A controller parameter optimization method and system based on transient stability region maximization

Abstract

The application provides a controller parameter optimization method and system based on transient stability domain maximization, and the method comprises the following steps: using a virtual synchronous generator to control a network type controller, and obtaining operation data of the network type controller; constructing a multi-machine system state equation according to the operation data of the network type controller; using a sum of squares programming algorithm to process the multi-machine system state equation, and obtaining a Lyapunov function for expanding the stability domain; using a simulation internal domain expansion parameter optimization algorithm to process the Lyapunov function for expanding the stability domain, and constructing an optimized Lyapunov function; and redesigning parameters of the network type controller according to the optimized Lyapunov function. The application realizes expansion of the transient stability domain, and improves the stability of the power system where the controller is located.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of controller parameter design, in particular to a controller parameter optimization method and system based on transient stability domain maximization. BACKGROUND

[0002] As the proportion of new energy in the energy supply system gradually rises, the original grid structure dominated by synchronous machines has undergone substantial changes, and power electronic equipment dominated by voltage source converters has shown a high penetration trend. It is of great value to study the stability mechanism and control technology of voltage source converters under the background of widespread application of power electronic equipment. However, the application of a high proportion of power electronic equipment in the power system has qualitatively changed the physical layer of the power system, resulting in a decrease in the physical entity of the rotor, a significant decrease in stored kinetic energy, and a deterioration in the system's ability to resist disturbances. Under large disturbances, the dynamic behavior and mechanism of power electronic equipment and traditional transient stability research conclusions cannot provide technical support for energy supply.

[0003] Classic parameter optimization methods such as eigenvalue analysis based on state space equations, sequence impedance model analysis, and automatic control theory based on open-loop transfer functions generally aim to increase energy dissipation, speed up transient response, and slow down the change in angular velocity under large disturbances. However, parameter design from the perspective of increasing the transient stability domain faces many challenges. Estimating the transient stability domain of a multi-machine system usually uses the Lyapunov energy function, but the non-uniqueness of the Lyapunov function increases the difficulty of solving the Lyapunov function of a multi-machine system and the difficulty of solving the boundary of the maximum stability domain. Especially for multi-machine systems containing power electronic converters, traditional algorithms such as double-step search and golden section method have defects such as large computational load and low computational efficiency when applied to the square sum programming algorithm for solving the boundary of the maximum stability domain. SUMMARY

[0004] The present application provides a controller parameter optimization method and system based on transient stability domain maximization, which maximizes the transient stability domain and improves the stability of the power system where the controller is located.

[0005] To solve the above technical problems, the present application provides a controller parameter optimization method based on transient stability domain maximization, comprising:

[0006] A virtual synchronous generator control network type controller is used, and the operating data of the network type controller is obtained; wherein the operating data of the network type controller includes the active power and reactive power at the point of common coupling of the network type controller;

[0007] According to the operation data of the network configuration type controller, a multi-machine system state equation is constructed; wherein the multi-machine system state equation comprises a synchronous machine swing equation, a power angle differential equation and a speed differential equation;

[0008] The multi-machine system state equation is processed using a sum of squares programming algorithm to obtain a Lyapunov function of enlarged stability domain;

[0009] The Lyapunov function of enlarged stability domain is processed using an analog inner domain expansion parameter optimization algorithm to construct an optimized Lyapunov function; wherein the analog inner domain expansion parameter optimization algorithm comprises an inner loop algorithm and an outer loop algorithm;

[0010] According to the optimized Lyapunov function, the parameters of the network configuration type controller are redesigned.

[0011] The application firstly constructs a state space equation of a multi-machine system, which can correctly reflect the kinematic equation of the virtual synchronous generator control of dynamic response; then, in the multi-machine system, a Lyapunov function of enlarged stability domain is obtained using a sum of squares programming algorithm, and the Lyapunov function parameters and the Lyapunov function of the maximum stability domain are solved using an analog inner domain expansion parameter optimization algorithm; finally, according to the Lyapunov function of the maximum stability domain, the parameters of the network configuration type controller in the multi-machine system are optimized and the transient stability domain is designed; thereby the maximum transient stability domain is realized, and the stability of the power system with the controller is improved.

[0012] Further, the construction of the multi-machine system state equation according to the operation data of the network configuration type controller comprises:

[0013] According to the active power and the reactive power of the common coupling point of the network configuration type controller, the synchronous phase angle and the voltage amplitude output by the virtual synchronous generator are obtained to obtain the synchronous machine swing equation;

[0014] The damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator are obtained, and the power angle differential equation and the speed differential equation are constructed according to the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator; wherein the power angle differential equation and the speed differential equation are as follows:

[0015]

[0016] In the formula, represents the virtual power angle; f represents the standard frequency, and Δω represents the virtual rotor angular velocity deviation unit value; represents the derivative of the virtual rotor angular velocity deviation unit value; w p = D / M, K p= 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, P eref represents a given reference active power, P e represents a converter grid-connected point active power;

[0017] In combination with the synchronous machine swing equation, the power angle differential equation and the speed differential equation, a multi-machine system state equation is constructed; wherein the multi-machine system state equation is specifically as follows:

[0018]

[0019]

[0020]

[0021]

[0022]

[0023] In the formula, represents a virtual power angle of the i th unit relative to the j th unit; ω i represents a virtual rotor speed of the i th unit; represents a derivative of the virtual rotor speed of the i th unit; H i represents an inertia time constant of the i th unit; w p = D / M, K p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient; P eref represents a given reference active power; P ei represents an active power output by the i th unit; P mi represents a mechanical power input by the i th unit, D i represents a virtual generator damping coefficient of the i th unit.

[0024] The application firstly establishes the relationship among power angle, speed and active power through a virtual synchronous generator control grid type controller control loop, constructs a synchronous machine swing equation, a power angle differential equation and a speed differential equation, obtains a multi-machine system state equation, and can correctly reflect the dynamic response of the multi-machine system and the grid type controller.

[0025] Further, the multi-machine system state equation is processed by using a sum of squares programming algorithm to obtain a Lyapunov function with an expanded stability domain, comprising:

[0026] An initial Lyapunov function is obtained by using a semi-definite programming solver to calculate the multi-machine system state equation;

[0027] The boundary value of the initial Lyapunov function is calculated using a forward search method and a dichotomy method to obtain a boundary value of a transient stability domain and a boundary value normalized Lyapunov function.

[0028] The boundary value normalized Lyapunov function is cyclically updated until the solution quality of the updated Lyapunov function is lower than a preset threshold, and the cycle is stopped to obtain an enlarged stability domain Lyapunov function; wherein in each cycle, a Lyapunov function coefficient polynomial is calculated according to the multi-machine system state equation and the boundary value normalized Lyapunov function; under the constraint condition that the outer domain contains the inner domain, a candidate Lyapunov function is constructed in combination with the coefficient polynomial and the boundary value normalized Lyapunov function; if the solution quality of the candidate Lyapunov function in the current cycle is lower than the preset threshold, the candidate Lyapunov function in the current cycle is taken as the enlarged stability domain Lyapunov function.

[0029] The application uses a semi-definite programming solver to automatically search the multi-machine system state equation, obtain an initial Lyapunov function, and then performs boundary value normalization processing and cyclic updating on the initial Lyapunov function to obtain an enlarged stability domain Lyapunov function, thereby improving the calculation efficiency of the Lyapunov function, saving the calculation cost, simplifying the calculation algorithm, and avoiding unnecessary consumption of computing resources.

[0030] Further, the enlarged stability domain Lyapunov function is processed using a simulated inner domain expansion parameter optimization algorithm to construct an optimized Lyapunov function, including:

[0031] The enlarged stability domain Lyapunov function is processed using an inner loop algorithm to obtain first Lyapunov function parameters;

[0032] The first Lyapunov function parameters are iteratively optimized using an outer loop algorithm to obtain second Lyapunov function parameters;

[0033] The enlarged stability domain Lyapunov function is parameter optimized according to the second Lyapunov function parameters to obtain an optimized Lyapunov function.

[0034] Further, the enlarged stability domain Lyapunov function is processed using an inner loop algorithm to obtain first Lyapunov function parameters, including:

[0035] The parameters of the enlarged stability domain Lyapunov function are set as to-be-solved parameters to construct a state space equation containing the to-be-solved parameters;

[0036] A first constraint condition containing the state space equation and the Lyapunov function stability condition is constructed; wherein the first constraint condition is specifically as follows:

[0037]

[0038] wherein s2 and s4 represent polynomial coefficients to be solved; V represents a Lyapunov function with expanded stability domain, f represents a state space equation containing parameters to be solved; G represents equality constraints under trigonometric coordinate transformation; and l2 represent sum-of-squares polynomials; ∑ M represent sum-of-squares polynomials;

[0039] Under the constraint of the first constraint condition, the state space equation containing parameters to be solved is solved by a solver to obtain first Lyapunov function parameters.

[0040] Further, the first Lyapunov function parameters are iteratively optimized according to an outer loop algorithm to obtain second Lyapunov function parameters, which comprises:

[0041] A second constraint condition is constructed in combination with the constraint condition that an outer domain contains an inner domain, the Lyapunov function stability condition and the trigonometric function transformation constraint condition, wherein the second constraint condition is specifically as follows:

[0042]

[0043] wherein V represents a Lyapunov function to be solved; p represents a Lyapunov function after last round of iterative optimization; and l i (o = 1, 2, 3) represent sum-of-squares polynomials, G i (i = 1, 2, 3) represent equality constraints under trigonometric coordinate transformation, s1, s2 and s4 represent known polynomial coefficients; f represents a state space equation containing Lyapunov function parameters after last round of iterative optimization; ∑ M represent sum-of-squares polynomials;

[0044] The first Lyapunov function parameters are iteratively optimized using an outer loop algorithm; until the Lyapunov function parameters after current round of iterative optimization meet a preset iteration stop condition, the iterative optimization is stopped, and the Lyapunov function parameters after current round of iterative optimization are output as second Lyapunov function parameters;

[0045] In the first round of iterative optimization, the first Lyapunov function parameter is optimized under the constraint of the second constraint condition, to obtain a first round of iterative optimization Lyapunov function parameter; in the iterative optimization after the first round, the Lyapunov function parameter of the last round of iterative optimization is optimized under the constraint of the second constraint condition, to obtain a current iterative optimization Lyapunov function parameter; the preset iteration stop condition is that the current iterative optimization Lyapunov function parameter meets the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameter of the last round of iterative optimization and the current iterative optimization Lyapunov function parameter is less than the preset accuracy value.

[0046] The application is based on a virtual synchronous generator square sum planning algorithm, and an analog inner domain expansion parameter optimization algorithm is designed by combining an inner loop algorithm and an outer loop algorithm. The Lyapunov function parameter of the optimal transient stability domain boundary and the maximum stability domain is solved by using the analog inner domain expansion parameter optimization algorithm, so that the compatibility and universality of the Lyapunov function parameter optimization method are improved. The optimal transient stability domain boundary can quantitatively describe the stability margin of the working point before and after the disturbance, and is applied to online dynamic security analysis and monitoring of a large power system containing power electronic equipment.

[0047] On the basis of the above-mentioned method embodiment, the application correspondingly provides a system embodiment, and provides a controller parameter optimization system based on transient stability domain maximization, which comprises a running data module, a state equation module, a square sum planning module, a parameter optimization module and a controller optimization module.

[0048] The running data module is used for using a virtual synchronous generator control network type controller and acquiring running data of the network type controller; wherein the running data of the network type controller comprises common coupling point active power and common coupling point reactive power of the network type controller.

[0049] The state equation module is used for constructing a multi-machine system state equation according to the running data of the network type controller; wherein the multi-machine system state equation comprises a synchronous machine swing equation, a power angle differential equation and a speed differential equation.

[0050] The square sum planning module is used for processing the multi-machine system state equation by using a square sum planning algorithm, to obtain a stability domain expansion Lyapunov function.

[0051] The parameter optimization module is used for processing the stability domain expansion Lyapunov function by using an analog inner domain expansion parameter optimization algorithm, to construct an optimized Lyapunov function; wherein the analog inner domain expansion parameter optimization algorithm comprises an inner loop algorithm and an outer loop algorithm.

[0052] The controller optimization module is configured to redesign parameters of the networked controller according to an optimized Lyapunov function.

[0053] Further, the state equation module comprises a first equation unit, a second equation unit and a third equation unit.

[0054] The first equation unit is configured to obtain a synchronous machine swing equation by outputting a synchronous phase angle and a voltage amplitude of the virtual synchronous generator according to active power and reactive power of a point of common coupling of the networked controller.

[0055] The second equation unit is configured to obtain a damping coefficient of the virtual synchronous generator and an inertia time coefficient of the virtual synchronous generator, and construct a power angle differential equation and a speed differential equation according to the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator; wherein the power angle differential equation and the speed differential equation are as follows:

[0056]

[0057] In the formula, ω represents a virtual power angle; f represents a standard frequency, and Δω represents a virtual rotor angular velocity deviation unit value; represents a virtual rotor angular velocity deviation unit value; w p = D / M, K p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, P eref represents a given reference active power, P e represents a converter grid-connected point active power.

[0058] The third equation unit is configured to construct a multi-machine system state equation in combination with the synchronous machine swing equation, the power angle differential equation and the speed differential equation; wherein the multi-machine system state equation is as follows:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064] In the formula, ω represents a virtual power angle of the i th unit relative to the j th unit; ω i represents an i th virtual rotor angular velocity, represents the derivative of the i-th virtual rotor angular velocity, i.e. rotor angular acceleration; H i represents the inertia time constant of the i-th unit; w p = D / M, K p = 1 / D, D represents the virtual generator damping coefficient, M represents the virtual generator inertia time coefficient, P eref represents the given reference active power, P ei represents the active power output by the i-th unit; P mi represents the mechanical power input by the i-th unit, D i represents the i-th virtual generator damping coefficient.

[0065] Further, the sum-of-squares programming module comprises: an initial function unit, a normalization function unit and a cyclic updating unit;

[0066] The initial function unit is configured to calculate the state equation of the multi-machine system using a semi-definite programming solver to obtain an initial Lyapunov function.

[0067] The normalization function unit is configured to calculate the boundary value of the initial Lyapunov function using a forward search method and a bisection method to obtain a boundary value of a transient stability region and a boundary value normalized Lyapunov function.

[0068] The cyclic updating unit is configured to cyclically update the boundary value normalized Lyapunov function until the solution quality of the updated Lyapunov function is lower than a preset threshold, and stop the cycle to obtain a Lyapunov function with an expanded stability region; wherein, in each cycle, a coefficient polynomial of the Lyapunov function is calculated according to the state equation of the multi-machine system and the boundary value normalized Lyapunov function; a candidate Lyapunov function is constructed in combination with the coefficient polynomial and the boundary value normalized Lyapunov function under the constraint condition that the outer region contains the inner region; and if the solution quality of the candidate Lyapunov function in the current cycle is lower than the preset threshold, the candidate Lyapunov function in the current cycle is taken as the Lyapunov function with the expanded stability region.

[0069] Further, the parameter optimization module comprises: an inner loop unit, an outer loop unit and an optimization function unit.

[0070] The inner loop unit is configured to process the Lyapunov function with the expanded stability region using an inner loop algorithm to obtain first Lyapunov function parameters.

[0071] The outer loop unit is configured to iteratively optimize the first Lyapunov function parameters using an outer loop algorithm to obtain second Lyapunov function parameters.

[0072] The optimization function unit is configured to perform parameter optimization on the Lyapunov function with expanded stability domain according to the second Lyapunov function parameter, and obtain an optimized Lyapunov function.

[0073] Further, the inner loop unit comprises a variable setting subunit, a first constraint subunit and a variable solving subunit.

[0074] The variable setting subunit is configured to set the parameter of the Lyapunov function with expanded stability domain as a to-be-solved parameter, and construct a state space equation containing the to-be-solved parameter.

[0075] The first constraint subunit is configured to construct a first constraint condition containing the state space equation and a Lyapunov function stability condition; and the first constraint condition is specifically as follows:

[0076]

[0077] In the formula, s2 and s4 represent polynomial coefficients to be solved. V represents the Lyapunov function with expanded stability domain, f represents the state space equation containing the to-be-solved parameter, G represents an equality constraint under a trigonometric coordinate transformation, and l2 represents a sum-of-squares polynomial. In the formula, s2 and s4 represent polynomial coefficients to be solved. M In the formula, s2 and s4 represent polynomial coefficients to be solved.

[0078] The variable solving subunit is configured to solve the state space equation containing the to-be-solved parameter by a solver under the constraint of the first constraint condition, and obtain a first Lyapunov function parameter.

[0079] Further, the outer loop unit comprises a second constraint subunit and an outer loop optimization subunit.

[0080] The second constraint subunit is configured to construct a second constraint condition in combination with a constraint condition in which an outer domain contains an inner domain, a Lyapunov function stability condition and a trigonometric function transformation constraint condition; and the second constraint condition is specifically as follows:

[0081]

[0082] In the formula, V represents a to-be-solved Lyapunov function, and p represents a Lyapunov function after optimization in a previous round of iteration. In the formula, V represents a to-be-solved Lyapunov function, and p represents a Lyapunov function after optimization in a previous round of iteration. i (o=1, 2, 3) represents a sum-of-squares polynomial, G i (i=1, 2, 3) represents an equality constraint under a trigonometric coordinate transformation, and s1, s2 and s4 represent known polynomial coefficients. f represents a state space equation comprising the Lyapunov function parameter optimized in the last round of iteration; and ∑ M represents a sum of squares polynomial.

[0083] The outer loop optimization subunit is configured to use an outer loop algorithm to iteratively optimize the first Lyapunov function parameter until the Lyapunov function parameter after the current iteration optimization meets a preset iteration stop condition, and stop the iteration optimization, and output the Lyapunov function parameter after the current iteration optimization as the second Lyapunov function parameter.

[0084] In the first round of iteration optimization, the first Lyapunov function parameter is optimized under the constraint of the second constraint condition to obtain the Lyapunov function parameter after the first round of iteration optimization; in the iteration optimization after the first round, the Lyapunov function parameter after the last round of iteration optimization is optimized under the constraint of the second constraint condition to obtain the Lyapunov function parameter after the current iteration optimization; and the preset iteration stop condition is that the Lyapunov function parameter after the current iteration optimization meets the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameter after the last round of iteration optimization and the Lyapunov function parameter after the current iteration optimization is less than a preset accuracy value. BRIEF DESCRIPTION OF DRAWINGS

[0085] Figure 1 A flowchart of an embodiment of the controller parameter optimization method based on transient stability domain maximization provided by the present application;

[0086] Figure 2 A structural diagram of an embodiment of the new energy unit system topology provided by the present application;

[0087] Figure 3 A structural diagram of an embodiment of the virtual synchronous generator provided by the present application;

[0088] Figure 4 A process diagram of an embodiment of the Lyapunov function sum of squares programming solution provided by the present application;

[0089] Figure 5 A flowchart of an embodiment of the simulation inner domain expansion parameter optimization algorithm provided by the present application;

[0090] Figure 6 A comparison result diagram of an embodiment of the comparison between the expanded transient stability domain and the initial transient stability domain provided by the present application;

[0091] Figure 7 A verification result diagram of an embodiment of the transient stability domain time domain simulation verification provided by the present application;

[0092] Figure 8 Figure 1 is a structural diagram of an embodiment of a controller parameter optimization system based on transient stability domain maximization provided by the present application. DETAILED DESCRIPTION

[0093] The technical solutions in the present application will be described clearly and completely below in combination with the drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of protection of the present application.

[0094] In the description of the present application, it should be understood that the terms "first", "second" and "third" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" and "third" can be explicitly or implicitly included one or more of the features. In the description of the present application, unless otherwise specified, the meaning of "several" is two or more.

[0095] The present application provides a controller parameter optimization method and system based on transient stability domain maximization, which maximizes the transient stability domain and improves the stability of the power system where the controller is located.

[0096] Embodiment 1

[0097] Based on the above needs, an embodiment of the present application provides a controller parameter optimization method based on transient stability domain maximization, and the method flow is as shown in Figure 1 The method comprises steps S1 to S5, and each step is specifically as follows:

[0098] S1, using a virtual synchronous generator to control a grid-forming controller, and obtaining operation data of the grid-forming controller; wherein the operation data of the grid-forming controller comprises common coupling point active power and common coupling point reactive power of the grid-forming controller.

[0099] This example analyzes a new energy unit composed of two synchronous generators and a grid-forming control converter connected in parallel, and the system topology diagram of the new energy unit is as shown in Figure 2

[0100] This embodiment uses a virtual synchronous generator to control a grid-forming controller, and the used virtual synchronous generator is as shown in Figure 3 The parameter meanings in the figure are specifically: P e represents the active power of the AC measurement grid connection point of the converter, P eref represents the given reference active power, w p = D / M, K​p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, Δω represents an angular velocity deviation unit value, U a , U b , and U c respectively represent three-phase voltages of a, b, and c.

[0101] The new energy unit in normal operation is controlled to obtain operation data of the grid-forming controller. The operation data of the grid-forming controller includes active power and reactive power at a point of common coupling of the grid-forming controller, and the active power and the reactive power at the point of common coupling are obtained through output of a power calculation module of the virtual synchronous generator.

[0102] S2, according to the operation data of the grid-forming controller, a multi-machine system state equation is constructed; wherein the multi-machine system state equation includes a synchronous machine swing equation, a power angle differential equation and a speed differential equation. This step includes steps S2.1 to S2.3, each step is as follows:

[0103] S2.1, according to the active power and the reactive power at the point of common coupling of the grid-forming controller, the synchronous phase angle and the voltage amplitude output by the virtual synchronous generator are obtained, and the synchronous machine swing equation is obtained.

[0104] According to the active power and the reactive power at the point of common coupling of the grid-forming controller, the synchronous phase angle and the voltage amplitude output by the virtual synchronous generator are obtained, and the synchronous machine swing equation is obtained.

[0105] S2.2, the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator are obtained, and the power angle differential equation and the speed differential equation are constructed according to the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator; wherein the power angle differential equation and the speed differential equation are as follows:

[0106]

[0107] In the formula, represents a virtual power angle; f represents a standard frequency, and Δω represents a virtual rotor angular velocity deviation unit value; represents a virtual rotor angular velocity deviation unit value; w p = D / M, K p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, P eref represents a given reference active power, P e represents an active power at a grid-connected point of a converter.

[0108] The embodiment establishes the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator through a virtual synchronous generator control loop, establishes the relationship among the power angle, the rotating speed and the active power, and obtains a power angle differential equation and a rotating speed differential equation.

[0109] S2.3, in combination with the synchronous machine swing equation, the power angle differential equation and the rotating speed differential equation, a multi-machine system state equation is constructed; wherein the multi-machine system state equation is specifically as follows:

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] In the formula, ω represents the virtual power angle of the i th unit relative to the j th unit; ω i represents the i th virtual rotor angular velocity, represents the derivative of the i th virtual rotor angular velocity, that is, the rotor angular acceleration; H i represents the inertia time constant of the i th unit; w p = D / M, K p = 1 / D, D represents the virtual generator damping coefficient, M represents the virtual generator inertia time coefficient, P eref represents the given reference active power, P ei represents the active power output by the i th unit; P mi represents the mechanical power input by the i th unit, D i represents the i th virtual generator damping coefficient.

[0116] The above multi-machine system state equation takes the first unit as a reference unit, and can correctly reflect the dynamic response of the multi-machine system and the network type controller.

[0117] S3, using a sum of squares programming algorithm to process the multi-machine system state equation, a Lyapunov function with an expanded stability domain is obtained. The sum of squares programming solving process of the Lyapunov function in this step is as shown in Figure 4 The step includes steps S3.1 to S3.3, and each step is specifically as follows:

[0118] S3.1, using a semi-definite programming solver to calculate the multi-machine system state equation, and obtaining an initial Lyapunov function.

[0119] The embodiment uses a semi-definite programming solver to calculate the state equation of the multi-machine system to obtain an initial Lyapunov function. In the embodiment, the semi-definite programming solver used is the SOSSTOOLS toolbox.

[0120] S3.2, the boundary value of the initial Lyapunov function is calculated using the forward search method and the bisection method, and the boundary value of the transient stability region and the boundary value normalized Lyapunov function are obtained.

[0121] The embodiment uses the forward search method and the bisection method to calculate the boundary value of the initial Lyapunov function, and the boundary value obtained is the boundary value of the transient stability region. Then, based on the boundary value of the transient stability region, the boundary value normalized Lyapunov function is constructed.

[0122] S3.3, the boundary value normalized Lyapunov function is cyclically updated until the solution quality of the updated Lyapunov function is lower than a preset threshold, and the cycle is stopped to obtain a Lyapunov function with an expanded stability region; wherein in each cycle, a coefficient polynomial of the Lyapunov function is calculated according to the state equation of the multi-machine system and the boundary value normalized Lyapunov function; under the constraint condition that the outer region contains the inner region, a candidate Lyapunov function is constructed in combination with the coefficient polynomial and the boundary value normalized Lyapunov function; if the solution quality of the candidate Lyapunov function in the current cycle is lower than the preset threshold, the candidate Lyapunov function in the current cycle is taken as the Lyapunov function with an expanded stability region.

[0123] The embodiment cyclically updates the boundary value normalized Lyapunov function, and the specific steps are as follows:

[0124] (1) the coefficient polynomial satisfying the initial constraint condition is calculated in combination with the state equation of the multi-machine system and the boundary value normalized Lyapunov function;

[0125] (2) a new candidate Lyapunov function is constructed under the constraint condition that the outer region contains the inner region according to the coefficient polynomial satisfying the initial constraint condition;

[0126] (3) it is judged whether the solution quality of the candidate Lyapunov function in the current cycle is lower than the preset threshold: if it is lower than the preset threshold, the cycle is exited, and the candidate Lyapunov function in the current cycle is output as the Lyapunov function with an expanded stability region. If it is higher than the preset threshold, the steps (1) and (2) are repeated until the solution quality of the updated Lyapunov function is lower than the preset threshold, and the cycle is stopped.

[0127] S4, the stable domain expanded Lyapunov function is processed using an analog inner domain expansion parameter optimization algorithm to construct an optimized Lyapunov function; wherein the analog inner domain expansion parameter optimization algorithm comprises an inner loop algorithm and an outer loop algorithm. A flowchart of the analog inner domain expansion parameter optimization algorithm is shown in Figure 5 The step S4 comprises steps S4.1 to S4.3, which are specifically as follows:

[0128] S4.1, the stable domain expanded Lyapunov function is processed using the inner loop algorithm to obtain first Lyapunov function parameters. The step S4.1 comprises steps S4.1.1 to S4.1.3, which are specifically as follows:

[0129] S4.1.1, the parameters of the stable domain expanded Lyapunov function are set as to-be-solved parameters to construct a state space equation containing the to-be-solved parameters.

[0130] The embodiment optimizes parameters based on the stable domain expanded Lyapunov function, enters the inner loop algorithm, sets the parameters in the stable domain expanded Lyapunov function as to-be-solved parameters, sets the parameter values as decision variables to be solved, and obtains a state space equation containing the to-be-solved parameters.

[0131] S4.1.2, a first constraint condition containing the state space equation and a Lyapunov function stability condition is constructed; wherein the first constraint condition is specifically as follows:

[0132]

[0133] In the formula, s2 and s4 represent polynomial coefficients to be solved; V represents the stable domain expanded Lyapunov function, f represents the state space equation containing the to-be-solved parameters, G represents an equality constraint under a triangular coordinate transformation, and l2 represents a sum of squares polynomial. M

[0134] S4.1.3, the state space equation containing the to-be-solved parameters is solved by a solver under the constraint of the first constraint condition to obtain first Lyapunov function parameters.

[0135] Under the constraint of the first constraint condition, the embodiment inputs the state space equation containing the to-be-solved parameters into a semi-definite programming solver, and a new Lyapunov function parameter is automatically returned by the semi-definite programming solver to obtain the first Lyapunov function parameters.

[0136] ​​S4.2, iteratively optimize the first Lyapunov function parameter using an outer loop algorithm to obtain a second Lyapunov function parameter. This step includes steps S4.2.1 to S4.2.2, each of which is as follows:

[0137] S4.2.1, construct a second constraint condition in combination with the constraint condition of the inner domain containing the outer domain, the Lyapunov function stability condition, and the trigonometric function transformation constraint condition. The second constraint condition is as follows:

[0138]

[0139] In the formula, V represents the Lyapunov function to be solved; p represents the Lyapunov function after the last round of iterative optimization; and l i (o = 1, 2, 3) represents a sum of squares polynomial, G i (i = 1, 2, 3) represents an equality constraint under trigonometric coordinate transformation, s1, s2, and s4 represent known polynomial coefficients; f represents a state space equation containing the Lyapunov function parameter after the last round of iterative optimization; ∑ M represents a sum of squares polynomial.

[0140] S4.2.2, iteratively optimize the first Lyapunov function parameter using an outer loop algorithm; until the Lyapunov function parameter after the current round of iterative optimization meets the preset iteration stopping condition, stop the iterative optimization, and output the Lyapunov function parameter after the current round of iterative optimization as the second Lyapunov function parameter.

[0141] In the first round of iterative optimization, the first Lyapunov function parameter is optimized under the constraint of the second constraint condition to obtain the Lyapunov function parameter after the first round of iterative optimization. In the iterative optimization after the first round, the Lyapunov function parameter after the last round of iterative optimization is optimized under the constraint of the second constraint condition to obtain the Lyapunov function parameter after the current round of iterative optimization. The preset iteration stopping condition is that the Lyapunov function parameter after the current round of iterative optimization meets the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameter after the last round of iterative optimization and the Lyapunov function parameter after the current round of iterative optimization is less than the preset accuracy value.

[0142] This embodiment uses an inner loop algorithm to obtain new first Lyapunov function parameters, and then uses an outer loop algorithm to further optimize the first Lyapunov function parameters to obtain second Lyapunov function parameters that maximize the transient stability domain.

[0143] Finally, the embodiment also verifies whether the obtained second Lyapunov function parameters satisfy the condition of expansion of the boundary value of the transient stability domain and the requirement of convergence accuracy. When the maximum absolute value deviation of the Lyapunov function parameters of two adjacent iterations in the outer loop iteration is lower than the accuracy value or the solution quality, the loop ends, and the Lyapunov function parameters optimized in the current iteration are output as the second Lyapunov function parameters, which are used for redesign of the controller parameters and estimation of the maximized transient stability domain of the system.

[0144] S4.3, according to the second Lyapunov function parameters, parameter optimization is performed on the Lyapunov function for stability domain expansion to obtain an optimized Lyapunov function.

[0145] According to the second Lyapunov function parameters, the embodiment performs parameter optimization on the Lyapunov function for stability domain expansion to obtain an optimized Lyapunov function.

[0146] S5, according to the optimized Lyapunov function, the parameters of the network-type controller are redesigned.

[0147] According to the second Lyapunov function parameters and the optimized Lyapunov function, the embodiment redesigns the parameters of the network-type controller and reestimates the maximized transient stability domain of the power system. A comparison chart of the expanded transient stability domain and the initial transient stability domain is shown in Figure 6 , and the conservativeness of the work point of the expanded transient stability domain in time domain simulation is verified, and the verification result is shown in Figure 7 .

[0148] The above embodiment of the present application has the following beneficial effects:

[0149] The controller parameter optimization method based on maximization of the transient stability domain provided by the present application firstly constructs the state space equation of the multi-machine system, which can correctly reflect the kinematic equation of the virtual synchronous generator control of the dynamic response; then, in the multi-machine system, the Lyapunov function for stability domain expansion is obtained by using the sum-of-squares programming algorithm, and the Lyapunov function parameters and the Lyapunov function of the maximized stability domain are solved by using the simulated internal domain expansion parameter optimization algorithm; finally, according to the Lyapunov function of the maximized stability domain, the parameters of the network-type controller in the multi-machine system are optimized and the transient stability domain is designed; thereby the maximized transient stability domain is realized, and the stability of the power system where the controller is located is improved.

[0150] Embodiment 2

[0151] Based on the content of the above embodiments, an embodiment of the present application provides a controller parameter optimization system based on transient stability domain maximization, comprising: a running data module 101, a state equation module 102, a sum-of-squares programming module 103, a parameter optimization module 104 and a controller optimization module 105. The system structure is as shown in Figure 8

[0152] The running data module 101 is configured to use a virtual synchronous generator control network type controller and obtain running data of the network type controller; wherein the running data of the network type controller comprises common coupling point active power and common coupling point reactive power of the network type controller.

[0153] The state equation module 102 is configured to construct a multi-machine system state equation according to the running data of the network type controller; wherein the multi-machine system state equation comprises a synchronous machine swing equation, an angle differential equation and a speed differential equation.

[0154] The sum-of-squares programming module 103 is configured to use a sum-of-squares programming algorithm to process the multi-machine system state equation to obtain a stability domain expansion Lyapunov function.

[0155] The parameter optimization module 104 is configured to use an analog internal domain expansion parameter optimization algorithm to process the stability domain expansion Lyapunov function to construct an optimized Lyapunov function; wherein the analog internal domain expansion parameter optimization algorithm comprises an inner loop algorithm and an outer loop algorithm.

[0156] The controller optimization module 1025 is configured to redesign parameters of the network type controller according to the optimized Lyapunov function.

[0157] Further, the state equation module 102 comprises: a first equation unit 201, a second equation unit 202 and a third equation unit 203.

[0158] The first equation unit 201 is configured to obtain a synchronous machine swing equation by outputting a synchronous phase angle and a voltage amplitude of the virtual synchronous generator according to the common coupling point active power and the common coupling point reactive power of the network type controller.

[0159] The second equation unit 202 is configured to obtain a damping coefficient of the virtual synchronous generator and an inertia time coefficient of the virtual synchronous generator, and construct an angle differential equation and a speed differential equation according to the damping coefficient of the virtual synchronous generator and the inertia time coefficient of the virtual synchronous generator; wherein the angle differential equation and the speed differential equation are as follows:

[0160]

[0161] In the formula,​ represents a virtual power angle; f represents a standard frequency, and Δω represents a virtual rotor speed deviation unit value; represents a virtual rotor speed deviation unit value; w p = D / M, K p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, P eref represents a given reference active power, P e represents a converter grid point active power.

[0162] The third equation unit 203 is configured to combine the synchronous machine swing equation, the power angle differential equation, and the rotor speed differential equation to construct a multi-machine system state equation; wherein the multi-machine system state equation is specifically as follows:

[0163]

[0164]

[0165]

[0166]

[0167]

[0168] In the formula, represents a virtual power angle of the i th unit relative to the j th unit; ω i represents a virtual rotor speed of the i th unit, represents a derivative of the i th virtual rotor speed, i.e., a rotor angular acceleration; H i represents an inertia time constant of the i th unit; w p = D / M, K p = 1 / D, D represents a virtual generator damping coefficient, M represents a virtual generator inertia time coefficient, P eref represents a given reference active power, P ei represents an active power output by the i th unit; P mi represents a mechanical power input by the i th unit, D i represents an i th virtual generator damping coefficient.

[0169] Further, the sum-of-squares programming module 103 comprises an initial function unit 301, a normalization function unit 302, and a loop updating unit 303.

[0170] The initial function unit 301 is configured to use a semi-definite programming solver to calculate the multi-machine system state equation to obtain an initial Lyapunov function;

[0171] The normalization function unit 302 is configured to calculate the boundary value of the initial Lyapunov function by using a forward search method and a dichotomy method, and obtain a boundary value of a transient stability region and a boundary value normalized Lyapunov function.

[0172] The cyclic updating unit 303 is configured to cyclically update the boundary value normalized Lyapunov function until the solution quality of the updated Lyapunov function is lower than a preset threshold value, stop the cycle, and obtain an enlarged stability region Lyapunov function. In each cycle, the coefficient polynomial of the Lyapunov function is calculated according to the multi-machine system state equation and the boundary value normalized Lyapunov function. The candidate Lyapunov function is constructed by combining the coefficient polynomial and the boundary value normalized Lyapunov function under the constraint condition that the outer region contains the inner region. If the solution quality of the candidate Lyapunov function in the current cycle is lower than the preset threshold value, the candidate Lyapunov function in the current cycle is taken as the enlarged stability region Lyapunov function.

[0173] Further, the parameter optimization module 104 includes an inner loop unit 401, an outer loop unit 402, and an optimization function unit 403.

[0174] The inner loop unit 401 is configured to process the enlarged stability region Lyapunov function by using an inner loop algorithm, and obtain first Lyapunov function parameters.

[0175] The outer loop unit 402 is configured to iteratively optimize the first Lyapunov function parameters by using an outer loop algorithm, and obtain second Lyapunov function parameters.

[0176] The optimization function unit 403 is configured to perform parameter optimization on the enlarged stability region Lyapunov function according to the second Lyapunov function parameters, and obtain an optimized Lyapunov function.

[0177] Further, the inner loop unit 401 includes a variable setting sub-unit 501, a first constraint sub-unit 502, and a variable solving sub-unit 503.

[0178] The variable setting sub-unit 501 is configured to set the parameters of the enlarged stability region Lyapunov function as to-be-solved parameters, and construct a state space equation containing the to-be-solved parameters.

[0179] The first constraint sub-unit 502 is configured to construct a first constraint condition containing the state space equation and a Lyapunov function stability condition. The first constraint condition is specifically as follows:

[0180]

[0181] In the formula, s2 and s4 represent polynomial coefficients to be solved; V represents a Lyapunov function with an expanded stable domain, f represents a state space equation containing parameters to be solved; G represents an equality constraint under a trigonometric coordinate transformation; and l2 represent sum-of-squares polynomials; ∑ M represents a sum-of-squares polynomial.

[0182] The variable solving subunit 503 is configured to solve the state space equation containing the parameters to be solved by a solver under the constraint of the first constraint condition, and obtain first Lyapunov function parameters.

[0183] Further, the outer loop unit 402 includes a second constraint subunit 601 and an outer loop optimization subunit 602.

[0184] The second constraint subunit 601 is configured to combine the constraint condition of the outer domain containing the inner domain, the Lyapunov function stability condition, and the trigonometric function transformation constraint condition to construct a second constraint condition; wherein the second constraint condition is specifically as follows:

[0185]

[0186] In the formula, V represents a Lyapunov function to be solved; p represents Lyapunov function parameters after optimization of the last iteration; and l i (o=1, 2, 3) represent sum-of-squares polynomials, G i (i=1, 2, 3) represent equality constraints under a trigonometric coordinate transformation, s1, s2, and s4 represent known polynomial coefficients; f represents a state space equation containing Lyapunov function parameters after optimization of the last iteration; ∑ M represents a sum-of-squares polynomial.

[0187] The outer loop optimization subunit 602 is configured to use an outer loop algorithm to iteratively optimize the first Lyapunov function parameters; until Lyapunov function parameters after optimization of the current iteration meet a preset iteration stop condition, stop the iterative optimization, and output the Lyapunov function parameters after optimization of the current iteration as second Lyapunov function parameters.

[0188] In the first round of iterative optimization, the first Lyapunov function parameter is optimized under the constraint of the second constraint condition, and a first round of iterative optimization of the Lyapunov function parameter is obtained; in the iterative optimization after the first round, the Lyapunov function parameter of the last round of iterative optimization is optimized under the constraint of the second constraint condition, and the current iterative optimization of the Lyapunov function parameter is obtained; the preset iteration stopping condition is that the current iterative optimization of the Lyapunov function parameter satisfies the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameter of the last round of iterative optimization and the current iterative optimization of the Lyapunov function parameter is less than the preset accuracy value.

[0189] The working principle and step flow of the embodiment can be but not limited to refer to the related description of the embodiment 1.

[0190] The above embodiment of the present application has the following beneficial effects:

[0191] The controller parameter optimization system based on transient stability domain maximization provided by the present application realizes the maximization of the transient stability domain and improves the stability of the power system in which the network type controller is located.

[0192] The above is the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements are also considered within the protection scope of the present application.

Claims

1. A controller parameter optimization method based on maximizing the transient stability region, characterized in that, include: A virtual synchronous generator is used to control a grid-type controller, and the operating data of the grid-type controller is acquired; wherein, the operating data of the grid-type controller includes the active power and reactive power at the common coupling point of the grid-type controller; Based on the operating data of the network controller, the state equations of the multi-machine system are constructed; wherein, the state equations of the multi-machine system include the oscillation equation of the synchronous machine, the power angle differential equation, and the speed differential equation; The state equations of the multi-machine system are processed using a sum-of-squares programming algorithm to obtain the Lyapunov function with an expanded stability domain. The Lyapunov function with expanded stable domain is processed using a simulated inner domain expansion parameter optimization algorithm to construct an optimized Lyapunov function; wherein, the simulated inner domain expansion parameter optimization algorithm includes an inner loop algorithm and an outer loop algorithm; Based on the optimization of the Lyapunov function, the parameters of the network controller are redesigned.

2. The controller parameter optimization method based on maximizing the transient stability region as described in claim 1, characterized in that, The step of constructing the multi-machine system state equations based on the operating data of the network controller includes: Based on the active power and reactive power at the common coupling point of the network controller, the synchronous generator oscillation equation is obtained by outputting the synchronous phase angle and voltage amplitude through the virtual synchronous generator. Obtain the damping coefficient and inertia time coefficient of the virtual synchronous generator. Based on the damping coefficient and inertia time coefficient of the virtual synchronous generator, construct the power angle differential equation and the rotational speed differential equation; wherein, the power angle differential equation and the rotational speed differential equation are as follows: In the formula, Indicates a virtual work angle; Indicates standard frequency, This represents the per-unit value of the virtual rotor angular velocity deviation; express The derivative of , i.e., the rotor angular acceleration; , D represents the damping coefficient of the virtual synchronous generator, and M represents the inertia time coefficient of the virtual synchronous generator. Indicates the given reference active power. This indicates the active power at the converter's grid connection point; By combining the oscillation equation of the synchronizer, the power angle differential equation, and the speed differential equation, the state equation of the multi-machine system is constructed; wherein, the state equation of the multi-machine system is specifically as follows: In the formula, This represents the virtual power angle of the i-th generating unit relative to the j-th generating unit; Represents the angular velocity of the i-th virtual rotor. The derivative of the angular velocity of the i-th virtual rotor is given, which is the rotor angular acceleration. Let represent the inertial time constant of the i-th generator unit; , D represents the damping coefficient of the virtual synchronous generator, and M represents the inertia time coefficient of the virtual synchronous generator. Indicates the given reference active power. This represents the active power output of the i-th generator unit; This represents the mechanical power input to the i-th generator unit. Let represent the damping coefficient of the i-th virtual synchronous generator.

3. The controller parameter optimization method based on maximizing the transient stability region as described in claim 1, characterized in that, The process of using the sum-of-squares programming algorithm to process the state equations of the multi-machine system to obtain the Lyapunov function with an expanded stability region includes: The state equations of the multi-machine system are calculated using a semidefinite programming solver to obtain the initial Lyapunov function; The boundary values ​​of the initial Lyapunov function are calculated using the forward search method and the bisection method to obtain the boundary values ​​of the transient stable domain and the boundary value normalized Lyapunov function. The boundary-value-normalized Lyapunov function is iteratively updated until the solution quality of the updated Lyapunov function is lower than a preset threshold, at which point the iteration stops, and a Lyapunov function with expanded stability region is obtained. In each iteration, the coefficient polynomial of the Lyapunov function is calculated based on the state equations of the multi-machine system and the boundary-value-normalized Lyapunov function. Under the constraint that the outer domain contains the inner domain, a candidate Lyapunov function is constructed by combining the coefficient polynomial and the boundary-value-normalized Lyapunov function. If the solution quality of the candidate Lyapunov function in the current iteration is lower than the preset threshold, the candidate Lyapunov function in the current iteration is used as the Lyapunov function with expanded stability region.

4. The controller parameter optimization method based on maximizing the transient stability region as described in claim 1, characterized in that, The process of using a simulated inner domain expansion parameter optimization algorithm to process the Lyapunov function with expanded stability domain and construct an optimized Lyapunov function includes: The Lyapunov function with the expanded stability region is processed using an inner loop algorithm to obtain the parameters of the first Lyapunov function. The first Lyapunov function parameters are iteratively optimized using an outer loop algorithm to obtain the second Lyapunov function parameters; Based on the parameters of the second Lyapunov function, the parameters of the Lyapunov function with the expanded stability region are optimized to obtain the optimized Lyapunov function.

5. The controller parameter optimization method based on maximizing the transient stability region as described in claim 4, characterized in that, The process of using an inner loop algorithm to process the Lyapunov function with expanded stability region to obtain the first Lyapunov function parameters includes: The parameters of the Lyapunov function with the extended stability region are set as the parameters to be solved, and a state-space equation containing the parameters to be solved is constructed. Construct a first constraint condition that includes the state-space equation and the Lyapunov function stability condition; wherein, the first constraint condition is specifically as follows: In the formula, and Represents the coefficients of the polynomial to be solved; , Lyapunov functions representing the expansion of the stability region G represents the state-space equation containing the parameters to be solved; G represents the equality constraints under triangular coordinate transformation. and Represents a polynomial of the sum of squares; Represents a polynomial of the sum of squares; Under the constraints of the first constraint, the state-space equation containing the parameters to be solved is solved by the solver to obtain the first Lyapunov function parameters.

6. The controller parameter optimization method based on maximizing the transient stability region as described in claim 4, characterized in that, The step of iteratively optimizing the parameters of the first Lyapunov function using an outer loop algorithm to obtain the parameters of the second Lyapunov function includes: Combining the constraints of the outer domain containing the inner domain, the stability condition of the Lyapunov function, and the trigonometric function transformation constraint, a second constraint is constructed; specifically, the second constraint is as follows: In the formula, denoted as Lyapunov function with expanded stability region; p represents Lyapunov function after the previous iteration optimization; and Represents a polynomial of the sum of squares. This represents the equality constraint under trigonometric coordinate transformation, where i=1,2,3 and o=1,2; , and Represents the coefficients of the polynomial to be solved; , This represents the state-space equation containing the parameters to be solved; Represents a polynomial of the sum of squares; The first Lyapunov function parameter is iteratively optimized using an outer loop algorithm until the currently optimized Lyapunov function parameter meets the preset iteration stopping condition. Then, the iteration optimization is stopped, and the currently optimized Lyapunov function parameter is output as the second Lyapunov function parameter. Specifically, in the first round of iterative optimization, under the constraint of the second constraint, the first Lyapunov function parameters are optimized to obtain the Lyapunov function parameters after the first round of iterative optimization; in subsequent rounds of iterative optimization, under the constraint of the second constraint, the Lyapunov function parameters after the previous round of iterative optimization are optimized to obtain the Lyapunov function parameters after the current round of iterative optimization; the preset iteration stopping condition is as follows: the Lyapunov function parameters after the current round of iterative optimization satisfy the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameters after the previous round of iterative optimization and the Lyapunov function parameters after the current round of iterative optimization is less than the preset accuracy value.

7. A controller parameter optimization system based on maximizing the transient stability region, characterized in that, include: The system includes a data processing module, a state equation module, a sum-of-squares programming module, a parameter optimization module, and a controller optimization module. The operation data module is configured to use a virtual synchronous generator to control a grid-type controller and acquire the operation data of the grid-type controller; wherein, the operation data of the grid-type controller includes the active power and reactive power of the common coupling point of the grid-type controller; The state equation module is used to construct the state equation of a multi-machine system based on the operating data of the network controller; wherein the state equation of the multi-machine system includes the oscillation equation of the synchronous machine, the power angle differential equation, and the speed differential equation. The square sum programming module is used to process the state equations of the multi-machine system using the square sum programming algorithm to obtain the Lyapunov function with an expanded stability domain. The parameter optimization module uses a simulated inner domain expansion parameter optimization algorithm to process the Lyapunov function whose stable domain has been expanded, and constructs an optimized Lyapunov function; wherein, the simulated inner domain expansion parameter optimization algorithm includes an inner loop algorithm and an outer loop algorithm; The controller optimization module is responsible for redesigning the parameters of the network controller based on the optimized Lyapunov function.

8. The controller parameter optimization system based on maximizing the transient stability region as described in claim 7, characterized in that, The state equation module includes: a first equation unit, a second equation unit, and a third equation unit; The first equation unit is used to obtain the oscillation equation of the synchronous machine by using the synchronous phase angle and voltage amplitude output by the virtual synchronous generator based on the active power and reactive power of the common coupling point of the grid-type controller. The second equation unit is used to obtain the damping coefficient and inertia time coefficient of the virtual synchronous generator. Based on the damping coefficient and inertia time coefficient of the virtual synchronous generator, the power angle differential equation and the rotational speed differential equation are constructed. Specifically, the power angle differential equation and the rotational speed differential equation are as follows: In the formula, Indicates a virtual work angle; Indicates standard frequency, This represents the per-unit value of the virtual rotor angular velocity deviation; The derivative of the per-unit value of the virtual rotor angular velocity deviation, i.e., the rotor angular acceleration; , D represents the damping coefficient of the virtual synchronous generator, and M represents the inertia time coefficient of the virtual synchronous generator. Indicates the given reference active power. This indicates the active power at the converter's grid connection point; The third-party process unit is used to construct the state equations of the multi-machine system by combining the oscillation equation of the synchronizer, the power angle differential equation, and the speed differential equation; wherein, the state equations of the multi-machine system are specifically as follows: In the formula, This represents the virtual power angle of the i-th generating unit relative to the j-th generating unit; Represents the angular velocity of the i-th virtual rotor; The derivative of the angular velocity of the i-th virtual rotor; Let represent the inertial time constant of the i-th generator unit; , D represents the damping coefficient of the virtual synchronous generator, and M represents the inertia time coefficient of the virtual synchronous generator. Indicates the given reference active power. This represents the active power output of the i-th generator unit; This represents the mechanical power input to the i-th generator unit. Let represent the damping coefficient of the i-th virtual synchronous generator.

9. The controller parameter optimization system based on maximizing the transient stability region as described in claim 7, characterized in that, The sum of squares programming module includes: an initial function unit, a normalization function unit, and a cyclic update unit; The initial function unit is used to calculate the state equations of the multi-machine system using a semidefinite programming solver to obtain the initial Lyapunov function; The normalization function unit is used to calculate the boundary values ​​of the initial Lyapunov function using the forward search method and the bisection method, so as to obtain the boundary values ​​of the transient stable domain and the boundary value normalized Lyapunov function. The cyclic update unit is used to cyclically update the boundary value-normalized Lyapunov function until the solution quality of the updated Lyapunov function is lower than a preset threshold, at which point the loop stops, and a Lyapunov function with expanded stability domain is obtained. In each loop, the coefficient polynomial of the Lyapunov function is calculated based on the state equations of the multi-machine system and the boundary value-normalized Lyapunov function. Under the constraint that the outer domain contains the inner domain, a candidate Lyapunov function is constructed by combining the coefficient polynomial and the boundary value-normalized Lyapunov function. If the solution quality of the candidate Lyapunov function in the current loop is lower than the preset threshold, the candidate Lyapunov function in the current loop is used as the Lyapunov function with expanded stability domain.

10. A controller parameter optimization system based on maximizing the transient stability region as described in claim 7, characterized in that, The parameter optimization module includes: an inner loop unit, an outer loop unit, and an optimization function unit; The inner loop unit is used to process the Lyapunov function with the expanded stability region using an inner loop algorithm to obtain the first Lyapunov function parameters. The outer loop unit is used to iteratively optimize the first Lyapunov function parameters using an outer loop algorithm to obtain the second Lyapunov function parameters. The optimization function unit is used to optimize the parameters of the Lyapunov function with expanded stability region based on the parameters of the second Lyapunov function, so as to obtain an optimized Lyapunov function.

11. A controller parameter optimization system based on maximizing the transient stability region as described in claim 10, characterized in that, The inner loop unit includes: a variable setting subunit, a first constraint subunit, and a variable solving subunit; The variable setting subunit is used to set the parameters of the Lyapunov function with the expanded stability region as the parameters to be solved, and to construct a state-space equation containing the parameters to be solved. The first constraint subunit is used to construct a first constraint condition that includes the state-space equation and the Lyapunov function stability condition; wherein, the first constraint condition is specifically as follows: In the formula, and Represents the coefficients of the polynomial to be solved; , Lyapunov functions representing the expansion of the stability region G represents the state-space equation containing the parameters to be solved; G represents the equality constraints under triangular coordinate transformation. and Represents a polynomial of the sum of squares; Represents a polynomial of the sum of squares; The variable solving subunit is used to solve the state-space equation containing the parameters to be solved by a solver under the constraints of the first constraint condition, so as to obtain the first Lyapunov function parameters.

12. The controller parameter optimization system based on maximizing the transient stability region as described in claim 10, characterized in that, The outer loop unit includes: a second constraint subunit and an outer loop optimization subunit; The second constraint subunit is used to construct a second constraint by combining the constraint condition that the outer domain contains the inner domain, the stability condition of the Lyapunov function, and the trigonometric function transformation constraint condition; wherein, the second constraint condition is as follows: In the formula, denoted as Lyapunov function with expanded stability region; p represents Lyapunov function after the previous iteration optimization; and Represents a polynomial of the sum of squares. This represents the equality constraint under trigonometric coordinate transformation, where i=1,2,3 and o=1,2; , and Represents the coefficients of the polynomial to be solved; , This represents the state-space equation containing the Lyapunov function parameters optimized in the previous iteration; Represents a polynomial of the sum of squares; The outer loop optimization subunit is used to iteratively optimize the first Lyapunov function parameters using an outer loop algorithm until the currently iteratively optimized Lyapunov function parameters meet the preset iteration stop condition, then stop the iterative optimization and output the currently iteratively optimized Lyapunov function parameters as the second Lyapunov function parameters. Specifically, in the first round of iterative optimization, under the constraint of the second constraint, the first Lyapunov function parameters are optimized to obtain the Lyapunov function parameters after the first round of iterative optimization; in subsequent rounds of iterative optimization, under the constraint of the second constraint, the Lyapunov function parameters after the previous round of iterative optimization are optimized to obtain the Lyapunov function parameters after the current round of iterative optimization; the preset iteration stopping condition is as follows: the Lyapunov function parameters after the current round of iterative optimization satisfy the preset transient stability domain boundary value expansion condition and the preset convergence accuracy requirement, and the maximum absolute value deviation between the Lyapunov function parameters after the previous round of iterative optimization and the Lyapunov function parameters after the current round of iterative optimization is less than the preset accuracy value.

Images