Inverter power supply control method and system providing positive and negative dual sequence transient voltage support

By establishing a mathematical model for the transient voltage support of an inverter power supply that coordinates positive and negative dual-sequence voltages, and using a globally optimal reference current command to control the inverter power supply, the problem of coordinating positive-sequence voltage rise and negative-sequence voltage suppression under grid faults is solved, thereby maximizing and adaptively adjusting the system's transient voltage support.

CN123074005APending Publication Date: 2026-11-10SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611552214.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-09
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing inverter power supply control methods cannot effectively coordinate positive sequence voltage rise and negative sequence voltage suppression under grid fault conditions, resulting in poor transient voltage support of the system. Furthermore, existing optimization models fail to achieve adaptive coordination under different fault severity levels.

Method used

A mathematical model for transient voltage support of inverter power supply with coordinated positive and negative dual-sequence voltages is established by using the method of minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude as the objective function. The model is solved by the globally optimal reference current command to realize the positive sequence voltage rise and negative sequence voltage suppression of inverter power supply during grid faults.

Benefits of technology

It maximizes the inverter's support effect on the system's transient voltage during grid faults, overcomes the problem of fixed priority for positive and negative sequence voltage support in existing optimization models, and can adaptively adjust the current distribution ratio according to the severity of the fault, ensuring global optimality in mathematics.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123074005A_ABST
    Figure CN123074005A_ABST
Patent Text Reader

Abstract

This invention belongs to the field of inverter power supply control technology, specifically disclosing an inverter power supply control method and system that provides positive and negative dual-sequence transient voltage support. The method includes: establishing a mathematical model for inverter power supply transient voltage support with coordinated positive and negative dual-sequence voltages, using the real and imaginary components of the positive and negative sequence currents injected into the inverter power supply at the point of common coupling as decision variables; transforming the mathematical model into a second-order moment relaxation problem with moment variables as optimization variables; solving the problem using a semidefinite programming solver to obtain the optimal moment variables; performing a rank test on the moment matrix corresponding to the optimal moment variables, and restoring the optimal current components under the unified complex plane when the rank-1 condition is met, thus obtaining the optimal positive and negative sequence dq current reference values; controlling the inverter power supply according to the current reference values, simultaneously raising the positive sequence voltage and suppressing the negative sequence voltage during grid faults, overcoming the problem of fixed priority for positive and negative sequence voltage support in existing optimization models.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of inverter power supply control technology, and in particular to an inverter power supply control method and system that provides positive and negative dual-sequence transient voltage support. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] Most new energy sources are connected to the grid via power electronic inverters. The large-scale replacement of conventional synchronous generator sets by these inverters has profoundly changed the operating characteristics of the power system, leading to a decrease in system voltage strength. Under grid faults, the problem of transient voltage safety and stability caused by insufficient voltage support capacity has become a core challenge for the safe and reliable operation of new power systems.

[0004] Inverter power supplies are flexible, fast, and arbitrarily controllable. They can implement control commands in milliseconds and are numerous with a large total capacity, providing a new path for transient voltage support.

[0005] Transient voltage support technologies, focusing on single-sequence voltage rise or negative-sequence voltage suppression, are relatively mature and can be broadly categorized into two types. One mainstream industrial approach provides control commands based on a pre-defined droop relationship between reactive power and AC voltage. However, droop control cannot guarantee full utilization of the inverter's capacity, and in most cases, the support effect is suboptimal. An unreasonable droop coefficient can even lead to system instability due to loss of equilibrium under severe fault conditions. The other type, based on mathematical optimization, constructs the transient voltage support problem as a constrained optimization model with the goal of maximizing positive-sequence voltage or minimizing negative-sequence voltage, using the active and reactive currents of the inverter as optimization variables. Solving this model yields the optimal control command. However, existing solutions primarily address control objectives under a single sequence, with less consideration given to the coordination between positive and negative sequences. Summary of the Invention

[0006] To address the aforementioned issues, this invention proposes an inverter power supply control method and system that provides positive and negative dual-sequence transient voltage support. It considers the optimal coordination between the objectives of positive-sequence voltage rise and negative-sequence voltage suppression, and takes into account various constraints such as current and active power. A mathematical model for inverter power supply transient voltage support with coordinated positive and negative dual-sequence voltages is established. By solving the globally optimal reference current command, the effectiveness of the inverter power supply in supporting the system's transient voltage under faults such as AC short circuits is maximized.

[0007] In some implementations, the following technical solutions are adopted: An inverter power supply control method providing positive and negative dual-sequence transient voltage support includes: Using the real and imaginary components of the positive and negative sequence currents injected by the inverter at the point of common coupling in a unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis as the decision variables, and minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling as the objective function, a mathematical model for the transient voltage support of the inverter with coordinated positive and negative sequence voltages is established. The mathematical model is transformed into a second-order moment relaxation problem with moment variables as optimization variables; The second-order moment relaxation problem is solved by calling a semidefinite programming solver to obtain the optimal moment variables. The moment matrix corresponding to the optimal moment variables is subjected to rank test. When the rank-1 condition is met, the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are recovered by spectral decomposition. The optimal positive and negative sequence dq current reference values ​​are obtained by coordinate transformation.

[0008] The optimal positive and negative sequence dq current reference values ​​are output to the positive and negative sequence current inner loop controllers of the inverter power supply, respectively, so as to control the inverter power supply to simultaneously raise the positive sequence voltage and suppress the negative sequence voltage during grid faults.

[0009] As a further solution, the transient voltage support mathematical model for the inverter power supply with positive and negative dual-sequence voltage coordination is a quadratic programming problem with quadratic constraints, using the real and imaginary parts of the positive and negative sequence currents injected by the inverter power supply at the common coupling point as decision variables.

[0010] As a further solution, the process of constructing the mathematical model for the transient voltage support of the inverter power supply with positive and negative dual-sequence voltage coordination includes: In the positive and negative sequence dq rotating coordinate system, based on Kirchhoff's voltage law, the relationship between the positive sequence voltage at the point of common coupling (PCC) and the positive sequence dq current component in the positive sequence grid voltage oriented dq coordinate system, and the relationship between the negative sequence voltage at the PCC and the negative sequence dq current in the negative sequence grid voltage oriented dq coordinate system are established respectively. By defining new variables, the negative sequence current is rotated from the negative sequence grid voltage orientation dq coordinate system to a unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis, while keeping the positive sequence current unchanged, so that the positive sequence and negative sequence current components are in the same complex plane. Based on the new variables, quadratic expressions for the magnitudes of the positive and negative sequence voltages at the PCC, as well as the magnitudes of the three-phase current, are given again; and constraints on the magnitudes of the three-phase current and the upper and lower limits of the active power are established. Based on the given expression and constraints, a mathematical model for transient voltage support of inverter power supply with coordinated positive and negative dual-sequence voltages is constructed.

[0011] As a further solution, the three-phase current amplitude constraint is expressed by establishing quadratic inequalities about the decision variables for the squares of the current amplitudes of phases A, B, and C, respectively, requiring that the current amplitude of each phase does not exceed the maximum allowable current limit of the inverter power supply. The active power upper and lower limit constraints are established by creating a quadratic equation for the output active power of the inverter power supply with respect to the decision variables, requiring the output active power to be between the upper and lower limits of the active power.

[0012] As a further solution, the mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative dual-sequence voltages is specifically as follows: ; in, For decision variables, corresponding , , ; , Let represent the real part and the imaginary part of the positive-sequence current, respectively. and These represent the real part and the imaginary part of the negative sequence current, respectively. Represents the positive sequence grid voltage With negative sequence grid voltage The phase difference between them; Let Hessian matrix be the objective function. 、 、 These are the constraint matrices corresponding to the current amplitude limits for phases A, B, and C, respectively. For constant terms, , , These are the weighting coefficients; , , This indicates the maximum permissible limit of the corresponding phase current, which satisfies... ; , represents the lower and upper limits of active power constraints, respectively; g is the coefficient vector of the linear term in the objective function, and Gp is ​​the coefficient matrix of the quadratic term in the expression for the average active power of the inverter power supply. This is the coefficient vector of the linear terms in the active power expression.

[0013] As a further step, a rank test is performed on the moment matrix corresponding to the optimal moment variable, specifically as follows: The moment matrix is ​​filled with the components of the optimal moment variables according to the rule of adding monomial exponents; If the rank of the moment matrix is ​​1, then the relaxation is confirmed to be accurate and the global optimal solution of the original QCQP can be obtained. The moment matrix is ​​then subjected to spectral decomposition to extract the first-order monomial components and recover the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane.

[0014] As a further step, the moment matrix is ​​subjected to spectral decomposition, specifically as follows: Calculate the maximum eigenvalue of the moment matrix and its corresponding unit eigenvector. Use the product of the square root of the maximum eigenvalue and the unit eigenvector as the recovery vector. Extract the first-order monomial components with indices 2 to 5 from the recovery vector to obtain the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents in the unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis.

[0015] As a further solution, the real and imaginary components of the positive and negative sequence currents of the optimal current are rotated to convert them into a unified coordinate system with the positive sequence voltage of PCC as the d-axis, so as to obtain the optimal positive and negative sequence dq current reference values, and the current reference values ​​are provided to the positive sequence and negative sequence current inner loop controllers respectively.

[0016] In other embodiments, the following technical solutions are adopted: An inverter power supply control system providing positive and negative dual-sequence transient voltage support includes: The model building module is used to establish a mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative sequence voltages. The real and imaginary components of the positive and negative sequence currents injected by the inverter power supply are used as decision variables in a unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis. The objective function is to minimize the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling. The model transformation module is used to transform the mathematical model into a second-order moment relaxation problem with moment variables as optimization variables; The model solving module is used to call the semidefinite programming solver to solve the second-order moment relaxation problem and obtain the optimal moment variables; the rank test is performed on the moment matrix corresponding to the optimal moment variables, and when the rank-1 condition is met, the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are obtained through spectral decomposition. The coordinate transformation and inverter power supply control module is used to convert the optimal values ​​of the real and imaginary components of the positive and negative sequence currents into reference values ​​of the positive and negative sequence dq currents with the positive sequence voltage of PCC as the d-axis, and output them to the current inner loop controller of the inverter power supply to control the inverter power supply to simultaneously raise the positive sequence voltage and suppress the negative sequence voltage during grid faults.

[0017] In other embodiments, the following technical solutions are adopted: A terminal device includes a processor and a memory, the processor being used to implement instructions; the memory being used to store multiple instructions adapted to be loaded and executed by the processor to provide positive and negative dual-sequence transient voltage support for the inverter power supply control method described above.

[0018] Compared with the prior art, the beneficial effects of the present invention are: This invention employs a method with the objective function of minimizing the weighted difference between the square of the negative-sequence voltage amplitude and the square of the positive-sequence voltage amplitude. This unifies the two mutually coupled control objectives of positive-sequence voltage rise and negative-sequence voltage suppression into the same optimization framework. It can adaptively adjust the current distribution ratio of positive and negative-sequence support according to the severity of the fault, thus overcoming the problem of fixed priority for positive and negative-sequence voltage support in existing optimization models.

[0019] This invention proposes a global optimal solution method for quadratic programming problems with non-convex quadratic constraints based on moment relaxation. The non-convex primitive problem with real and imaginary components of positive and negative sequence currents under a unified complex plane as decision variables is transformed into a convex relaxation problem. When the rank of the moment matrix is ​​1, the mathematical global optimality of the obtained solution can be guaranteed, avoiding the defect of conventional solvers getting trapped in local optima.

[0020] Other features and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] Figure 1 This is a flowchart of an inverter power supply control method that provides positive and negative dual-sequence transient voltage support in an embodiment of the present invention; Figure 2 This is a block diagram of the main circuit and control of the grid-connected inverter power supply in an embodiment of the present invention; Figure 3 This is a schematic diagram of the phasor relationship between positive and negative sequence voltages and currents in a unified complex plane. Figure 4 The simulation results of the method and comparison method of the present invention under fault scenarios 1 and 2 are shown below; Figure 5 The simulation results of the method and the comparison method of the present invention are shown in fault scenarios 3 and 4. Detailed Implementation

[0022] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0024] Example 1 Current inverter reference current commands during faults primarily rely on reactive power or voltage droop rules, issuing corresponding reactive current commands based on the magnitude of positive or negative sequence voltage. However, they do not specify the calculation method for active current. In low impedance ratio (X / R) scenarios (i.e., when the reactance-to-resistance ratio X / R is very small), this active current's effect on voltage support is not negligible, such as in distribution networks. Therefore, existing conventional control methods cannot maximize the transient voltage support effect. A few mathematical optimization-based control methods simultaneously specify the calculation methods for both positive and negative sequence active and reactive currents, but they fail to achieve adaptive coordination of positive and negative sequence support priorities under different fault severity levels, and they also fail to achieve mathematically global optimality of the support effect.

[0025] Based on this, the present invention focuses on the transient voltage support of inverter power supplies, considering the optimal coordination between the positive sequence voltage rise and negative sequence voltage suppression objectives, and taking into account multiple types of constraints such as current and active power. It proposes a generalized design method for optimal current command of inverter power supplies with dual-sequence coordination, which maximizes the transient voltage support effect of inverter power supplies on the system under faults such as AC short circuits. This has important practical engineering significance for the safe and reliable transmission of new energy.

[0026] In one or more embodiments, an inverter power supply control method for providing positive and negative dual-sequence transient voltage support is disclosed, combined with Figure 1 Specifically, it includes the following processes: S101: Using the real and imaginary components of the positive and negative sequence currents injected into the inverter power supply under a unified complex plane at the point of common coupling as decision variables, and minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling as the objective function, a mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative sequence voltages is established.

[0027] The inverter power supply control method proposed in this embodiment is for conventional grid-connected inverters based on phase-locked loops (PLLs) and vector current control. This embodiment considers a three-phase three-wire grid-connected system, so the zero-sequence current component is ignored; the inverter power supply is configured with positive and negative sequence decoupled current control; the external grid of the PCC adopts the Thevenin equivalent model, and it is assumed that the positive and negative sequence equivalent grid impedances are the same, uniformly expressed as Z=R+jX; there are no special requirements for the energy type on the DC side, and it can be applied to grid-connected inverters using the above control structure, such as photovoltaic, wind power, and energy storage.

[0028] This embodiment only improves the way the positive and negative sequence dq current reference commands are given during grid faults. The main circuit and control block diagram of the grid-type inverter are as follows: Figure 2 As shown. Figure 2 "Other parts" refers to the DC-side power supply, such as wind turbines, photovoltaic arrays, or energy storage. The control system adopts a traditional cascaded vector control structure based on PLL, with the outer loop controlling power or voltage and the inner loop controlling current. To achieve decoupled control of positive-sequence and negative-sequence currents, a typical dual control method is employed.

[0029] For grid-connected inverters, since their inner current loop has a fast response speed, they can be equivalently modeled as controlled current sources for analysis and control on transient time scales.

[0030] Inverter power supply output three-phase instantaneous current , , It can be decomposed into positive sequence current using the symmetrical component method. With negative sequence current The superposition of, where the superscript " "and" The positive and negative sequence components are represented respectively, and bold text indicates phasors. The external power grid at the PCC (point of common coupling) is represented using Thevenin equivalents, and the three-phase instantaneous grid voltage is... , , The corresponding positive-sequence and negative-sequence voltage components are denoted as follows: and The equivalent grid impedance is expressed as Three-phase voltage at PCC , , The positive and negative order components are denoted as follows: and .

[0031] According to Kirchhoff's voltage law, combined with Figure 2 The equivalent circuit is modeled in positive and negative sequence dq rotating coordinate systems oriented by positive and negative sequence grid voltages, respectively, and the following voltage equations can be obtained: (1) In this context, non-bold variables represent the magnitudes of the corresponding phasors, such as... Represents positive sequence voltage The amplitude. Angle and These represent the phase difference between the positive-sequence PCC voltage and the positive-sequence grid voltage, and the phase difference between the negative-sequence PCC voltage and the negative-sequence grid voltage, respectively; angle Represents positive-sequence and negative-sequence grid voltages and Phase difference between them; , These represent the equivalent resistance and reactance of the power grid, respectively.

[0032] , These represent the positive-sequence dq-axis current components in the positive-sequence grid voltage-oriented coordinate system, while and These represent the negative-sequence dq-axis current components in the negative-sequence grid voltage-oriented coordinate system. After coordinate unification, these current components form the decision variables for the subsequent QCQP problem.

[0033] Based on equation (1), the squares of the positive-sequence voltage amplitude and the squares of the negative-sequence voltage amplitude of PCC can be further expressed as: (2) The amplitude of the A-phase output current of the inverter power supply can be expressed as: (3) Since the positive-sequence current and the negative-sequence current are defined in two different dq rotating coordinate systems with reference to the positive-sequence grid voltage and the negative-sequence grid voltage, respectively, there is a phase angle difference between them. Therefore, vector superposition cannot be performed directly in the same coordinate plane. On the other hand, the actual output current of the three phases of the inverter power supply is formed by the superposition of positive and negative sequence currents. In order to establish a unified algebraic expression for the amplitude of the three-phase current with respect to the positive and negative sequence currents, this embodiment maps the negative sequence current to the unified complex plane where the positive sequence current is located by coordinate rotation.

[0034] The phasor relationships of positive and negative sequence voltages and currents in a unified complex plane are as follows: Figure 3 As shown. Figure 3 In the middle, the positive sequence grid voltage phasor The direction is the real axis. The positive sequence current maintains its original coordinate representation, while the negative sequence current rotates to this unified complex plane according to the phase angle difference between the positive and negative sequence grid voltages. and These represent the phase angle differences between the positive and negative sequence voltages of the PCC and their corresponding grid sequence voltages, respectively. and These represent the voltage components corresponding to the equivalent resistance and reactance of the power grid, respectively.

[0035] based on Figure 3 The phasor relationships shown are used to introduce the following coordinate transformation for the negative sequence current in this embodiment: make , ,in , , and The newly introduced parameter has the following expression: , ; , ; The coordinates of the positive-sequence current remain unchanged, while the negative-sequence current follows the phase angle difference between the positive and negative-sequence grid voltages.

[0036] Through the above transformation, the influence of different reference angles of the positive and negative sequence dq coordinate systems on the superposition relationship of the three-phase currents can be eliminated, so that the PCC voltage, three-phase current and active power can all be uniformly expressed as quadratic functions about the same decision vector, thus providing a basis for subsequent QCQP modeling.

[0037] Therefore, the PCC voltage can be rewritten as follows: (4) Three-phase current can be re-expressed as: (5) (6) (7) To ensure the safe operation of the converter, the current amplitude of each phase must be strictly limited to not exceeding the maximum allowable current of the equipment. The corresponding current constraint can be expressed as: (8) The average output active power of the inverter over one fundamental cycle can be expressed as:

[0038] Among them, superscript This indicates the conjugate operation.

[0039] Considering the limitations of available active power on the DC side of the inverter power supply and DC voltage safety, active power P The lower limit of active power must be met simultaneously. With active power limit The constraints are as follows: (9) The upper limit of active power is determined by the available power of the DC power supply, while the lower limit of active power is determined by the minimum active power requirement to maintain DC voltage stability.

[0040] In this embodiment, during the optimization process, the feasible domain of the decision variables is jointly limited by current constraints and active power constraints to ensure that the optimal solution neither exceeds the current limit nor the power limit.

[0041] To address the problem of supporting positive and negative dual-sequence transient voltages, this embodiment employs a weighted sum method to construct a scalar objective function, aiming to minimize the quadratic function. With the objective of determining the optimal allocation of positive and negative sequence currents, given the priority of positive and negative sequence voltage support, this optimization model can determine the optimal allocation of positive and negative sequence currents based on the positive and negative sequence grid voltages after a fault, the grid equivalent impedance, and the inverter power supply operation constraints.

[0042] Weighting coefficient and The weighting is determined based on the control priority of positive-sequence voltage support and negative-sequence voltage suppression. When both have the same control priority, they can be assigned the same weight. When it is necessary to prioritize enhancing the voltage support effect of a certain sequence component, the weighting coefficient corresponding to that objective can be increased accordingly. For example, under conditions where the voltage imbalance is small or the positive-sequence voltage drops significantly, the optimization result will tend to allocate more current to support the positive-sequence voltage; as the imbalance increases, the result will tilt towards negative-sequence voltage suppression.

[0043] Based on the above objective function and constraints, the aforementioned problem of optimal transient voltage support with positive and negative dual-order coordination can be formulated as a quadratically constrained quadratic programming (QCQP) problem, as follows: (10) in, For decision variables, corresponding , , Each component corresponds to the real part of the positive-sequence current, the real part of the negative-sequence current, the imaginary part of the positive-sequence current, and the imaginary part of the negative-sequence current in a unified complex plane with the direction of the positive-sequence grid voltage phasor as the real axis.

[0044] , These represent the positive-sequence dq-axis current components in the positive-sequence grid voltage oriented coordinate system. and These represent the negative-sequence dq-axis current components in the negative-sequence grid voltage orientation coordinate system, respectively. Represents the positive sequence grid voltage With negative sequence grid voltage The phase difference between them.

[0045] The constant term in the objective function is defined as follows: ,matrix Let Hessian matrix and vector be the objective function. This represents the coefficients of linear terms related to the equivalent parameters of the power grid. Specifically, the equivalent parameters of the power grid include the positive-sequence and negative-sequence voltage magnitudes, the phase angle difference between the positive-sequence and negative-sequence voltages, and the equivalent impedance of the power grid in the Thevenin equivalent circuit.

[0046] matrix 、 、 These are the constraint matrices corresponding to the current amplitude limits for phases A, B, and C, respectively. sum vector They jointly defined a quadratic expression for the active power constraint. This is the coefficient matrix of the quadratic term in the expression for the average active power of the inverter power supply. This is the coefficient vector of the linear terms in the active power expression. (Constant term) This indicates the maximum permissible limit of the phase current. and These correspond to the lower and upper limits of active power constraints, respectively.

[0047] The specific definitions of each matrix and vector are as follows: ; ; ; ; ; ; ; This embodiment uses a weighted sum method with the objective function of minimizing the weighted difference between the squares of the negative-sequence voltage amplitude and the squares of the positive-sequence voltage amplitude. This unifies the two coupled control objectives of positive-sequence voltage rise and negative-sequence voltage suppression into the same optimization framework, establishing a mathematical model for inverter transient voltage support that coordinates positive and negative sequence voltages. The optimization of inverter transient voltage support is transformed into a QCQP problem, which can unify the positive and negative sequence voltage support objectives, as well as operating constraints such as current and active power, into a quadratic form about the same decision vector. While retaining the nonlinear coupling characteristics of the original model, this provides a standardized mathematical foundation for subsequently using moment relaxation to transform the non-convex problem into a second-order moment relaxation problem and verify global optimality.

[0048] S102: Transform the mathematical model into a second-order moment relaxation problem with moment variables as optimization variables.

[0049] For solving the QCQP problem in formula (10) above, conventional methods require that the Hessian matrix in both the objective function and the constraints be positive semi-definite to guarantee global optimality. Due to the grid impedance... >0, matrix It contains both positive and negative eigenvalues, is an indeterminate matrix, and has a lower bound constraint on the matrix -G. P It is negative semi-definite, which makes the constraint non-convex, thus causing the QCQP problem to be non-convex. This means that even if the solver can converge to a certain stable point, the result obtained is not necessarily the global optimum.

[0050] Based on this, this embodiment proposes a global optimal solution method for non-convex QCQP problems based on moment relaxation. The original non-convex QCQP problem, with the real and imaginary parts of the positive and negative sequence currents under a unified complex plane as decision variables, is transformed into a convex relaxation problem with the moment variable y as the optimization variable. Based on the definition of moment relaxation, formula (10) can be expressed as follows: (11) in, 0 indicates that the corresponding matrix is ​​a positive semi-definite matrix. A 15×15 moment matrix (also known as a moment quantity matrix) is defined; five Each matrix corresponds to a 5×5 dimensional localization matrix, which corresponds one-to-one with the constraints in formula (10). Furthermore, Forced application The normalization condition of =1 ensures the consistency between the moment variables and the original polynomial structure.

[0051] for Define multiple exponents for nonnegative integers: Multi-index It is a vector consisting of 4 non-negative integers; the corresponding monomial is If a polynomial is written as Then we can define a linear operator: ,in It is with monomials Moment variables with one-to-one correspondence.

[0052] First, define a second-order monomial basis: , Let them be the highest second-order monomial basis vectors; then It is a second-order moment matrix. Because the four original variables and the highest second-order monomial together have One, therefore That is, 15×15. And because... The highest order monomials that can appear in the equation are 4th order, so the moment variables need to cover monomials from 0 to 4th order, and the number of these is... Therefore, there is .

[0053] The moment variable y has the following relationship with the original decision variable: ; ; ;

[0054] In other words, y here is actually the set of moment variables in the moment space of the original decision variables at each order of monomials. When relaxed to exact and satisfying the rank-1 condition, these moment variables are re-interacted with a real physical decision point x. Consistent, satisfying: .

[0055] By using moment variable lifting and semidefinite constraints, the original non-convex QCQP can be transformed into a convex semidefinite programming relaxation problem with respect to moment variables, thus enabling stable solutions using mature convex optimization algorithms. This relaxation problem also provides a lower bound for the global optimum of the original non-convex QCQP problem, and the accuracy of the relaxation can be verified by the rank condition of the optimal moment matrix. When the optimal second-order moment matrix satisfies the rank-1 condition, moment relaxation has no relaxation gaps, and the global optimum decision variables of the original QCQP can be further recovered, i.e., the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents in the unified complex plane.

[0056] This embodiment transforms the QCQP problem into a second-order convex relaxation problem with moment variables as optimization variables. By increasing the dimensionality and applying semidefinite constraints, the original non-convex problem is transformed into a convex semidefinite programming relaxation problem. When the rank condition is satisfied, global optimum can be guaranteed. At the same time, a rigorous verification mechanism—the moment matrix rank test—is provided to make the global optimum verifiable, thereby mathematically guaranteeing the global optimum of the optimal current solution under the unified complex plane.

[0057] S103: Call the semidefinite programming solver to solve the convex relaxation problem and obtain the optimal moment variables; perform a rank test on the moment matrix corresponding to the optimal moment variables, and obtain the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane by spectral decomposition when the rank-1 condition is met, and obtain the optimal reference values ​​of the positive and negative sequence dq currents by coordinate transformation.

[0058] As a specific implementation method, the specific expressions of each matrix in formula (11) can be derived from the basic theory of moment relaxation, and therefore will not be elaborated in detail. This problem is a series of semidefinite programming problems about moment variables y, which can be solved efficiently by calling conventional commercial solvers. In some specific implementation methods, a second-order moment relaxation model can be established using YALMIP in the MATLAB environment, and the MOSEK semidefinite programming solver can be called to solve it.

[0059] In this embodiment, a semidefinite programming solver is used to solve the obtained second-order moment relaxation problem to obtain the optimal moment variables. . use The components are filled according to the exponential addition rule to construct the optimal moment matrix. ;right Perform eigenvalue decomposition and estimate its numerical rank.

[0060] If satisfied This indicates that the moment relaxation is accurate, there is no relaxation gap, the relaxation problem is equivalent to the original non-convex problem, and the global optimal solution of the original problem is obtained. It can be recovered using spectral decomposition methods.

[0061] If the rank of the optimal moment matrix is ​​greater than 1, it indicates that the current second-order moment relaxation does not satisfy the rank-1 accuracy condition. In this case, it is impossible to directly recover the unique original variable vector corresponding to the moment solution through a single rank-1 spectral decomposition. Therefore, the current moment relaxation result cannot be used to directly verify and recover the global optimal decision variable of the original QCQP. In this case, the moment relaxation order can be further increased to construct a tighter higher-order moment relaxation and then solve it again.

[0062] To illustrate the recovery process of optimal decision variables from the moment matrix, we first define the monomial basis vectors in moment relaxation. For the original four-dimensional decision vector of QCQP: (12) definition For decision variables x The total number of times does not exceed The column vector formed by arranging the monomials in a preset order has the following structure: (13) Under these conditions, the recovered vector is: (14) in, A moment matrix The largest eigenvalue, Let it be its corresponding unit eigenvector. The second-order monomial vector recovered from the optimal moment matrix contains the original QCQP global optimal decision variable and the second-order monomial information formed by the decision variable.

[0063] The physical decision variables in the original QCQP problem (i.e., the real and imaginary parts of the positive and negative sequence currents in the unified complex plane) correspond to the first-order monomials with indices 2 to 5 in this vector. That is: , , , These are the optimal decision variables. Therefore, the global optimal solution... Extract as follows: (15) Obtained from equation (15) To optimize the optimal current solution in the unified complex plane, it cannot be directly used as a reference command for the inverter power supply current controller. It needs to be further transformed to a unified control coordinate system with the PCC positive sequence voltage as the d-axis. The coordinate rotation matrix is ​​defined as follows: ; The optimal current solution is then transformed to positive and negative sequence grid voltage-oriented coordinate systems, respectively. Since the real axis of the unified complex plane is aligned with the positive sequence grid voltage phasor, the positive sequence current does not require rotation, while the negative sequence current follows the phase angle difference between the positive and negative sequence grid voltages. After rotating, we get: ; in, and These represent the optimal components of the positive and negative sequence currents in the corresponding grid voltage-oriented coordinate systems. Further transformation to the corresponding positive and negative sequence PCC voltage-oriented coordinate systems is then performed. ; in, These represent the phase angle differences between the positive and negative sequence voltages of the PCC and the corresponding positive and negative sequence grid voltages, respectively.

[0064] Considering that the inverter power supply uses a phase-locked loop to track the positive-sequence voltage phase angle of the PCC, the optimal current component in the negative-sequence PCC voltage-oriented dq coordinate system needs to be further transformed to a coordinate system with the PCC positive-sequence voltage as the d-axis, that is: ; in, Represents the angular displacement between the negative-sequence PCC voltage orientation coordinate system and the positive-sequence PCC voltage orientation coordinate system; matrix This is used to correct coordinate differences caused by different rotation directions of positive and negative order rotating coordinate systems. The result obtained after the above transformation... and These serve as the optimal reference commands for the positive-sequence current control inner loop and the negative-sequence current control inner loop, respectively.

[0065] S104: Outputs the optimal current command to the inverter's inner current loop controller to control the inverter to simultaneously boost the positive sequence voltage and suppress the negative sequence voltage during grid faults.

[0066] The optimal positive and negative sequence dq current reference values ​​obtained through coordinate transformation in step S103 are output to the inner current loop controller in the PLL-based cascaded vector control structure of the inverter power supply. The inner current loop tracks the positive and negative sequence current reference values ​​separately through dual control methods, realizing decoupled control of the positive and negative sequence currents. This allows the inverter power supply to simultaneously boost the positive sequence voltage and suppress the negative sequence voltage during grid faults, maximizing the transient voltage support effect.

[0067] After a grid fault occurs, the system parameters required for steps S101 to S104 above (including positive sequence grid voltage, negative sequence grid voltage, phase difference and equivalent grid impedance) are estimated by sampling the three-phase voltage and three-phase current at the PCC in real time. The estimation time is about 50ms after the fault occurs. After that, the control method starts and continuously outputs the optimal current command.

[0068] This embodiment takes into account both the three-phase current amplitude constraint and the upper and lower limits of active power constraint, and incorporates the contribution of active current to voltage support into the optimization model. Even in low impedance ratio X / R scenarios, it can fully utilize the support capability of the inverter power supply and maximize the utilization of the inverter power supply capacity.

[0069] This embodiment considers a three-phase three-wire grid-connected system, so the zero-sequence current component is not considered; the inverter power supply is configured with positive and negative sequence decoupled current control; the external grid of the PCC adopts the Thevenin equivalent model, and it is assumed that the positive and negative sequence equivalent grid impedances are the same, uniformly expressed as Z=R+jX; there are no special requirements for the energy type on the DC side, and it can be applied to grid-connected inverter power supplies with the above control structure, such as photovoltaic, wind power, and energy storage.

[0070] The inverter power supply control method proposed in this embodiment, which provides optimal positive and negative dual-sequence transient voltage support, overcomes the fixed priority problem of positive and negative sequence voltage support in existing optimization models. It can adaptively adjust the current distribution ratio of positive and negative sequence support according to the severity of disturbance. In addition, the proposed solution method based on moment relaxation can guarantee the global mathematical optimality of the obtained solution. In engineering implementation, it can maximize the utilization of the inverter power supply's support capacity while ensuring that various safety and stability constraints of the inverter power supply are not exceeded.

[0071] To verify the effectiveness of the proposed control method, this invention uses a single-inverter power supply grid-connected test system for simulation analysis. A simulation model is built in MATLAB / Simulink, and the system topology is as follows: Figure 2As shown, the positive and negative sequence voltages of the power grid are set to change to different degrees. To demonstrate the superiority of the proposed method, it is compared with existing methods that maximize the voltage difference ( A comparison is made between typical ODVS control schemes with the objective of ).

[0072] Four representative fault scenarios were selected: Scenario 1: Only the maximum current constraint is a tight constraint; Scenario 2: Both the maximum current constraint and the maximum active power constraint are tight constraints; Scenario 3: Both the maximum current constraint and the minimum active power constraint are tight constraints. Scenario 4: Only the maximum active power constraint is a tight constraint.

[0073] Maximum current limit in all fault scenarios All are set to 1.5 pu, lower limit of active power. The maximum active power is 0.1 pu. Depending on the scenario: 1.0 pu is used in scenario 1 / 3, and 0.2 pu is used in scenario 2 / 4. In all fault scenarios, the unbalanced voltage sag occurs at... =2.0s, the proposed method starts 50 ms after the fault occurs, corresponding to the time required for estimating the equivalent parameters of the power grid.

[0074] Fault Scenario 1: Only the maximum current constraint is a tight constraint: In scenario 1, the system parameter is set to the short-circuit ratio. impedance ratio The positive and negative sequence grid voltages are respectively and The time-domain dynamic simulation results of the corresponding output current, active power, and sequence voltage are as follows: Figure 4 The left half is shown.

[0075] In the time-domain simulation results, the current amplitude of phase A is 0.2715 pu, while the currents of phases B and C remain at their upper limits. The positive-sequence voltage was raised to 0.3870 pu, the negative-sequence voltage was suppressed to 0.3625 pu, and the active power output was 0.2226 pu. None of the physical quantities exceeded the constraints, and the time-domain simulation results were in high agreement with the theoretical calculations, verifying the effectiveness and correctness of the moment relaxation-based solution algorithm and its theoretical analysis.

[0076] In contrast, the ODVS control strategy can boost the positive-sequence voltage to 0.3486 pu and suppress the negative-sequence voltage to 0.3318 pu, corresponding to an active power output of 0.2227 pu, without violating any constraints. Although this strategy can significantly improve the voltage support capability of the inverter, its performance is inferior to the proposed strategy under the quadratic objective function evaluation adopted in this invention, which can be reflected by the comparison of objective function values ​​(ODVS is -0.0114, while the proposed method is -0.0183). Furthermore, even when maximizing ODVS itself in the design... Under the target conditions, the result obtained (0.0168) is still less than the result corresponding to the proposed method (0.0245).

[0077] Fault Scenario 2: Both the maximum current constraint and the maximum active power constraint are tight constraints: Take in Scenario 2 =0.2 pu, setting =5, =5, the positive and negative sequence grid voltages are respectively and The corresponding optimal solution makes both the maximum current constraint and the maximum active power constraint tight constraints.

[0078] Under this condition, the theoretically optimal result obtained by moment relaxation is: =0.2157 pu, = = 1.5 pu, the corresponding positive and negative sequence PCC voltages are respectively =0.4312 pu、 =0.4089 pu, active power output The value is 0.2 pu, and the objective function value is -0.0187. The time-domain dynamic simulation results of the corresponding output current, active power, and sequence voltage are as follows: Figure 4 The right half of the diagram is shown. In the simulation, the A-phase current is 0.2262 pu, while the B and C-phase currents remain at their maximum limits; the positive-sequence voltage is increased to 0.4314 pu, the negative-sequence voltage is suppressed to 0.4085 pu, and the active power is stably maintained at 0.2 pu. Both the tightness of the constraints and the values ​​of each physical quantity are highly consistent with the theoretical calculations.

[0079] In contrast, the ODVS control strategy raises the positive-sequence voltage to 0.4527 pu and suppresses the negative-sequence voltage to 0.4352 pu. Although its positive-sequence voltage increase is slightly higher than the proposed method, under the quadratic objective function, its corresponding objective function value is -0.0155, which is significantly worse than the proposed method. Under its linear objective, ODVS achieves a value of 0.0175, while the proposed method reaches 0.0229, still demonstrating superior performance. Furthermore, under this operating condition, ODVS results in an active power output of 0.29 pu, exceeding the upper limit of available active power.

[0080] Fault Scenario 3: Both the maximum current constraint and the minimum active power constraint are tight constraints: In failure scenario 3 =5、 =10, the positive and negative sequence grid voltages are respectively and The time-domain simulation results for the corresponding output current, active power, and sequence voltage are as follows: Figure 5 The left half is shown.

[0081] In this fault scenario, theoretical analysis based on moment relaxation yields the following results. =1.2852 pu, the current of the other two phases is at the upper limit. The corresponding optimal positive and negative sequence voltages are respectively =0.3839 pu, =0.2775pu, and the active power output is limited to the lower limit. The above theoretical results deviate very little from the simulation results, and the optimal quadratic objective function value is -0.0704, and the linear objective function value is 0.1064.

[0082] Under ODVS control, the positive and negative sequence voltages are respectively =0.2316 pu and =0.1332 pu, corresponding to a quadratic objective function value of -0.0359 and a linear objective function value of 0.0984. The results show that, regardless of whether the quadratic objective function is minimized or the linear objective function is maximized, the method proposed in this invention exhibits better voltage support performance than ODVS.

[0083] Fault Scenario 4: Only the maximum active power constraint is a tight constraint: In this scenario, the optimal solution only results in a tight active power constraint, and the output current does not reach the limit. In scenario 4, the parameters are set as follows: =0.2 pu, =4、 =0.5, the positive and negative sequence grid voltages are respectively and The time-domain simulation results for the corresponding output current, active power, and sequence voltage are as follows: Figure 5 The right half is shown.

[0084] Under this operating condition, the three-phase current amplitudes corresponding to the optimal solution obtained by moment relaxation are respectively = 0.9997 pu, =0.5954 pu, = 1.0901 pu, the positive and negative sequence current amplitudes are respectively = 0.8739 pu、 = 0.2887pu, with a phase difference of 72.56°. The corresponding positive and negative sequence voltages are respectively =0.3782 pu、 =0.0328 pu, active power output is = 0.2 pu, the quadratic objective function value is -0.1420. By checking the rank of the moment matrix, it can be confirmed that the solution still satisfies the rank-1 condition, is the global optimal solution, and the deviation between the theoretical result and the time-domain simulation result is still very small.

[0085] In contrast, since the ODVS strategy only considers peak current constraints during design, an infeasible solution will be obtained in this scenario. (Theoretical calculations yield...) =0.3821 pu、 = 0.1440 pu; In the time-domain simulation, using the reference command given by ODVS, the actual voltage obtained is = 0.3778 pu, = 0.1149 pu, corresponding to a quadratic objective function value of -0.1253 and a linear objective function value of 0.2381, both inferior to the results of the proposed method. Furthermore, the active power output caused by ODVS reaches as high as 0.4313 pu, far exceeding the safety limit. = 0.2 pu, and because the inverter cannot accurately track the given current reference value, it severely aggravates power oscillation, posing a potential risk to the safe and stable operation of the inverter.

[0086] In summary, the control method proposed in this embodiment outperforms existing ODVS control schemes under all the above-mentioned fault scenarios. Under the premise of strictly satisfying the three-phase current amplitude constraints and the upper and lower limits of active power constraints, it can achieve the mathematical global optimum of the positive and negative dual-sequence transient voltage support effect, showing better transient voltage support performance, and verifying the effectiveness and superiority of the method in this embodiment.

[0087] Example 2 In one or more embodiments, an inverter power supply control system providing positive and negative dual-sequence transient voltage support is disclosed, specifically including: The model building module is used to establish a mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative sequence voltages, using the real and imaginary components of the positive and negative sequence currents injected by the inverter power supply at the point of common coupling as decision variables and minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling as the objective function. The model transformation module is used to transform the mathematical model into a second-order moment relaxation problem with moment variables as optimization variables; The model solving module is used to call the semidefinite programming solver to solve the convex relaxation problem and obtain the optimal moment variables; the rank test is performed on the moment matrix corresponding to the optimal moment variables, and the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are obtained by spectral decomposition when the rank-1 condition is satisfied. The coordinate transformation and inverter power supply control module is used to convert the optimal values ​​of the real and imaginary components of the positive and negative sequence currents into reference values ​​of the positive and negative sequence dq currents with the positive sequence voltage of PCC as the d-axis, and output them to the current inner loop controller of the inverter power supply to control the inverter power supply to simultaneously raise the positive sequence voltage and suppress the negative sequence voltage during grid faults.

[0088] It should be noted that the specific implementation methods of the above modules are exactly the same as those in Example 1, and will not be described in detail again.

[0089] Example 3 In one or more embodiments, a terminal device is disclosed, comprising a processor and a memory, wherein the processor is used to implement instructions; and the memory is used to store multiple instructions adapted to be loaded by the processor and executed by the processor to provide positive and negative dual-sequence transient voltage support for the inverter power supply control method described in Embodiment 1.

[0090] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any other conventional processor.

[0091] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of the memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0092] In the implementation process, each step of the above method can be completed by the integrated logic circuits in the processor hardware or by software instructions.

[0093] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for controlling an inverter power supply that provides positive and negative dual-sequence transient voltage support, characterized in that, include: Using the real and imaginary components of the positive and negative sequence currents injected by the inverter at the point of common coupling as decision variables in a unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis, and minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling as the objective function, a mathematical model for the transient voltage support of the inverter with coordinated positive and negative sequence voltages is established. The mathematical model is transformed into a second-order moment relaxation problem with moment variables as optimization variables; The second-order moment relaxation problem is solved by calling a semidefinite programming solver to obtain the optimal moment variables; the rank test is performed on the moment matrix corresponding to the optimal moment variables, and when the rank-1 condition is met, the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are obtained by spectral decomposition, and the optimal reference values ​​of the positive and negative sequence dq currents are obtained by coordinate transformation; The optimal positive and negative sequence dq current reference values ​​are output to the positive and negative sequence current inner loop controllers of the inverter power supply, respectively, so as to control the inverter power supply to simultaneously raise the positive sequence voltage and suppress the negative sequence voltage during grid faults.

2. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 1, characterized in that, The mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative dual-sequence voltages is a quadratic programming problem with quadratic constraints, using the real and imaginary parts of the positive and negative sequence currents under a unified complex plane as decision variables.

3. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 1, characterized in that, The process of constructing the mathematical model for the transient voltage support of the inverter power supply with positive and negative dual-sequence voltage coordination includes: In the positive and negative sequence dq rotating coordinate system, based on Kirchhoff's voltage law, the relationship between the positive sequence voltage at PCC and the positive sequence dq current component in the positive sequence grid voltage oriented dq coordinate system, and the relationship between the negative sequence voltage at PCC and the negative sequence dq current component in the negative sequence grid voltage oriented dq coordinate system are established respectively. By defining new variables, the negative sequence current is rotated from the negative sequence grid voltage orientation dq coordinate system to a unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis, while keeping the positive sequence current unchanged, so that the positive sequence and negative sequence current components are in the same complex plane. Based on the new variables, quadratic expressions for the magnitudes of the positive and negative sequence voltages at the PCC, as well as the magnitudes of the three-phase current, are given again; and constraints on the magnitudes of the three-phase current and the upper and lower limits of the active power are established. Based on the given expression and constraints, a mathematical model for transient voltage support of inverter power supply with coordinated positive and negative dual-sequence voltages is constructed.

4. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 1, characterized in that, The three-phase current amplitude constraint is expressed by establishing quadratic inequalities about the decision variables for the squares of the current amplitudes of phases A, B, and C, respectively, requiring that the current amplitude of each phase does not exceed the maximum allowable current limit of the inverter power supply. The active power upper and lower limit constraints are established by creating a quadratic equation for the output active power of the inverter power supply with respect to the decision variables, requiring the output active power to be between the upper and lower limits of the active power.

5. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 1, characterized in that, The specific mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative dual-sequence voltages is as follows: ; in, For decision variables, corresponding , , Each component corresponds to the real part of the positive-sequence current, the real part of the negative-sequence current, the imaginary part of the positive-sequence current, and the imaginary part of the negative-sequence current in a unified complex plane with the direction of the positive-sequence grid voltage phasor as the real axis. , These represent the positive-sequence dq-axis current components in the positive-sequence grid voltage oriented coordinate system. and These represent the negative-sequence dq-axis current components in the negative-sequence grid voltage orientation coordinate system, respectively. Represents the positive sequence grid voltage With negative sequence grid voltage The phase difference between them; Let Hessian matrix be the objective function. 、 、 These are the constraint matrices corresponding to the current amplitude limits for phases A, B, and C, respectively. For constant terms, , , These are the weighting coefficients; , , This indicates the maximum permissible limit for the corresponding phase current; , represents the lower and upper limits of active power constraints, respectively; g is the coefficient vector of the linear term in the objective function, and Gp is ​​the coefficient matrix of the quadratic term in the expression for the average active power of the inverter power supply. This is the coefficient vector of the linear terms in the active power expression.

6. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 1, characterized in that, The rank test is performed on the moment matrix corresponding to the optimal moment variable, specifically as follows: The moment matrix is ​​filled with the components of the optimal moment variables according to the rule of adding monomial exponents; If the rank of the moment matrix is ​​1, then the relaxation is confirmed to be accurate and can be restored to the global optimal solution of the original QCQP. The moment matrix is ​​subjected to spectral decomposition to extract the first-order monomial components, and the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are recovered.

7. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 6, characterized in that, The moment matrix is ​​subjected to spectral decomposition, specifically as follows: Calculate the maximum eigenvalue of the moment matrix and its corresponding unit eigenvector. Use the product of the square root of the maximum eigenvalue and the unit eigenvector as the recovery vector. Extract the first-order monomial components with indices 2 to 5 from the recovery vector to obtain the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents in the unified complex plane with the direction of the positive sequence grid voltage phasor as the real axis.

8. The inverter power supply control method for providing positive and negative dual-sequence transient voltage support as described in claim 7, characterized in that, The real and imaginary components of the positive and negative sequence currents of the optimal current are rotated to transform them into a unified coordinate system with the positive sequence voltage of PCC as the d-axis, so as to obtain the optimal positive and negative sequence dq current reference values. The current reference values ​​are then provided to the positive and negative sequence current inner loop controllers, respectively.

9. An inverter power supply control system that provides positive and negative dual-sequence transient voltage support, characterized in that, include: The model building module is used to establish a mathematical model for the transient voltage support of the inverter power supply with coordinated positive and negative sequence voltages, using the real and imaginary components of the positive and negative sequence currents injected by the inverter power supply at the point of common coupling as decision variables and minimizing the weighted difference between the square of the negative sequence voltage amplitude and the square of the positive sequence voltage amplitude at the point of common coupling as the objective function. The model transformation module is used to transform the mathematical model into a second-order moment relaxation problem with moment variables as optimization variables; The model solving module is used to call the semidefinite programming solver to solve the second-order moment relaxation problem and obtain the optimal moment variables; the rank test is performed on the moment matrix corresponding to the optimal moment variables, and when the rank-1 condition is met, the optimal values ​​of the real and imaginary parts of the positive and negative sequence currents under the unified complex plane are obtained through spectral decomposition. The coordinate transformation and inverter power supply control module is used to convert the optimal values ​​of the real and imaginary components of the positive and negative sequence currents into reference values ​​of the positive and negative sequence dq currents with the positive sequence voltage of PCC as the d-axis, and output them to the current inner loop controller of the inverter power supply to control the inverter power supply to simultaneously raise the positive sequence voltage and suppress the negative sequence voltage during grid faults.

10. A terminal device comprising a processor and a memory, the processor for implementing instructions; the memory for storing multiple instructions, characterized in that, The instructions are adapted to be loaded by a processor and executed as described in any one of claims 1-8, for providing positive and negative dual-sequence transient voltage support for inverter power supply.