with normal system operation caused by traditional global suppression methods. This method significantly improves the stability and reliability of the grid-connected converter parallel system.

[0016] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the following detailed description to explain the invention, but do not constitute a limitation thereof. In the drawings:

[0018] Figure 1 This is a flowchart of the analysis model and suppression method for parallel oscillation of a grid converter.

[0019] Figure 2 This is a flowchart of the virtual impedance suppression oscillation closed-loop control. Detailed Implementation

[0020] The following is in conjunction with the appendix Figure 1 -Appendix Figure 2 The specific implementation methods of the embodiments of the present invention will be described in detail below. It should be understood that the specific implementation methods described herein are only for illustrating and explaining the embodiments of the present invention, and are not intended to limit the embodiments of the present invention.

[0021] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with the relevant provisions of national laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0022] In the process of realizing this invention, the inventors of this application discovered that the prior art has the defect of being unable to accurately identify and effectively suppress coupled oscillations caused by the complex interaction of multiple grid converters.

[0023] Example 1

[0024] Reference Figures 1-2 This is the first embodiment of the present invention, which provides a parallel oscillation analysis model and suppression method for a grid converter, including:

[0025] S100: Based on the control structure of each grid converter and the grid parameters of the grid converter parallel system, construct a small-signal model of the grid converter parallel system.

[0026] In the embodiments of this application, constructing a small-signal model of a grid-connected converter parallel system includes: extracting control structure parameters and grid operating parameters of each grid-connected converter; the control structure parameters include the voltage outer loop proportional coefficient and the current inner loop proportional-integral coefficient; the grid operating parameters include the equivalent grid impedance characteristics at the point of common coupling; based on the extracted control structure parameters and grid operating parameters, establishing the linearized state equation of the grid-connected converter parallel system in the linearized neighborhood of the rated operating point using the state-space averaging method.

[0027] It should be noted that the grid-connected converter in this solution, or grid-connected converter, refers to a converter unit with a grid-type control structure and external voltage source characteristics. In a multi-converter parallel system, its steady-state operating point and dynamic response are subject to equivalent constraints from the public power grid or other parallel converters. It exhibits a following characteristic to the public voltage and frequency reference at the system level, rather than serving as the sole voltage or frequency reference establishment unit. While the grid-connected converter possesses the basic characteristics of grid-type control in its control structure, it does not operate in isolation. Its voltage, frequency, and power distribution behavior is dynamically coupled with other converters in the parallel system. This grid-connected converter is applicable to power system scenarios where multiple grid-connected control converters operate in parallel.

[0028] A grid-connected converter parallel system refers to an operating system formed by multiple grid-connected converters connected in parallel to the power grid or an equivalent power grid structure through a common connection point. The grid-connected converters form a dynamic coupling relationship through electrical connections and control interactions.

[0029] In the embodiments of this application, the linearized state equation is established in a synchronous rotating coordinate system; the linearized state equation is subjected to Laplace transform to obtain the transfer function matrix characterizing the multi-input multi-output characteristics of the grid-connected converter parallel system. The transfer function matrix is ​​used to characterize the output impedance characteristics and input admittance characteristics of the grid-connected converter parallel system.

[0030] In a preferred embodiment of this application, the parallel system of grid-connected converters consists of multiple grid-connected converters, each numbered. The parallel topology of the grid-connected converter system (i.e., a topology where all grid-connected converters are connected to the same common connection point) and the electrical connections between the converters are clearly defined. The modeling coverage is delineated, fully encompassing the internal control and filtering components of each grid-connected converter, as well as the equivalent link extending from the common connection point to the grid side.

[0031] Furthermore, control structure parameters for each grid-connected converter are collected individually, focusing on key control parameters such as the voltage outer loop proportional coefficient and the current inner loop proportional-integral coefficient. No detailed parameter design is required; only actual operating configuration or rated design parameter values ​​are obtained. Simultaneously, relevant operating parameters of the power grid and equipment are extracted, with a focus on the equivalent grid impedance characteristics (such as resistance and reactance characteristics) at the point of common coupling (PCC), and the filter element parameters configured for each grid-connected converter (such as the inductance and capacitance values ​​of L-type and LCL-type filters). The rated state parameters of the grid-connected converter parallel system during steady-state operation are determined, including the rated output voltage, output current, and output power of each grid-connected converter, as well as the rated voltage and frequency at the PCC. This rated steady-state state is used as the benchmark operating point for subsequent small-signal linearization analysis (including the steady-state values ​​of rated output voltage, current, and control parameters).

[0032] Furthermore, for each grid-connected converter, a small disturbance is applied near its rated operating point, i.e., the reference operating point. For example, a step disturbance of ±5% of the rated voltage amplitude is applied to the converter's voltage reference command, or a small fluctuation disturbance of ±3% of the rated voltage amplitude is applied to the grid-side input voltage. This ensures that the disturbance amplitude is within the applicable range of linearization and does not affect the basic steady-state operation of the system. Due to the switching characteristics of the switching devices and the nonlinear characteristics of the filtering elements and control loops, the original dynamic model of the grid-connected converter is a nonlinear model. During the disturbance process, the dynamic response data of the internal filtering loops (such as the inductor current of the L-type filter or the inductor current and capacitor voltage of the LCL-type filter) and the control loops (voltage loop, current loop, and power outer loop) of the converter under this nonlinear model are collected simultaneously.

[0033] Furthermore, based on the response data, the state-space averaging method is used to perform equivalent processing on the nonlinear model of the grid converter. By ignoring higher-order nonlinear terms under small disturbances and retaining the dominant linear terms, the influence of switching nonlinearity and component nonlinearity on modeling accuracy is eliminated. Finally, the linearized state equation of a single grid converter, representing the linear correspondence between state variables and input variables, is established as follows:

[0034]

[0035]

[0036] in, This indicates the increment of critical internal states of the converter (such as filter inductor current, capacitor voltage, control loop output, etc.) deviating from the rated operating point. express The rate of change reflects the dynamic response trend of the internal state; It indicates the increment of external disturbances or control commands (such as voltage reference, power command, grid voltage fluctuations, etc.) from the rated value; This indicates the increment of the converter's output (such as port voltage and current) from its rated value; , , , These are the state matrix, input matrix, output matrix, and direct transmission matrix of a single grid converter after linearization.

[0037] In a preferred embodiment of this application, constraint relationships are constructed based on Kirchhoff's current law and voltage law: Based on Kirchhoff's current law, a balance equation is established between the total current at the point of common coupling and the output current of each converter (the total current at the point of common coupling is equal to the sum of the output currents at the ports of all parallel converters). Based on Kirchhoff's voltage law, a voltage drop relationship is established between the port voltage of each converter and the voltage at the point of common coupling and the equivalent impedance of the line, thereby integrating the linearized state equations of all individual converters. Simultaneously, considering the electrical interactions generated by voltage coupling and current superposition of each converter through the point of common coupling, as well as the dynamic effects brought about by the line impedance, the state variables, input variables, and output variables of each individual converter's equations are uniformly mapped to a synchronous rotating coordinate system (dq axis), eliminating coupling terms and alternating components in the three-phase stationary coordinate system. The coupling equations are then uniformly organized through matrix splicing and variable merging, ultimately forming a system-level linearized state equation reflecting the dynamic coupling characteristics of the entire parallel system.

[0038] Furthermore, a Laplace transform is performed on each variable of the above system-level linearized state equations. The time variable is eliminated using the differential property of the Laplace transform, transforming the state-space model in the time domain into a frequency domain model. Then, based on the frequency domain model, the transfer function matrix characterizing the multi-input multi-output characteristics of the grid-connected converter system is obtained by rearranging the data according to the "input-output" correspondence, as follows:

[0039]

[0040]

[0041] in, represents the system-level transfer function matrix of the system connected in parallel with the grid converter, and s represents the complex frequency variable, which is the core variable of the Laplace transform. The complex frequency is represented by the real part, and j represents the imaginary unit. The imaginary part of the complex frequency is represented; the transfer function matrix has a dimension of q×p (q is the output dimension, and p is the input dimension). The frequency domain increment vector of the system output is represented. In this technical solution, it is specifically a vector composed of the voltage increment and current increment of each converter port. This represents the frequency domain increment vector of the system input. The system-level output matrix is ​​obtained by integrating the output matrices of each individual machine through coupling constraints at a common connection point, and it represents the linear mapping coefficients of the internal state variables of the system to the output variables. Represents the identity matrix. The system-level state matrix is ​​obtained by integrating the state matrices of each individual machine through Kirchhoff's laws. The system-level input matrix is ​​obtained by integrating the input matrices of each individual machine through coupling constraints at common connection points. This represents the system-level direct transmission matrix, which is obtained by integrating the direct transmission matrices of each individual machine through coupling constraints at a common connection point.

[0042] The above scheme transforms the complex and nonlinear dynamic characteristics (such as switching behavior and controller nonlinearity) of the actual system into a linear and analytical mathematical model by using the state-space averaging method and the linearization of the rated operating point.

[0043] S200: Based on the small-signal model, construct the multi-dimensional frequency domain impedance tensor of the grid converter parallel system.

[0044] In the embodiments of this application, constructing a multidimensional frequency domain impedance tensor for a grid-connected converter parallel system includes: performing a frequency domain scan on the transfer function matrix obtained from the small-signal model within a preset frequency band; calculating the output impedance matrix and the input admittance matrix at each frequency point of the frequency domain scan; and combining the output impedance matrix and the input admittance matrix in order of frequency points to form a multidimensional frequency domain impedance tensor to characterize the global impedance characteristics of the grid-connected converter parallel system.

[0045] In a preferred embodiment of this application, it is first necessary to determine the core parameters of the frequency domain scan, and then, in conjunction with the typical dynamic characteristic frequency bands of the grid converter parallel system, define a preset frequency band range as [[ , ], where the lower limit frequency Not lower than the power frequency (50Hz) of the grid converter parallel system, upper limit frequency The system covers the converter control bandwidth and the resonant frequency of the filter loop (2kHz-5kHz) to ensure coverage of critical frequency bands where the system may experience potential resonance and oscillation. Logarithmic or linear intervals are used to set the scanning frequency points. In resonant-sensitive frequency bands (filter resonant band, control loop cutoff band), the interval is reduced to improve resolution, while in non-sensitive frequency bands, the interval is increased to optimize computational efficiency. The angular frequencies corresponding to all scanning frequency points are recorded simultaneously.

[0046] Furthermore, for each scan frequency point determined above, the complex frequency variable s is replaced with a sinusoidal steady-state frequency domain form. (Take the real part =0, considering only sinusoidal steady-state characteristics, k is the frequency index variable), and substitute it into the system-level transfer function matrix derived by S100. The specific values ​​of the transfer function matrix are obtained by calculating each frequency point through matrix operations. After traversing all frequency points, a full-band transfer function matrix dataset is formed. |k=1,2,…,N}.

[0047] Based on the full-band transfer function matrix dataset, and combining the correlation between the output impedance matrix and the transfer function matrix, the system output impedance matrix corresponding to each frequency point is calculated. The matrix is ​​n×n dimensional (n is the number of parallel converters), and its elements are... Characterizing frequency points The impact of the current disturbance of the j-th converter on the voltage of the i-th converter is determined; subsequently, matrix inversion is performed on the output impedance matrix at the same frequency point to obtain the corresponding input admittance matrix. Its elements Characterizing frequency points The extent to which the voltage disturbance of the j-th converter affects the current of the i-th converter is determined, ultimately forming a full-band impedance matrix dataset { |k=1,2,…,N} and the dataset of admittance matrices { |k=1,2,…,N}.

[0048] Furthermore, the full-band output impedance matrix dataset and input admittance matrix dataset are combined according to a preset frequency scanning order to construct a multi-dimensional frequency domain impedance tensor. This tensor adopts an N×n×n three-dimensional structure, with the first dimension corresponding to the number of scanned frequency points N, and the second and third dimensions corresponding to the n×n-dimensional impedance and admittance matrices (n being the number of parallel converters). By embedding the impedance and admittance matrices corresponding to each frequency point into the corresponding positions of the tensor in sequence, the full-band impedance and admittance characteristics are fully integrated. The resulting multi-dimensional frequency domain impedance tensor can accurately characterize the global impedance characteristics of the grid-connected converter system within the preset frequency band.

[0049] A simplified example is as follows: If we preset N=3 scanning frequencies and n=2 parallel converters, then this tensor has a 3×2×2 structure, with each frequency layer containing a 2×2 impedance matrix and a 2×2 admittance matrix. Its form can be simplified as follows:

[0050]

[0051] in, Represents the multidimensional frequency domain impedance tensor. Represents the k-th frequency point The output impedance matrix elements are as follows: i represents the affected converter number, and j represents the converter number that generates the current disturbance. Represents the k-th frequency point The input admittance matrix elements below, where i represents the converter number that generates the current response, and j represents the converter number that generates the voltage disturbance; the three scan points are respectively , , .

[0052] The above scheme transforms the time-domain state equation into a transfer function matrix that characterizes the multi-input multi-output characteristics through Laplace transform and frequency domain scanning. It then constructs a multi-dimensional tensor containing full-band output impedance and input admittance information. This tensor fully describes the electrical coupling and interaction between parallel units through common connection points in the complex frequency domain, providing a complete data structure for system-level oscillation mode analysis based on frequency domain stability criteria.

[0053] S300: Performs frequency domain sampling and matrix decomposition on the multidimensional frequency domain impedance tensor to identify unstable coupled oscillation modes in a grid-connected converter system, and determines the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation mode.

[0054] In embodiments of this application, frequency domain sampling and matrix decomposition of a multidimensional frequency domain impedance tensor include: frequency domain sampling of the multidimensional frequency domain impedance tensor within a preset frequency band at a set sampling interval; and singular value decomposition of the impedance matrix at each sampling frequency point to obtain the corresponding singular value spectrum and singular vector. The singular value decomposition formula is as follows:

[0055]

[0056] in, Describes a diagonal singular value matrix. Describes a left singular vector matrix. This represents the conjugate transpose of a right singular vector matrix.

[0057] Based on the generalized Nyquist stability criterion, unstable coupled oscillation modes are identified according to the trajectory characteristics of the singular value spectrum.

[0058] In a preferred embodiment of this application, determining the dominant oscillation frequency and the dominant converter set of the unstable coupled oscillation mode includes: extracting frequency points in the singular value spectrum whose singular value amplitude is greater than a first preset threshold (singular value amplitude threshold, usually 1) and whose phase margin is lower than a second preset threshold (phase margin threshold, usually 30°) as candidate frequency points; sorting the candidate frequency points in descending order of singular value amplitude; and determining the center frequency corresponding to the candidate frequency point with the largest singular value amplitude after sorting as the dominant oscillation frequency.

[0059] In a preferred embodiment of this application, determining the dominant oscillation frequency and the set of dominant converters for the unstable coupled oscillation mode further includes: obtaining the left singular vector and right singular vector corresponding to the dominant oscillation frequency; calculating the participation factor of each grid converter for the unstable coupled oscillation mode, wherein the participation factor is obtained by taking the modulus of the product of the corresponding elements of the left singular vector and the right singular vector, and the calculation formula for the participation factor is as follows:

[0060]

[0061] in, This represents the participation factor of the i-th grid converter in the unstable coupled oscillation mode. This represents the left singular vector corresponding to the dominant oscillation frequency. This represents the i-th element of the right singular vector corresponding to the dominant oscillation frequency.

[0062] Participation factors Sort the converters in descending order of their numerical values ​​and select the first 1 or 2 converters in the grid. The selected grid converters form the main converter set.

[0063] The above scheme, based on the generalized Nyquist stability criterion, performs singular value decomposition on the impedance tensor, thereby decoupling the multi-input multi-output system into a series of independent single-input single-output channels. The stability of each oscillation mode is determined based on the encirclement of their singular value trajectories in the complex plane. Furthermore, participation factor analysis quantifies the contribution of each converter to specific unstable modes, accurately identifying the dominant oscillation frequency and its corresponding dominant converter set, thus achieving observability and localization of the oscillation source.

[0064] S400: In the main converter set, a preset virtual impedance is injected into the local control loop of each grid converter to suppress the corresponding unstable coupled oscillation mode.

[0065] In the embodiments of this application, injecting a preset virtual impedance into the local control loop of each grid converter includes: obtaining the dominant oscillation frequency of the unstable coupled oscillation mode; configuring the transfer function of the band-stop filter type virtual impedance based on the dominant oscillation frequency; and injecting the configured virtual impedance into the current loop control loop of each grid converter in the set of dominant converters in the form of voltage feedforward.

[0066] In a preferred embodiment of this application, based on the determined dominant parameters of the unstable coupled oscillation mode, an oscillation suppression operation is performed in the set of dominant converters, and the identified dominant angular frequency is retrieved. And through the formula Converted to the corresponding Hertz dominant frequency This is used as the center frequency of the target suppression band for the virtual impedance.

[0067] Furthermore, the complex frequency domain transfer function of the band-stop filter-type virtual impedance is configured as follows:

[0068]

[0069]

[0070] in, The complex frequency domain transfer function representing the virtual impedance of a band-stop filter. This represents the virtual impedance amplitude adjustment coefficient. Indicates the gain coefficient; Q represents the maximum singular value of the output impedance matrix at the dominant oscillation frequency, and Q represents the band-stop filter quality factor. The selection is based on the frequency band concentration of the oscillation mode. When the singular value spectrum shows that the oscillation mode exhibits highly localized characteristics near the dominant frequency (i.e., the -3dB bandwidth is less than 5% of the system fundamental frequency), a higher Q value (2-5) is selected to achieve frequency selective suppression. When the oscillation mode exhibits broadband characteristics or there is a risk of frequency drift, a lower Q value (0.5-1) is selected to enhance the robustness of the suppression strategy.

[0071] Furthermore, the configured virtual impedance transfer function In the form of voltage feedforward, the current loop control circuit of each grid converter in the dominant converter assembly is embedded, so that the dominant converter operates at the dominant oscillation frequency. The corresponding frequency band exhibits high impedance characteristics, blocking the energy transfer path of unstable coupled oscillations.

[0072] Optionally, after virtual impedance injection, the system's multidimensional frequency domain impedance tensor is reconstructed and oscillation modes are identified. If the singular amplitude corresponding to the original unstable mode drops below a first preset threshold and the phase margin increases above a second preset threshold, then the oscillation suppression is deemed effective. If the suppression effect does not meet expectations, fine-tuning is performed. (The value ranges from 0.1Ω to 5Ω and can be dynamically adjusted according to the oscillation suppression requirements. The larger the value, the higher the impedance rise in the dominant oscillation frequency band) or Q (the value ranges from 0.5 to 2, used to adjust the band width of the band-stop filter. The smaller the Q value, the wider the filter bandwidth, which can cover small frequency shifts near the dominant oscillation frequency. The larger the Q value, the steeper the filter characteristics and the less interference to non-target frequency bands) parameter values, until the stability requirements of the parallel system with the grid converter are met.

[0073] The above scheme, by injecting a virtual impedance with band-stop filtering characteristics into the inner current loop of the dominant converter assembly, can reshape the equivalent output impedance characteristics of the system at a specific oscillation frequency. By increasing the amplitude margin of the open-loop transfer function of the oscillation loop at the critical frequency point, the amplitude and phase conditions of the oscillation are disrupted, effectively blocking the transmission path of negatively damped energy.

[0074] S500: Based on unstable coupled oscillation mode, it performs adaptive adjustment of control loop parameters for grid converters with injected virtual impedance to suppress oscillations in parallel grid converter systems.

[0075] In the embodiments of this application, the adaptive adjustment of control loop parameters includes: periodically collecting oscillation data of the parallel system of the grid converter to update the damping ratio change of the unstable coupled oscillation mode; using the updated damping ratio change as feedback, adjusting the proportional-integral parameters of the inner current loop of the grid converter with a preset virtual impedance injected through a fuzzy controller; and locking the control parameters at this time as the optimal parameter set when the oscillation energy attenuation rate of the parallel system of the grid converter meets the preset stability criterion.

[0076] In a preferred embodiment of this application, adaptive adjustment of control loop parameters is performed. First, real-time waveform data of port voltage and current of the parallel system are collected at a preset period (e.g., 100ms-500ms). The frequency and amplitude of the oscillation components are extracted using Fast Fourier Transform (FFT). The damping ratio change of the unstable coupled oscillation mode is calculated by combining the response curve of the parallel system of the grid converter. This damping ratio change and its rate of change are used as the dual-input feedback quantities of the fuzzy controller. The fuzzy controller predefines the fuzzy subsets of input and output as seven levels: negative large, negative medium, negative small, zero, positive small, positive medium, and positive large. Based on existing fuzzy control technology, fuzzification processing, fuzzy rule reasoning, and defuzzification calculation are performed sequentially, and the output is the proportional-integral parameter adjustment quantity of the inner loop current of the grid converter with injected virtual impedance. (Real-time adjustment of the proportional coefficient of the inner loop current of the grid converter) and (The real-time adjustment of the integral coefficient of the current loop in the grid converter) is updated using an incremental formula to update the proportional coefficient of the current loop. and integral coefficient The updated formula is as follows:

[0077]

[0078]

[0079] in, , These are the proportional and integral coefficient values ​​for the current control cycle. , These are the proportional and integral coefficient values ​​for the next control cycle after the update.

[0080] It should be noted that, in order to avoid parameters exceeding the reasonable range of the project, constraints need to be applied to the updated parameters: The value range is limited to 0.1-5. The value range is limited to 0.01-1. If the updated parameter exceeds the threshold, the threshold boundary value is taken as the final parameter.

[0081] Furthermore, the system continuously monitors the oscillation energy decay rate. When the preset stability criterion is met, i.e. the oscillation energy decay rate is ≥80% within 3 consecutive cycles and the oscillation amplitude drops to less than 5% of the power frequency amplitude, the optimal parameter set consisting of the current inner loop PI parameter and the virtual impedance parameter is locked at this time, and the adaptive adjustment process of the control loop parameters is completed.

[0082] The above scheme monitors the damping ratio change of the oscillation mode in real time and uses this as feedback to perform online adaptive tuning of the proportional-integral parameters of the converter current loop with injected virtual impedance using a fuzzy controller. This closed-loop optimization mechanism enables the grid-connected converter system to dynamically adapt to changes in operating conditions, automatically searching for and converging to a set of optimal controller parameters that simultaneously considers stability, dynamic response speed, and steady-state accuracy, thereby ensuring the durability and reliability of the oscillation suppression effect.

[0083] The present invention also provides a parallel oscillation analysis model and suppression system for a grid-connected converter, which is used to implement the parallel oscillation analysis model and suppression method for a grid-connected converter. The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the parallel oscillation analysis model and suppression method for the grid-connected converter.

[0084] This invention provides a storage medium storing a program that, when executed by a processor, implements the parallel oscillation analysis model and suppression method for the grid converter.

[0085] This invention provides a processor for running a program, wherein the program executes the parallel oscillation analysis model and suppression method of the grid converter.

[0086] This invention provides a device including a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements a parallel oscillation analysis model and suppression method for a grid-connected converter. The device described herein can be a server, PC, tablet, mobile phone, etc.

[0087] This application also provides a computer program product that, when executed on a data processing device, is suitable for executing a parallel oscillation analysis model and suppression method for a grid converter.

[0088] Those skilled in the art will understand that embodiments of this application can provide methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0089] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0093] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0094] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0095] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0096] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A parallel oscillation analysis model and suppression method for a grid converter, characterized in that, include: Based on the control structure of each grid converter and the grid parameters of the grid converter parallel system, a small-signal model of the grid converter parallel system is constructed. Based on the small-signal model, construct the multi-dimensional frequency domain impedance tensor of the parallel grid converter system; Frequency domain sampling and matrix decomposition are performed on the multidimensional frequency domain impedance tensor to identify the unstable coupled oscillation modes in the parallel system of the grid converter, and to determine the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation modes. In the set of dominant converters, a preset virtual impedance is injected into the local control loop of each grid converter to suppress the corresponding unstable coupled oscillation mode; Based on the aforementioned unstable coupled oscillation mode, adaptive adjustment of control loop parameters is performed on the grid converter with injected virtual impedance to suppress oscillations in the parallel system of the grid converter.

2. The parallel oscillation analysis model and suppression method for grid converters according to claim 1, characterized in that, The construction of the small-signal model of the grid converter parallel system includes: Extract the control structure parameters and grid operation parameters of each grid converter; The control structure parameters include the voltage outer loop proportional coefficient and the current inner loop proportional-integral coefficient. The power grid operating parameters include the equivalent power grid impedance characteristics at the point of common coupling. Based on the extracted control structure parameters and grid operating parameters, the linearized state equations of the grid converter parallel system are established in the linearized neighborhood of the rated operating point using the state-space averaging method.

3. The parallel oscillation analysis model and suppression method for grid converters according to claim 2, characterized in that, The linearized state equation is established in a synchronous rotating coordinate system; the linearized state equation is subjected to a Laplace transform to obtain a transfer function matrix characterizing the multi-input multi-output characteristics of the grid-connected converter parallel system. The transfer function matrix is ​​used to characterize the output impedance characteristics and input admittance characteristics of the grid-connected converter parallel system.

4. The parallel oscillation analysis model and suppression method for grid converters according to claim 3, characterized in that, The construction of the multidimensional frequency domain impedance tensor of the parallel grid converter system includes: Within a preset frequency band, the transfer function matrix obtained from the small-signal model is scanned in the frequency domain; At each frequency point of the frequency domain scan, calculate the output impedance matrix and the input admittance matrix; The output impedance matrix and the input admittance matrix are combined in frequency order to form a multidimensional frequency domain impedance tensor, which characterizes the global impedance characteristics of the parallel grid converter system.

5. The parallel oscillation analysis model and suppression method for grid converters according to claim 1, characterized in that, The step of performing frequency domain sampling and matrix decomposition on the multidimensional frequency domain impedance tensor includes: The multidimensional frequency domain impedance tensor is sampled in the frequency domain at a set sampling interval within a preset frequency band. Singular value decomposition is performed on the impedance matrix at each sampling frequency point to obtain the corresponding singular value spectrum and singular vector; Based on the generalized Nyquist stability criterion, the unstable coupled oscillation mode is identified according to the trajectory characteristics of the singular value spectrum.

6. The parallel oscillation analysis model and suppression method for grid converters according to claim 5, characterized in that, Determining the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation mode includes: Frequency points in the singular value spectrum whose singular value amplitude is greater than a first preset threshold and whose phase margin is less than a second preset threshold are extracted as candidate frequency points; Sort the candidate frequency points in descending order of singular value amplitude; The center frequency corresponding to the candidate frequency point with the largest singular value amplitude after sorting is determined as the dominant oscillation frequency.

7. The parallel oscillation analysis model and suppression method for grid converters according to claim 6, characterized in that, Determining the dominant oscillation frequency and dominant converter set of the unstable coupled oscillation mode further includes: Obtain the left singular vector and right singular vector corresponding to the dominant oscillation frequency; Calculate the participation factor of each grid converter for the unstable coupled oscillation mode. The participation factor is obtained by taking the modulus of the product of the corresponding elements of the left singular vector and the right singular vector. Arrange the participating factors in descending order and select the grid converters that are ranked first by a preset proportion. The selected grid converters constitute the dominant converter set.

8. The parallel oscillation analysis model and suppression method for grid converters according to claim 1, characterized in that, The injection of a preset virtual impedance into the local control loop of each grid converter includes: Obtain the dominant oscillation frequency of the unstable coupled oscillation mode; The transfer function of the band-stop filter type virtual impedance is configured based on the dominant oscillation frequency; The configured virtual impedance is injected into the current loop control circuit of each grid converter in the dominant converter set in the form of voltage feedforward.

9. The parallel oscillation analysis model and suppression method for grid converters according to claim 1, characterized in that, The adaptive adjustment of the execution control loop parameters includes: Oscillation data of the grid-connected converter system are periodically collected to update the damping ratio change of the unstable coupled oscillation mode; Using the updated damping ratio change as feedback, the fuzzy controller adjusts the inner loop proportional-integral parameters of the current of the grid converter with the injected virtual impedance. When the oscillation energy decay rate of the system connected in parallel with the grid converter meets the preset stability criterion, the control parameters at this time are locked as the optimal parameter set.

10. A parallel oscillation analysis model and suppression system for a grid converter, characterized in that, The system includes a control module, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the parallel oscillation analysis model and suppression method for grid converters according to any one of claims 1-9.

Images