A parameter optimization method and system for a grid-connected static VAR generator

Abstract

The present application discloses a parameter optimization method and system for a grid-type static VAR generator, and relates to the field of parameter optimization of a grid-type static VAR generator. The method comprises: randomly generating multiple groups of control parameters of a current inner loop controller of a grid-type static VAR generator; using a bidirectional long short-term memory neural network to construct a first function mapping between an overshoot and a short-circuit ratio of a power grid and control parameters, a second function mapping between a response time and a short-circuit ratio of a power grid and control parameters, and a third function mapping between a stability margin and an equivalent inductance of a power grid and control parameters; using a particle swarm algorithm to randomly select multiple groups of control parameters, inputting the random selection result and the short-circuit ratio of a power grid at a current moment into the first function mapping and the second function mapping respectively, and inputting the random selection result and the equivalent inductance of a power grid at a current moment into the third function mapping, until the stability margin difference, the overshoot and the response time reach a minimum value, and then outputting the optimal control parameters.

Description

Technical Field

[0001] The present application relates to the field of parameter optimization of a grid-type static VAR generator, and more specifically, to a method and system for optimizing parameters of a grid-type static VAR generator. Background Art

[0002] The static VAR generator (SVG) can automatically compensate for the reactive power required by the power grid. It has the advantages of fast dynamic response, strong reactive power support capability, and high intelligence. It is a commonly used reactive power compensation and harmonic suppression device in the current power grid.

[0003] The grid-forming static VAR generator adopts a power synchronization method similar to that of a synchronous generator. It can achieve synchronization without a phase-locked loop in the control loop. It is an important device for improving the transient voltage stability of the power system.

[0004] The existing control loop of the grid-type static VAR generator includes a power synchronization link, a voltage outer loop and a current inner loop. Among them, the current inner loop control directly affects the control performance of the grid-type static VAR generator. Among them, the proportional integral (PI) controller is a commonly used controller for the current inner loop of the grid-type static VAR generator, but it is difficult to take into account the stability margin and control performance of the grid-type static VAR generator when designing its parameters. If the parameters are not designed properly, it is easy to cause oscillation in the power system. At the same time, it is difficult for proportional integral control to achieve zero-static error control of current, and the cross-coupling terms in the control loop further reduce the robustness and dynamic response performance of the power system. In addition, with the large-scale access of power electronic equipment, nonlinear loads, etc. to the power grid, their high-frequency switching characteristics introduce harmonic interference into the power grid; the mismatch between semiconductor devices and the digital-to-analog conversion of control will also introduce DC components into the power grid; under non-ideal power grids, the various types of disturbance signals in the power system also affect the stable control of the grid-type static VAR generator. Summary of the invention

[0005] The purpose of this application is to provide a parameter optimization method and system for a grid-type static VAR generator, which is used to solve the problem that the current inner loop controller of the current grid-type static VAR generator has a weak ability to suppress the DC component and harmonic interference on the grid side, and the problem that the design of control parameters cannot adapt to the intensity of changes in different grids.

[0006] In a first aspect of the present application, a method for optimizing parameters of a grid-connected static VAR generator is provided, the method comprising:

[0007] Randomly generate multiple groups of control parameters of the current inner loop controller of the grid-type static VAR generator; wherein the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between the input signal and the output signal of the proportional resonant controller with a generalized integral link;

[0008] A bidirectional long short-term memory neural network is used to construct a first function mapping between the overshoot of the current inner loop controller and the short-circuit ratio of the power grid and the control parameters, a second function mapping between the response time of the current inner loop controller and the short-circuit ratio of the power grid and the control parameters, and a third function mapping between the stability margin of the grid-forming static VAR generator and the equivalent inductance of the power grid and the control parameters;

[0009] A particle swarm algorithm is used to randomly select multiple groups of control parameters, and the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment are input into the first function mapping and the second function mapping respectively, and the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment are input into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and the optimal control parameters of the proportional resonant controller with a harmonic compensation link are output; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

[0010] In one implementation, the bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network.

[0011] In one implementation, the expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0012] In one implementation, the expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0013] In one implementation, obtaining the short-circuit ratio of the power grid at the current moment includes:

[0014] The impedance of the power grid is measured by a small signal injection method to obtain impedance measurement results;

[0015] The short-circuit ratio of the power grid at the current moment is calculated based on the impedance measurement results.

[0016] In one implementation, the particle swarm algorithm adopts a variable weight crossover particle swarm algorithm.

[0017] In a second aspect of the present application, a grid-type static VAR generator parameter optimization system is provided, the system comprising:

[0018] A control parameter acquisition module, used for randomly generating multiple groups of control parameters of a current inner loop controller of a grid-forming static VAR generator; wherein the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between an input signal and an output signal of a proportional resonant controller with a generalized integral link;

[0019] A function mapping construction module is used to construct a first function mapping between an overshoot of an inner current loop controller and a short circuit ratio of a power grid and a control parameter, a second function mapping between a response time of the inner current loop controller and a short circuit ratio of a power grid and a control parameter, and a third function mapping between a stability margin of a grid-forming static VAR generator and an equivalent inductance of a power grid and a control parameter by using a bidirectional long short-term memory neural network;

[0020] The control parameter optimization module is used to randomly select multiple groups of control parameters by using a particle swarm algorithm, input the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment into the first function mapping and the second function mapping respectively, and input the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and output the optimal control parameters of the proportional resonant controller with a harmonic compensation link; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

[0021] In one implementation, the bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network.

[0022] In one implementation, the expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0023] In one implementation, the expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f 2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0024] Compared with the prior art, this application has the following beneficial effects:

[0025] In a parameter optimization method for a grid-type static VAR generator provided in an embodiment of the present application, first, a proportional resonant controller based on a generalized integral link and a harmonic compensation link is designed, so that the proportional resonant controller has a stronger suppression capability for the DC component and harmonic interference on the grid side, and therefore, it is used for the current inner loop control of the grid-type static VAR generator, which can improve the suppression capability of the grid-type static VAR generator for grid-side harmonics and DC offset. Secondly, in order to optimize the parameter design of the proportional resonant controller and overcome the problem that the performance of the current inner loop controller of the grid-type static VAR generator is difficult to quantify and evaluate, the present application uses a bidirectional long short-term memory neural network to construct a functional mapping relationship between the stability margin difference, overshoot and response time with respect to the short-circuit ratio of the grid and the control parameters of the proportional resonant controller. Finally, a method for solving the control parameters of a proportional resonant controller that can adapt to changes in the short-circuit ratio is constructed. That is, a particle swarm algorithm is used to randomly select multiple groups of randomly generated control parameters to obtain random selection results of multiple groups of control parameters. The random selection results and the short-circuit ratio and equivalent inductance of the power grid at the current moment are respectively input into the function mapping relationship. When the obtained stability margin difference is the smallest, the overshoot is the smallest, and the response time reaches the minimum, the optimal control parameters of the proportional resonant controller with a harmonic compensation link can be output. Based on the optimal control parameters, it can be ensured that the grid-forming static VAR generator can always maintain a good stability margin and dynamic response performance when the short-circuit ratio changes in a wide range, that is, when a transient occurs in the power grid, thereby improving the real-time adjustment capability of the grid-forming static VAR generator.

[0026] Secondly, an embodiment of the present application also provides a grid-type static VAR generator parameter optimization system, which has the same beneficial effects as the grid-type static VAR generator parameter optimization method described above, so it will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The drawings described herein are used to provide a further understanding of the embodiments of the present application, constitute a part of the present application, and do not constitute a limitation on the embodiments of the present application. In the drawings:

[0028] Figure 1 A schematic diagram of a flow chart of a parameter optimization method for a grid-connected static VAR generator provided in an embodiment of the present application;

[0029] Figure 2 A control block diagram of a proportional resonant controller with a harmonic compensation link provided in an embodiment of the present application;

[0030] Figure 3 Bode diagrams of the proportional-integral controller and the proportional resonant controller with a harmonic compensation link provided in the embodiments of the present application;

[0031] Figure 4A schematic diagram of the structure of a bidirectional long short-term memory neural network provided in an embodiment of the present application;

[0032] Figure 5 A schematic diagram of a flow chart for solving the optimal control parameters of a proportional resonant controller provided in an embodiment of the present application;

[0033] Figure 6 A principle block diagram of a grid-type static VAR generator parameter optimization system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with examples and drawings. The illustrative implementation scheme of the present application and its description are only used to explain the present application and are not intended to limit the present application.

[0035] It should be noted that the terms "include" or "may include" used in various embodiments of the present application indicate the presence of the function, operation or element applied for, and do not limit the addition of one or more functions, operations or elements. In addition, as used in various embodiments of the present application, the terms "include", "have" and their cognates are only intended to indicate specific features, numbers, steps, operations, elements, components or a combination of the foregoing items, and should not be understood as first excluding the presence of one or more other features, numbers, steps, operations, elements, components or a combination of the foregoing items or the possibility of adding one or more features, numbers, steps, operations, elements, components or a combination of the foregoing items.

[0036] In various embodiments of the present application, the expression "or" or "at least one of B or / and C" includes any combination or all combinations of the words listed at the same time. For example, the expression "B or C" or "at least one of B or / and C" may include B, may include C, or may include both B and C.

[0037] It should be understood that terms such as "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly indicate the number of technical features indicated. Therefore, features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0038] Please refer to Figure 1 , Figure 1 A schematic diagram of a flow chart of a parameter optimization method for a grid-connected static VAR generator provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes:

[0039] S101, randomly generate multiple groups of control parameters of the current inner loop controller of the grid-type static VAR generator; wherein, the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between the input signal and the output signal of the proportional resonant controller with a generalized integral link.

[0040] In this embodiment, the control loop of the grid-type static VAR generator is composed of a power synchronization link, a voltage outer loop and a current inner loop. Secondly, most of the controllers of the current inner loop provided by the prior art are proportional integral controllers, i.e., PI controllers. However, it is difficult for the PI controller to take into account both the stability margin and the control performance of the grid-type static VAR generator when designing parameters. If the parameters are not designed properly, it is easy to cause oscillation in the power system. Therefore, this embodiment proposes to use a proportional resonant controller with a harmonic compensation link for the current inner loop, which is composed of a generalized integral link and a specific subharmonic compensator, so as to solve the problem that the current inner loop controller of the current grid-type static VAR generator has a weak ability to suppress the DC component and harmonic interference on the grid side.

[0041] like Figure 2 As shown, i 1 is the input signal of the proportional resonant controller with harmonic compensation link, i 2 is the output signal of the proportional resonant controller with harmonic compensation link, H c (s) is the transfer function of a specific subharmonic compensator, f is the grid frequency, λ is the adjustment coefficient of the proportional resonant controller with harmonic compensation link. Figure 2 The control block diagram of the proportional resonant controller with harmonic compensation link shown in the figure can derive the transfer function of the proportional resonant controller with harmonic compensation link The expression is as follows: (1), where k x represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, represents the bandwidth of a particular subharmonic compensator, is the angular frequency, k is the harmonic frequency to be compensated, n is the upper limit of the harmonic frequency to be compensated, for kThe angular frequency value of the subharmonic, s is the Laplace operator of the transfer function.

[0042] For the proportional resonant controller provided in this embodiment, it can flexibly select the frequency signal to be compensated according to the main frequency distribution of the grid-side harmonics connected to the actual converter. In order to illustrate the advantages of the proportional resonant controller with harmonic compensation link proposed in this embodiment, the transfer function of the traditional PR controller is given as follows: (2), where k p is the proportional coefficient of the traditional PR controller, k r is the resonance coefficient of the traditional PR controller, is the angular frequency of the power grid.

[0043] It should be noted that the control parameters of the current inner loop controller include k x , k y , λ , A , , for the multiple control parameters of the current inner loop controller of the randomly generated grid-type static VAR generator, it is achieved by setting k x , k y , λ , A , Respective numerical ranges, and then multiple groups can be randomly generated within the numerical range k x , k y , λ , A , ;

[0044] like Figure 3 As shown, Figure 3 The Bode diagram of the traditional PI controller and the proportional resonant controller provided in this embodiment is given. Figure 3 The amplitude frequency curve and phase frequency curve of the conventional PR controller and the proportional resonant controller provided in this embodiment, respectively. The proportional resonant controller with harmonic compensation link provided in this embodiment selects to compensate the third, fifth and seventh harmonics on the grid side. Figure 3 It can be seen that the proportional resonant controller with a harmonic compensation link provided in this embodiment has a stronger ability to suppress DC offset and harmonic signals.

[0045] Therefore, this embodiment provides a technical means for replacing the controller of the current inner loop of the grid-type static VAR generator with a proportional resonant controller with a harmonic compensation link, which can solve the problem that the current inner loop controller of the current grid-type static VAR generator has a weak ability to suppress the DC component and harmonic interference on the grid side.

[0046] S102, using a bidirectional long short-term memory neural network to construct a first function mapping between the overshoot of the current inner loop controller and the short-circuit ratio of the power grid and control parameters, a second function mapping between the response time of the current inner loop controller and the short-circuit ratio of the power grid and control parameters, and a third function mapping between the stability margin of the grid-forming static VAR generator and the equivalent inductance of the power grid and control parameters.

[0047] In this embodiment, the bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network. Figure 4 , x t-1 , x t , x t+1 Respectively represent the input signals of the previous moment, the current moment, and the next moment; h t-1 , h t

[0048] , h t+1 Respectively represent the encoded hidden state vectors of the previous moment, the current moment, and the next moment; the forward LSTM is a forward long short-term memory neural network, and the backward LSTM is a backward long short-term memory neural network, which consists of a forget gate, an input gate, and an output gate. The bidirectional long short-term memory neural network can encode information from backward to forward, and the two LSTMs can be trained simultaneously by reversely inputting the forward sequence into the backward sequence. After generating a large number of training samples through batch simulation of the grid-type static VAR generator, the overshoot of the current inner loop control can be obtained by using the bidirectional long short-term memory neural network architecture. x 1 and response time x 2. The functional mapping relationship between the short-circuit ratio and the control parameters, as well as the functional mapping relationship between the stability margin of the grid-type static VAR generator and the short-circuit ratio and the control parameters of the power grid.

[0049] According to the transfer function of the proportional resonant controller described above From the expression, we can know that in order to obtain a proportional resonant controller with a harmonic compensation link, the control parameters that need to be determined include k x , k y, λ , A , These control parameters directly determine the overshoot and response time of the current inner loop control. Therefore, this embodiment adopts a bidirectional long short-term memory neural network architecture to obtain the overshoot of the current inner loop control. x 1 and response time x 2. The functional mapping relationship between the short-circuit ratio and the control parameters is proposed, so that the proportional resonant controller used in the grid-type static VAR generator can maintain good dynamic response characteristics.

[0050] Among them, the expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0051] The expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f 2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0052] The expression of the third function mapping is: ;in, represents the stability margin of the grid-connected static VAR generator, It indicates the impedance phase of the grid-type static VAR generator corresponding to the intersection frequency of the amplitude-frequency characteristic curve of the grid-type static VAR generator impedance and the grid equivalent impedance. SVG is the grid-type static VAR generator. represents the phase of the equivalent impedance of the power grid, L g is the equivalent inductance of the power grid, Indicates the equivalent inductance of the power grid at the current moment. After obtaining the equivalent inductance, the short-circuit ratio of the power grid at the current moment can be calculated. Input the first function mapping and the second function mapping, short circuit ratio The calculation method is as follows: (3), where f 0 represents the base frequency of the power grid, P N represents the capacity of the grid-connected static VAR generator, U g Indicates the voltage of the actual power grid to which the grid-forming static VAR generator is connected.

[0053] S103, using a particle swarm algorithm to randomly select multiple groups of control parameters, input the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment into the first function mapping and the second function mapping respectively, and input the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and output the optimal control parameters of the proportional resonant controller with a harmonic compensation link; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

[0054] In this embodiment, in order to make the proportional resonance control described in the above embodiment of this embodiment adaptable to the scenario of power grid strength change, the method flow of adaptively solving the parameters of the proportional resonance controller with harmonic compensation link proposed in this embodiment is as follows: Figure 5 As shown, the method combines the impedance measurement results on the grid side to adjust the control parameters of the proportional resonant controller with harmonic compensation link used in the grid-type static VAR generator in real time. The implementation steps of the method are as follows:

[0055] Step 1: Calculate the short-circuit ratio of the power grid by: using a small signal injection method to measure the impedance of the power grid to obtain an impedance measurement result; and calculating the short-circuit ratio of the power grid based on the impedance measurement result. The impedance online measurement calculation method is as follows:

[0056] (4), where is the magnitude of the equivalent impedance of the power grid; is the frequency of the disturbance signal injected into the command current; is the phase of the grid-connected current with the same frequency as the disturbance signal; is the phase of the common coupling point voltage with the same frequency as the disturbance signal.

[0057] It should be understood that calculating the equivalent inductance of the power grid based on the impedance online measurement calculation method is a prior art, so this embodiment does not make redundant descriptions thereof. Accordingly, since the impedance online measurement is a real-time measurement, the short-circuit ratio of the power grid at the current moment can be calculated based on the result of the impedance online measurement calculation method.

[0058] Step 2: Particle swarm algorithm is an optimization algorithm based on swarm intelligence. It imitates the behavior of bird flocks foraging and seeks the optimal solution to the problem through information sharing and collaboration between individuals in the group. In the particle swarm algorithm, each solution is regarded as a particle in the search space. All particles have a fitness value, which is given by the optimization function and indicates the quality of the solution. Each particle also has a speed, which determines the direction and distance of the particle's movement in the search space.

[0059] The variable weight crossover particle swarm algorithm is an optimization algorithm improved on the basis of the variable weight particle swarm algorithm. The algorithm dynamically adjusts the inertia weight of the particles and introduces the crossover operation to enhance the algorithm's global search ability and the ability to jump out of the local optimal solution. The variable weight particle swarm algorithm adaptively adjusts the algorithm weight according to the distance between the particle and the optimal particle to avoid the algorithm from falling into the local optimal solution. At the same time, the crossover mutation operation in the genetic algorithm is introduced into the variable weight crossover particle swarm algorithm to prevent the problem of premature convergence in the algorithm solution. In this embodiment, the optimization objectives of the variable weight crossover particle swarm algorithm include the following three aspects:

[0060] (1) The difference between the stability margin of the grid-connected static VAR generator and the expected stability margin Minimum, the difference can be expressed as: After determining the short-circuit ratio of the power grid, combined with the impedance stability criterion, the impedance phase calculation can be solved

[0061] ; To construct the expected stability margin of the network SVG and to take into account both the steady-state and dynamic control performance of the system, The angle can be selected between 15° and 35°.

[0062] (2) Overshoot of the current inner loop control x 1Min.

[0063] (3) Response time of current inner loop control x 2Min.

[0064] Step 3: Combine the first function mapping, the second function mapping, the third function mapping and the impedance stability criterion to verify the control parameters output by the variable weight crossover particle swarm algorithm kA , k B , λ , A , Whether the control requirements of the grid-type static VAR generator are met in that the stability margin difference, overshoot and response time are each at the minimum value, if the requirements are met, the final control parameter result is output; if not, the algorithm parameters are updated to continue to execute step 2 until the control requirements are met, and the optimal control parameters of the proportional resonant controller with harmonic compensation link are obtained.

[0065] Therefore, this embodiment describes a parameter optimization method for a proportional resonant controller with a harmonic compensation link, which can adjust the controller parameters in real time according to the strength of the power grid, thereby improving the adaptability of the control parameters of the proportional resonant controller with a harmonic compensation link to the power grid strength change scenario. By measuring the power grid impedance in real time and adjusting the control parameters of the proportional resonant controller with a harmonic compensation link online, it is ensured that the grid-type static VAR generator can always maintain a good stability margin and dynamic response performance when the short-circuit ratio changes over a wide range.

[0066] In summary, a parameter optimization method for a grid-type static VAR generator described in the above embodiments, first, a proportional resonant controller based on a generalized integral link and a harmonic compensation link is designed, so that the proportional resonant controller has a stronger suppression capability for the DC component and harmonic interference on the grid side. Therefore, it is used for the current inner loop control of the grid-type static VAR generator, which can improve the suppression capability of the grid-type static VAR generator for grid-side harmonics and DC offset. Secondly, in order to optimize the parameter design of the proportional resonant controller and overcome the problem that the performance of the current inner loop controller of the grid-type static VAR generator is difficult to quantify and evaluate, the present application uses a bidirectional long short-term memory neural network to construct a functional mapping relationship between the stability margin difference, overshoot and response time with respect to the short-circuit ratio of the grid and the control parameters of the proportional resonant controller. Finally, a method for solving the control parameters of a proportional resonant controller that can adapt to changes in the short-circuit ratio is constructed. That is, a particle swarm algorithm is used to randomly select multiple groups of randomly generated control parameters to obtain random selection results of multiple groups of control parameters. The random selection results and the short-circuit ratio and equivalent inductance of the power grid at the current moment are respectively input into the function mapping relationship. When the obtained stability margin difference is the smallest, the overshoot is the smallest, and the response time reaches the minimum, the optimal control parameters of the proportional resonant controller with a harmonic compensation link can be output. Based on the optimal control parameters, it can be ensured that the grid-forming static VAR generator can always maintain a good stability margin and dynamic response performance when the short-circuit ratio changes in a wide range, that is, when a transient occurs in the power grid, thereby improving the real-time adjustment capability of the grid-forming static VAR generator.

[0067] Please refer to Figure 6 , Figure 6A principle block diagram of a grid-type static VAR generator parameter optimization system provided in an embodiment of the present application is shown in FIG. Figure 6 As shown, the system includes:

[0068] A control parameter acquisition module 610 is used to randomly generate multiple groups of control parameters of the current inner loop controller of the grid-forming static VAR generator; wherein the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between the input signal and the output signal of the proportional resonant controller with a generalized integral link;

[0069] A function mapping construction module 620 is used to construct a first function mapping between the overshoot of the current inner loop controller and the short circuit ratio of the power grid and the control parameters, a second function mapping between the response time of the current inner loop controller and the short circuit ratio of the power grid and the control parameters, and a third function mapping between the stability margin of the grid-forming static VAR generator and the equivalent inductance of the power grid and the control parameters by using a bidirectional long short-term memory neural network;

[0070] The control parameter optimization module 630 is used to randomly select multiple groups of control parameters using a particle swarm algorithm, input the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment into the first function mapping and the second function mapping respectively, and input the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and output the optimal control parameters of the proportional resonant controller with a harmonic compensation link; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

[0071] Specifically, a grid-type static VAR generator parameter optimization system in an embodiment of the present application is Figure 1 The grid-type static VAR generator parameter optimization method shown is a technical solution based on the same inventive concept. Through the detailed description of the grid-type static VAR generator parameter optimization method provided in the above embodiment, those skilled in the art can clearly understand the implementation process of a grid-type static VAR generator parameter optimization system in this embodiment. Therefore, for the sake of brevity of the specification, the same implementation method will not be repeated below.

[0072] Accordingly, a parameter optimization system for a grid-type static VAR generator provided in an embodiment of the present application firstly designs a proportional resonant controller based on a generalized integral link and a harmonic compensation link, so that the proportional resonant controller has a stronger suppression capability for the DC component and harmonic interference on the grid side, and therefore uses it for the current inner loop control of the grid-type static VAR generator, which can improve the suppression capability of the grid-type static VAR generator for grid-side harmonics and DC offset. Secondly, in order to optimize the parameter design of the proportional resonant controller and overcome the problem that the performance of the current inner loop controller of the grid-type static VAR generator is difficult to quantify and evaluate, the present application uses a bidirectional long short-term memory neural network to construct a functional mapping relationship between the stability margin difference, overshoot and response time with respect to the short-circuit ratio of the grid and the control parameters of the proportional resonant controller. Finally, a method for solving the control parameters of a proportional resonant controller that can adapt to changes in the short-circuit ratio is constructed. That is, a particle swarm algorithm is used to randomly select multiple groups of randomly generated control parameters to obtain random selection results of multiple groups of control parameters. The random selection results and the short-circuit ratio and equivalent inductance of the power grid at the current moment are respectively input into the function mapping relationship. When the obtained stability margin difference is the smallest, the overshoot is the smallest, and the response time reaches the minimum, the optimal control parameters of the proportional resonant controller with a harmonic compensation link can be output. Based on the optimal control parameters, it can be ensured that the grid-forming static VAR generator can always maintain a good stability margin and dynamic response performance when the short-circuit ratio changes in a wide range, that is, when a transient occurs in the power grid, thereby improving the real-time adjustment capability of the grid-forming static VAR generator.

[0073] In some embodiments, the bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network.

[0074] In some embodiments, the expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator

[0075] In some embodiments, the expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f 2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

[0076] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for optimizing parameters of a grid-connected static VAR generator, characterized in that: include: Randomly generate multiple groups of control parameters of the current inner loop controller of the grid-type static VAR generator; wherein the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between the input signal and the output signal of the proportional resonant controller with a generalized integral link; A bidirectional long short-term memory neural network is used to construct a first function mapping between the overshoot of the current inner loop controller and the short-circuit ratio of the power grid and the control parameters, a second function mapping between the response time of the current inner loop controller and the short-circuit ratio of the power grid and the control parameters, and a third function mapping between the stability margin of the grid-forming static VAR generator and the equivalent inductance of the power grid and the control parameters; A particle swarm algorithm is used to randomly select multiple groups of control parameters, and the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment are input into the first function mapping and the second function mapping respectively, and the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment are input into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and the optimal control parameters of the proportional resonant controller with a harmonic compensation link are output; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

2. A parameter optimization method for a grid-connected static VAR generator according to claim 1, characterized in that: The bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network.

3. A parameter optimization method for a grid-connected static VAR generator according to claim 1, characterized in that: The expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

4. A parameter optimization method for a grid-connected static VAR generator according to claim 1, characterized in that: The expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f 2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

5. The method for optimizing parameters of a grid-connected static VAR generator according to claim 1, characterized in that: Get the short-circuit ratio of the power grid at the current moment, including: The impedance of the power grid is measured by a small signal injection method to obtain impedance measurement results; The short-circuit ratio of the power grid at the current moment is calculated based on the impedance measurement results.

6. A parameter optimization method for a grid-connected static VAR generator according to claim 1, characterized in that: The particle swarm algorithm adopts a variable weight crossover particle swarm algorithm.

7. A grid-type static VAR generator parameter optimization system, characterized in that the system include: A control parameter acquisition module, used for randomly generating multiple groups of control parameters of a current inner loop controller of a grid-forming static VAR generator; wherein the current inner loop controller is a proportional resonant controller with a harmonic compensation link, and the proportional resonant controller with a harmonic compensation link is obtained by connecting a specific subharmonic compensator between an input signal and an output signal of a proportional resonant controller with a generalized integral link; A function mapping construction module is used to construct a first function mapping between an overshoot of an inner current loop controller and a short circuit ratio of a power grid and a control parameter, a second function mapping between a response time of the inner current loop controller and a short circuit ratio of a power grid and a control parameter, and a third function mapping between a stability margin of a grid-forming static VAR generator and an equivalent inductance of a power grid and a control parameter by using a bidirectional long short-term memory neural network; The control parameter optimization module is used to randomly select multiple groups of control parameters by using a particle swarm algorithm, input the random selection results of the multiple groups of control parameters and the short-circuit ratio of the power grid at the current moment into the first function mapping and the second function mapping respectively, and input the random selection results of the multiple groups of control parameters and the equivalent inductance of the power grid at the current moment into the third function mapping, until the stability margin difference, overshoot and response time reach the minimum value, and output the optimal control parameters of the proportional resonant controller with a harmonic compensation link; wherein the stability margin difference is obtained by subtracting the stability margin threshold from the stability margin.

8. A grid-connected static VAR generator parameter optimization system according to claim 7, characterized in that: The bidirectional long short-term memory neural network is composed of a forward long short-term memory neural network and a backward long short-term memory neural network.

9. A grid-connected static VAR generator parameter optimization system according to claim 7, characterized in that: The expression of the first function mapping is: ;in, represents the overshoot of the current inner loop controller, f 1 represents the function mapping between overshoot, short-circuit ratio of the power grid and control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

10. A grid-connected static VAR generator parameter optimization system according to claim 7, characterized in that: The expression of the second function mapping is: ;in, represents the response time of the current inner loop controller, f 2 represents the function mapping between the response time, the short-circuit ratio of the power grid, and the control parameters. represents the short-circuit ratio of the power grid, k x Represents the internal gain of the proportional resonant controller with harmonic compensation link, k y represents the gain inside the generalized integral link, λ represents the adjustment coefficient of the proportional resonant controller with harmonic compensation link, A represents the gain of a particular subharmonic compensator, Indicates the bandwidth of a particular subharmonic compensator.

Images