A method, device, equipment and storage medium for variable pitch control and frequency modulation of a wind turbine generator set
By calculating the maximum voltage deviation ratio and determining the control parameters by preset pitch control model, the voltage fluctuation problem caused by frequency regulation of the double-feed asynchronous wind turbine is solved, and the stability of the microgrid is improved.
Patent Information
- Application Number
- CN202311301707.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-09
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2043-10-09
AI Technical Summary
Dual-feed asynchronous wind turbines will cause large voltage fluctuations during the active jump growth of frequency modulation, affecting the stability of the microgrid.
Through the preset pitch control model, initial characteristics are generated based on the operating data of the wind power plant, input the vertex diagram, generator initial characteristics, paddle initial characteristics, converter initial characteristics, calculate the maximum voltage deviation ratio, and backpropagation Iterative training determines the pitch angle control parameters and reactive power control parameters, and performs pitch control frequency modulation.
It improves the stability of the microgrid, reduces voltage fluctuations, and enhances the operating safety of the wind turbine.
Smart Images

Figure CN117439198B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wind power generation, and in particular to a method, device, equipment and storage medium for variable pitch control and frequency modulation of a wind turbine set. Background Art
[0002] Doubly-fed asynchronous wind turbines are the most widely used wind turbines, consisting of a wound-type asynchronous generator with a stator winding directly connected to a fixed-frequency three-phase grid and a bidirectional back-to-back IGBT voltage source converter installed on the rotor winding. With the increasing size of wind turbines, the blades and loads are getting bigger and bigger, which puts higher requirements on the safety performance of the variable pitch system.
[0003] In the prior art, the variable pitch control technology of the doubly-fed asynchronous wind turbine mainly adjusts the angle of the blades in real time according to the change of wind speed at or near the rated wind speed, controls the absorbed mechanical energy, and reduces the impact of wind on the variable speed constant frequency wind turbine while ensuring the maximum energy. When shutting down, the blades are adjusted to the feathering position through the variable pitch system to achieve aerodynamic braking and ensure the safe shutdown of the wind turbine.
[0004] However, the reactive power of the doubly-fed asynchronous wind turbine generator depends on the stator reactive power and the converter reactive power. In the process of a jump in the frequency modulation active power, it will cause a drastic change in reactive power. At this time, the upper limit of the stator reactive power will decrease with the increase in active power. The converter needs to provide a drastic change in reactive power. Compared with the stator output reactive power, the converter will cause a large voltage fluctuation. As the number of wind turbines in the power plant increases, the impact on the stability of the microgrid will be greater. Summary of the invention
[0005] In view of this, it is necessary to provide a wind turbine pitch control frequency modulation method, device, equipment and storage medium to solve the problem in the prior art that the double-fed asynchronous wind generator will cause large voltage fluctuations and affect the stability of the microgrid during the jump-like increase in frequency modulation active power.
[0006] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:
[0007] In a first aspect, the present invention provides a method for variable pitch control and frequency modulation of a wind turbine generator set, comprising:
[0008] Generate initial characteristics of the generator, initial characteristics of the blades, and initial characteristics of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions;
[0009] Inputting the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a maximum voltage deviation ratio;
[0010] Performing back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and reactive power control parameters;
[0011] The wind turbine is subjected to variable pitch control and frequency modulation based on the pitch angle control parameters and the reactive power control parameters.
[0012] In some possible implementations, the preset pitch control model includes: a graph network layer, a parameter generation layer and an MLP layer; the vertex graph of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter are input into the preset pitch control model to obtain the maximum voltage deviation ratio, including:
[0013] The graph network layer generates a generator intermediate vector and a blade intermediate vector according to the vertex graph, the generator initial features, the blade initial features, and the converter initial features, and inputs the generator intermediate vector and the blade intermediate vector into the parameter generation layer;
[0014] The parameter generation layer generates a unified control parameter vector according to the generator intermediate vector and the blade intermediate vector, and inputs the unified control parameter vector into the MLP layer;
[0015] The MLP layer obtains the maximum voltage deviation ratio during the period from when the frequency regulation of the wind turbine is started to when the voltage at the grid connection point returns to a steady state according to the unified control parameter vector.
[0016] In some possible implementations, before the generator intermediate vector and the blade intermediate vector are input into the parameter generation layer, the step includes: concatenating the generator intermediate vector and the blade intermediate vector to obtain an intermediate vector vector.
[0017] In some possible implementations, before the generator intermediate vector and the blade intermediate vector are input into the parameter generation layer, the following steps are also included:
[0018] The parameter generation layer is connected to the preset classifier to classify and obtain the pre-trained value of the dynamic frequency deviation ratio of the wind power plant microgrid;
[0019] Pre-training is performed with minimizing the difference between the pre-trained value of the dynamic frequency deviation ratio and the true value of the dynamic frequency deviation ratio as the training objective.
[0020] In some possible implementations, a vertex graph of the wind turbine generator set is constructed, including:
[0021] Determine the generator vertex, blade vertex, and converter vertex of the wind turbine;
[0022] According to the connection relationship between the wind turbine generator, the wind turbine blades and the wind turbine converter, the generator vertices, the blade vertices and the converter vertices are connected to obtain a vertex graph of the wind turbine.
[0023] In some possible implementations, back propagation iterative training is performed on the voltage maximum deviation ratio to determine the pitch angle control parameter and the reactive power control parameter, including:
[0024] The maximum voltage deviation ratio is back-propagated to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector;
[0025] When the preset termination condition is met, the i-1th component of the unified control parameter vector updated in the last iteration is used as the pitch angle control parameter of the i-th wind turbine group, and the i-th component is used as the reactive power control parameter of the i-th wind turbine group.
[0026] In some possible implementations, the maximum voltage deviation ratio is back-propagated to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector, including: performing back-propagation iterative training to update the unified control parameter vector based on the difference between the maximum voltage deviation ratio and a preset voltage deviation ratio threshold as a loss value.
[0027] In a second aspect, the present invention further provides a wind turbine pitch control frequency modulation device, comprising:
[0028] A feature generation module, used to generate initial features of the generator, initial features of the blades, and initial features of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions;
[0029] A model calculation module is used to input the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a maximum voltage deviation ratio;
[0030] An iterative training module is used to perform back-propagation iterative training on the voltage maximum deviation ratio to determine a pitch angle control parameter and a reactive power control parameter;
[0031] The control module is used to perform variable pitch control and frequency modulation on the wind turbine generator set based on the pitch angle control parameter and the reactive power control parameter.
[0032] In a third aspect, the present invention further provides a wind turbine pitch control frequency modulation device, comprising a memory and a processor, wherein:
[0033] Memory, used to store programs;
[0034] The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the wind turbine pitch control frequency modulation method in any of the above implementations.
[0035] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the wind turbine pitch control and frequency modulation method in any of the above-mentioned implementation modes.
[0036] The beneficial effects of adopting the above embodiment are: the present invention relates to a method, device, equipment and storage medium for variable pitch control and frequency modulation of a wind turbine set, the method comprising: generating the initial characteristics of the generator, the initial characteristics of the blades and the initial characteristics of the converter of the wind turbine set according to the operation data of the wind power plant under preset conditions; inputting the vertex graph, the initial characteristics of the generator, the initial characteristics of the blades and the initial characteristics of the converter of the wind turbine set into the preset variable pitch control model to obtain the maximum voltage deviation ratio; performing back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters; performing variable pitch control and frequency modulation on the wind turbine set based on the pitch angle control parameters and the reactive power control parameters. The present invention obtains multiple initial characteristics of the wind turbine set according to the operation data of the wind power plant under preset conditions, and first calculates the maximum voltage deviation ratio through the preset variable pitch control model, and then performs back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters, so as to realize reactive frequency modulation of the wind turbine set, thereby improving the stability of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 A schematic flow chart of an embodiment of a method for variable pitch control and frequency modulation of a wind turbine provided by the present invention;
[0038] Figure 2 The present invention provides Figure 1 A schematic diagram of a flow chart of an embodiment of step S102;
[0039] Figure 3 It is a voltage schematic diagram of an embodiment of a grid connection point in a wind power plant control process provided by the present invention;
[0040] Figure 4 A schematic structural diagram of an embodiment of a wind turbine pitch control and frequency modulation device provided by the present invention;
[0041] Figure 5 A schematic diagram of the structure of a wind turbine pitch control and frequency modulation device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0042] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not used to limit the scope of the present invention.
[0043] In the description of the present application, “plurality” means two or more than two, unless otherwise clearly and specifically defined.
[0044] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0045] The present invention provides a method, device, equipment and storage medium for variable pitch control and frequency modulation of a wind turbine generator set, which are respectively described below.
[0046] See also Figure 1 , Figure 1 A flow chart of an embodiment of a method for variable pitch control and frequency modulation of a wind turbine set provided by the present invention is provided. A specific embodiment of the present invention discloses a method for variable pitch control and frequency modulation of a wind turbine set, comprising:
[0047] S101, generating initial characteristics of a generator, initial characteristics of a blade, and initial characteristics of a converter of a wind turbine according to operation data of a wind power plant under preset conditions;
[0048] S102, inputting the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a maximum voltage deviation ratio;
[0049] S103, performing back propagation iterative training on the voltage maximum deviation ratio to determine a pitch angle control parameter and a reactive power control parameter;
[0050] S104: performing variable pitch control and frequency modulation on the wind turbine generator set based on the pitch angle control parameter and the reactive power control parameter.
[0051] In the above embodiments, the preset condition refers to the frequency regulation condition of the wind power plant. The frequency regulation conditions of different wind power plants are different. Generally, the deviation value of the microgrid frequency drop is used as a condition. The frequency regulation condition is generally that the difference between the microgrid frequency and the preset reference frequency is greater than the set deviation limit. It can be understood that the deviation limit can be adjusted according to actual needs, and the present invention does not impose further restrictions on this. The time point for collecting operating data is the time point when the frequency regulation is started.
[0052] Initial characteristics of wind turbine generators
[0053] , where ω r represents the rotor speed, J represents the moment of inertia, ω0 represents the rated speed, Pr Indicates the rotor active power, P s Indicates stator active power, Q s represents the reactive power of the wind turbine, Hz0 represents the steady-state frequency of the microgrid before the frequency regulation is started, X d Indicates the stator reactance of the generator, L m represents the stator-rotor mutual reactance, S c Indicates the slip rate, R z Indicates rotor resistance, I d represents the component of the rotor current on the d-axis, I q represents the component of the rotor current on the q axis, φ d represents the rotor self-inductance flux component on the d-axis, φ q represents the component of the rotor self-inductance flux on the q axis, V0 represents the voltage at the grid connection point of the wind power plant;
[0054] Initial characteristics of wind turbine blades Where β0 represents the pitch angle before frequency regulation is started, γ0 represents the tip speed ratio before frequency regulation is started, ρ represents the air density, V represents the air velocity before entering the wind swept area, C P represents the wind energy utilization coefficient, and R represents the radius of the wind wheel;
[0055] Initial characteristics of wind turbine converters Among them, Hz1 represents the grid-side frequency of the converter, Hz2 represents the machine-side frequency of the converter, I1 represents the grid-side current of the converter, I2 represents the machine-side current of the converter, V1 represents the grid-side voltage of the converter, V2 represents the machine-side voltage of the converter, C1 represents the grid-side capacity of the converter, and C2 represents the machine-side capacity of the converter.
[0056] It is understandable that the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter of the wind turbine set also belong to the operating data of the wind power plant. Therefore, after determining the operating data of the wind power plant, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter of the wind turbine set can be directly determined.
[0057] If only variable pitch control frequency modulation is considered, it is easy to cause instability of the microgrid. The present invention further improves the stability of the microgrid by comprehensive control of pitch angle control parameters and reactive power control parameters.
[0058] Compared with the prior art, the present embodiment provides a method for variable pitch control and frequency modulation of a wind turbine set, the method comprising: generating the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter of the wind turbine set according to the operating data of the wind power plant under preset conditions; inputting the vertex graph, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter of the wind turbine set into the preset variable pitch control model to obtain the maximum voltage deviation ratio; performing back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters; and performing variable pitch control and frequency modulation on the wind turbine set based on the pitch angle control parameters and the reactive power control parameters. The present invention obtains multiple initial characteristics of the wind turbine set according to the operating data of the wind power plant under preset conditions, and first calculates the maximum voltage deviation ratio through the preset variable pitch control model, and then performs back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters, so as to realize reactive frequency modulation of the wind turbine set, thereby improving the stability of the microgrid.
[0059] See also Figure 2 , Figure 2 The present invention provides Figure 1 In some embodiments of the present invention, the preset pitch control model includes: a graph network layer, a parameter generation layer and an MLP layer; the vertex graph of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter are input into the preset pitch control model to obtain the maximum voltage deviation ratio, including:
[0060] S201, the graph network layer generates a generator intermediate vector and a blade intermediate vector according to the vertex graph, the generator initial features, the blade initial features, and the converter initial features, and inputs the generator intermediate vector and the blade intermediate vector to the parameter generation layer;
[0061] S202, the parameter generation layer generates a unified control parameter vector according to the generator intermediate vector and the blade intermediate vector, and inputs the unified control parameter vector to the MLP layer;
[0062] S203, the MLP layer obtains the maximum voltage deviation ratio in the time period from when the frequency regulation of the wind turbine generator set is started to when the voltage at the grid connection point returns to a steady state according to the unified control parameter vector.
[0063] In the above embodiment, the input of the graph network layer is the vertex graph, the initial features of the generator, the initial features of the blades, and the initial features of the converter. Then, the intermediate vector of the generator and the intermediate vector of the blades are calculated based on the input. The calculation formula of the graph network layer is as follows:
[0064] in, represents the middle vector of the i-th vertex, and Represent the initial features of the i-th and j-th vertices respectively, W represents the weight parameter of the first hidden layer, represents the weight vector of the first hidden layer, N i Represents the set of vertices directly connected to the i-th vertex; the initial features of the vertex are the initial features of the generator or blade or converter it represents, and the intermediate vector of the vertex is the intermediate vector of the generator or blade it represents.
[0065] It should be noted that the initial features of the generator, the initial features of the blades, and the initial features of the converter need to be mapped to the same dimension through the linear mapping layer before they can be input into the graph network layer for calculation and determination of the intermediate vector.
[0066] In addition to generating a unified control parameter vector, the parameter generation layer is also used for feedback iterative training in the future to update the unified control parameter vector to determine the pitch angle control parameters and reactive power control parameters.
[0067] The output of the MLP layer is the maximum voltage deviation ratio during the period from when the microgrid frequency regulation is started to when the grid-connected point voltage returns to a steady state. The calculation formula for the maximum voltage deviation ratio Δ0 is as follows:
[0068] Among them, Δ max It indicates the maximum absolute value of the voltage fluctuation during the period from when the frequency regulation is started to when the grid-connected point voltage returns to a steady state. sec Indicates the maximum inertia load of the wind turbine (design value), U sec Indicates the voltage at the grid connection point when it is in steady state.
[0069] In some embodiments of the present invention, before the generator intermediate vector and the blade intermediate vector are input into the parameter generation layer, the method includes: concatenating the generator intermediate vector and the blade intermediate vector to obtain an intermediate vector vector.
[0070] In the above embodiment, the generator intermediate vector and the blade intermediate vector need to be spliced before being input into the MLP layer for calculation. The spliced vector is expressed as in represents the intermediate vector between the generator and the blades of the first wind turbine, Represents the intermediate vector between the generator and blades of the nth wind turbine, where n is the total number of wind turbines.
[0071] In some embodiments of the present invention, before inputting the generator intermediate vector and the blade intermediate vector into the parameter generation layer, the method further includes:
[0072] The parameter generation layer is connected to the preset classifier to classify and obtain the pre-trained value of the dynamic frequency deviation ratio of the wind power plant microgrid;
[0073] Pre-training is performed with minimizing the difference between the pre-trained value of the dynamic frequency deviation ratio and the true value of the dynamic frequency deviation ratio as the training objective.
[0074] In the above embodiment, the parameter generation layer needs to be pre-trained to generate a unified control parameter vector. During the pre-training, it needs to be connected to a preset classifier, and the classification label of the preset classifier represents the dynamic frequency deviation ratio of the wind power plant microgrid; the training goal is to minimize the difference between the pre-training value of the dynamic frequency deviation ratio output by the pre-training and the true value of the dynamic frequency deviation ratio;
[0075] The calculation formula of dynamic frequency deviation ratio is as follows:
[0076] Among them, H1 represents the dynamic frequency deviation, H w represents the steady-state frequency before the microgrid frequency changes, σ H Indicates the dynamic frequency deviation ratio.
[0077] It should be noted that the dynamic frequency deviation refers to the maximum difference between the frequency of the microgrid during the process of dynamic recovery of the frequency to a steady state and the steady state frequency before the change when the frequency changes.
[0078] In some embodiments of the present invention, it also includes constructing a vertex graph of the wind turbine generator set, specifically including:
[0079] Determine the generator vertex, blade vertex, and converter vertex of the wind turbine;
[0080] According to the connection relationship between the wind turbine generator, the wind turbine blades and the wind turbine converter, the generator vertices, the blade vertices and the converter vertices are connected to obtain a vertex graph of the wind turbine.
[0081] In the above embodiment, the vertex graph of the wind turbine group includes vertices representing the wind turbine generator, the wind turbine blades, and the wind turbine converter respectively. When there is a connection relationship between the wind turbine generator, the wind turbine blades, and the wind turbine converter, the corresponding vertices will be connected by edges. The existence of an edge connection between two vertices indicates that the generator or blade or converter has a direct mechanical or grid line connection on the wind farm wind power plant grid.
[0082] In some embodiments of the present invention, performing back propagation iterative training on the voltage maximum deviation ratio to determine the pitch angle control parameter and the reactive power control parameter includes:
[0083] The maximum voltage deviation ratio is back-propagated to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector;
[0084] When the preset termination condition is met, the i-1th component of the unified control parameter vector updated in the last iteration is used as the pitch angle control parameter of the i-th wind turbine group, and the i-th component is used as the reactive power control parameter of the i-th wind turbine group.
[0085] In the above embodiment, the back propagation iterative training in the present invention is similar to feedback regulation, and the output voltage maximum deviation ratio is input into the parameter generation layer for back propagation iterative training, and the unified control parameter vector is continuously updated through multiple cycles to determine an accurate unified control parameter vector.
[0086] The preset termination condition is that the loss value of the iterative training is less than the preset loss value. The unified control parameter vector has multiple components. This application uses the i-1th component as the pitch angle control parameter of the i-th wind turbine group, and the i-th component as the reactive power control parameter of the i-th wind turbine group, i∈[2,2n], and i is an even number, and n is the total number of wind turbine groups. This application finally uses the pitch angle control parameter as the increment of the pitch angle, and the wind turbine group reactive power control parameter as the increment of the wind turbine group reactive power.
[0087] It should be noted that the preset loss value can be set according to actual needs, and the present invention does not impose further limitations on this.
[0088] In some embodiments of the present invention, the maximum voltage deviation ratio is back-propagated to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector, including: performing back-propagation iterative training to update the unified control parameter vector based on the difference between the maximum voltage deviation ratio and a preset voltage deviation ratio threshold as the loss value.
[0089] In the above embodiment, the back propagation iterative training is to make the maximum voltage deviation ratio gradually approach the preset voltage deviation ratio threshold value, and the unified control parameter vector corresponding to the preset voltage deviation ratio threshold value can realize reactive power regulation of the wind turbine.
[0090] It should be noted that the preset voltage deviation ratio threshold value needs to be set according to actual conditions, and the preset voltage deviation ratio threshold values of different wind turbines may be different, and the present invention does not impose further restrictions on this.
[0091] See also Figure 3 , Figure 3It is a voltage schematic diagram of an embodiment of a grid connection point in the control process of a wind power plant provided by the present invention. In a specific embodiment of the present invention, the wind turbine of the wind power plant is composed of a double-fed asynchronous wind turbine generator with a single unit capacity of 5MW, and the double-fed asynchronous wind turbine generator model is composed of a wind turbine, a double-fed generator, a pulse width modulation converter and a converter control system. The wind power plant has a total of 400 double-fed asynchronous wind turbine generators with a total installed capacity of 4000MW and a penetration rate of 15.7%.
[0092] At a constant wind speed of 12 m / s, when t=30s, the load in the system increases by 500 MW, and the frequency regulation is started. The dotted line is the control curve of the method of the present invention. Figure 3 The solid line part in the figure is a control method that adjusts the pitch angle based only on the frequency difference. The reactive power is automatically controlled by PI. The voltage fluctuation at the grid connection point during the regulation process of the dashed line part is significantly reduced and smoother, which improves the stability of the microgrid system. In addition, the difference between the maximum values of the dynamic frequency deviation during the two control methods can be controlled within 0.1Hz, which has little impact on frequency regulation. It can be seen that the single consideration of variable pitch control frequency modulation is likely to cause microgrid instability. The present invention further improves the stability of the microgrid through the comprehensive control of variable pitch control frequency modulation and reactive frequency modulation.
[0093] In order to better implement the wind turbine pitch control frequency modulation method in the embodiment of the present invention, based on the wind turbine pitch control frequency modulation method, correspondingly, please refer to Figure 4 , Figure 4 This is a schematic diagram of the structure of an embodiment of a wind turbine pitch control and frequency modulation device provided by the present invention. The embodiment of the present invention provides a wind turbine pitch control and frequency modulation device 400, including:
[0094] The feature generation module 410 is used to generate the initial features of the generator, the initial features of the blades, and the initial features of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions;
[0095] The model calculation module 420 is used to input the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a voltage maximum deviation ratio;
[0096] Iterative training module 430, used for performing back propagation iterative training on the voltage maximum deviation ratio to determine the pitch angle control parameter and the reactive power control parameter;
[0097] The control module 440 is used to perform variable pitch control and frequency modulation on the wind turbine generator set based on the pitch angle control parameter and the reactive power control parameter.
[0098] It should be noted here that: the device 400 provided in the above embodiment can implement the technical solutions described in the above method embodiments. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above method embodiments, which will not be repeated here.
[0099] See also Figure 5 , Figure 5 A schematic diagram of the structure of a wind turbine pitch control and frequency modulation device provided in an embodiment of the present invention. Based on the above-mentioned wind turbine pitch control and frequency modulation method, the present invention also provides a wind turbine pitch control and frequency modulation device, which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer, and a server. The wind turbine pitch control and frequency modulation device 500 includes a processor 510, a memory 520, and a display 530. Figure 5 Only some components of the wind turbine pitch control and frequency modulation device are shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components may be implemented instead.
[0100] In some embodiments, the memory 520 may be an internal storage unit of the wind turbine pitch control and frequency modulation device 500, such as a hard disk or memory of the wind turbine pitch control and frequency modulation device 500. In other embodiments, the memory 520 may also be an external storage device of the wind turbine pitch control and frequency modulation device 500, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the wind turbine pitch control and frequency modulation device 500. Further, the memory 520 may also include both the internal storage unit of the wind turbine pitch control and frequency modulation device 500 and an external storage device. The memory 520 is used to store application software and various data installed in the wind turbine pitch control and frequency modulation device 500, such as the program code for installing the wind turbine pitch control and frequency modulation device 500. The memory 520 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a wind turbine pitch control frequency modulation program 540 is stored in the memory 520, and the wind turbine pitch control frequency modulation program 540 can be executed by the processor 510, thereby realizing the wind turbine pitch control frequency modulation method of each embodiment of the present application.
[0101] In some embodiments, the processor 510 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 520, such as executing a wind turbine pitch control frequency modulation method.
[0102] In some embodiments, the display 530 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 530 is used to display information on the wind turbine pitch control frequency modulation device 500 and to display a visual user interface. The components 510-530 of the wind turbine pitch control frequency modulation device 500 communicate with each other via a system bus.
[0103] In one embodiment, when the processor 510 executes the wind turbine pitch control frequency modulation program 540 in the memory 520 , the steps in the above wind turbine pitch control frequency modulation method are implemented.
[0104] This embodiment further provides a computer-readable storage medium on which a wind turbine pitch control and frequency modulation program is stored. When the wind turbine pitch control and frequency modulation program is executed by a processor, the following steps are implemented:
[0105] Generate initial characteristics of the generator, initial characteristics of the blades, and initial characteristics of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions;
[0106] Inputting the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a maximum voltage deviation ratio;
[0107] Performing back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and reactive power control parameters;
[0108] The wind turbine is subjected to variable pitch control and frequency modulation based on the pitch angle control parameters and the reactive power control parameters.
[0109] In summary, the present embodiment provides a method, device, equipment and storage medium for variable pitch control and frequency modulation of a wind turbine set, the method comprising: generating the initial characteristics of the generator, the initial characteristics of the blades and the initial characteristics of the converter of the wind turbine set according to the operating data of the wind power plant under preset conditions; inputting the vertex graph, the initial characteristics of the generator, the initial characteristics of the blades and the initial characteristics of the converter of the wind turbine set into the preset variable pitch control model to obtain the maximum voltage deviation ratio; performing back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters; performing variable pitch control and frequency modulation on the wind turbine set based on the pitch angle control parameters and the reactive power control parameters. The present invention obtains multiple initial characteristics of the wind turbine set according to the operating data of the wind power plant under preset conditions, and first calculates the maximum voltage deviation ratio through the preset variable pitch control model, and then performs back propagation iterative training on the maximum voltage deviation ratio to determine the pitch angle control parameters and the reactive power control parameters, so as to realize reactive frequency modulation of the wind turbine set, thereby improving the stability of the microgrid.
[0110] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.
Claims
1. A method for variable pitch control and frequency modulation of a wind turbine generator set, It is characterized in that include: Generate initial characteristics of the generator, initial characteristics of the blades, and initial characteristics of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions; Inputting the vertex diagram of the wind turbine set, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset pitch control model to obtain a maximum voltage deviation ratio; Performing back-propagation iterative training on the voltage maximum deviation ratio to determine a pitch angle control parameter and a reactive power control parameter; Performing variable pitch control and frequency modulation on the wind turbine generator set based on the pitch angle control parameter and the reactive power control parameter; The preset pitch control model includes: a graph network layer, a parameter generation layer and an MLP layer; the vertex graph of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter are input into the preset pitch control model to obtain the maximum voltage deviation ratio, including: The graph network layer generates a generator intermediate vector and a blade intermediate vector according to the vertex graph, the generator initial features, the blade initial features, and the converter initial features, and inputs the generator intermediate vector and the blade intermediate vector into the parameter generation layer; The parameter generation layer generates a unified control parameter vector according to the generator intermediate vector and the blade intermediate vector, and inputs the unified control parameter vector to the MLP layer; The MLP layer obtains the maximum voltage deviation ratio in the time period from when the frequency regulation of the wind turbine generator set is started to when the voltage at the grid connection point returns to a steady state according to the unified control parameter vector; The method further includes constructing a vertex graph of the wind turbine generator set, specifically comprising: Determine the generator vertex, blade vertex, and converter vertex of the wind turbine; According to the connection relationship between the wind turbine generator, the wind turbine blades, and the wind turbine converter, the generator vertex, the blade vertex, and the converter vertex are connected to obtain a vertex graph of the wind turbine; The performing back propagation iterative training on the voltage maximum deviation ratio to determine the pitch angle control parameter and the reactive power control parameter comprises: Back-propagating the voltage maximum deviation ratio to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector; When the preset termination condition is met, the i-1th component of the unified control parameter vector updated in the last iteration is used as the pitch angle control parameter of the i-th wind turbine group, and the i-th component is used as the reactive power control parameter of the i-th wind turbine group.
2. The wind turbine pitch control frequency modulation method according to claim 1, It is characterized in that Before the generator intermediate vector and the blade intermediate vector are input into the parameter generation layer, the method includes: splicing the generator intermediate vector and the blade intermediate vector to obtain an intermediate vector vector.
3. The wind turbine pitch control frequency modulation method according to claim 1, It is characterized in that Before inputting the generator intermediate vector and the blade intermediate vector into the parameter generation layer, the method further includes: The parameter generation layer is connected to a preset classifier to classify and obtain a pre-trained value of a dynamic frequency deviation ratio of a microgrid of a wind power plant; Pre-training is performed with minimizing the difference between the pre-trained value of the dynamic frequency deviation ratio and the true value of the dynamic frequency deviation ratio as a training goal.
4. The wind turbine pitch control frequency modulation method according to claim 1, It is characterized in that The back-propagating the maximum voltage deviation ratio to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector includes: performing back-propagation iterative training to update the unified control parameter vector based on the difference between the maximum voltage deviation ratio and a preset voltage deviation ratio threshold as a loss value.
5. A wind turbine variable pitch control frequency modulation device, It is characterized in that include: A feature generation module, used to generate initial features of the generator, initial features of the blades, and initial features of the converter of the wind turbine according to the operation data of the wind power plant under preset conditions; A model calculation module, used for inputting the vertex diagram of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter into a preset variable pitch control model to obtain a maximum voltage deviation ratio; An iterative training module, used for performing back-propagation iterative training on the voltage maximum deviation ratio to determine a pitch angle control parameter and a reactive power control parameter; A control module, used for performing variable pitch control and frequency modulation on the wind turbine generator set based on the pitch angle control parameter and the reactive power control parameter; The preset pitch control model includes: a graph network layer, a parameter generation layer and an MLP layer; the vertex graph of the wind turbine, the initial characteristics of the generator, the initial characteristics of the blades, and the initial characteristics of the converter are input into the preset pitch control model to obtain the maximum voltage deviation ratio, including: The graph network layer generates a generator intermediate vector and a blade intermediate vector according to the vertex graph, the generator initial features, the blade initial features, and the converter initial features, and inputs the generator intermediate vector and the blade intermediate vector into the parameter generation layer; The parameter generation layer generates a unified control parameter vector according to the generator intermediate vector and the blade intermediate vector, and inputs the unified control parameter vector to the MLP layer; The MLP layer obtains the maximum voltage deviation ratio in the time period from when the frequency regulation of the wind turbine generator set is started to when the voltage at the grid connection point returns to a steady state according to the unified control parameter vector; The model calculation module is also used to construct a vertex graph of the wind turbine generator set, specifically including: Determine the generator vertex, blade vertex, and converter vertex of the wind turbine; According to the connection relationship between the wind turbine generator, the wind turbine blades, and the wind turbine converter, the generator vertex, the blade vertex, and the converter vertex are connected to obtain a vertex graph of the wind turbine; The performing back propagation iterative training on the voltage maximum deviation ratio to determine the pitch angle control parameter and the reactive power control parameter comprises: Back-propagating the voltage maximum deviation ratio to the parameter generation layer for back-propagation iterative training to update the unified control parameter vector; When the preset termination condition is met, the i-1th component of the unified control parameter vector updated in the last iteration is used as the pitch angle control parameter of the i-th wind turbine group, and the i-th component is used as the reactive power control parameter of the i-th wind turbine group.
6. A wind turbine pitch control frequency modulation device, It is characterized in that comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the wind turbine pitch control and frequency modulation method described in any one of claims 1 to 4 above.
7. A computer-readable storage medium, It is characterized in that Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps of the wind turbine pitch control and frequency modulation method described in any one of claims 1 to 4 above.
