Self-adaptive adjusting method and system for control parameters of net-forming type direct-driven fan

Abstract

The invention relates to a self-adaptive adjustment method and system for control parameters of a network-building type direct-drive fan, and the method comprises the steps: obtaining the operation data of the network-building type direct-drive fan, initializing the control parameters, and obtaining adjustment parameters based on the operation data; the adjusting parameters are input into a BP neural network, output vectors are obtained, and each vector value of the output vectors corresponds to one control parameter variable quantity of the network construction type direct drive fan; mapping the adjustment parameter to a preset control parameter disturbance rule table to obtain an update direction of the control parameter; updating the control parameter based on the initialized control parameter, the control parameter variation and the updating direction of the control parameter; and the updated control parameters are input into a control equation of a virtual synchronous machine algorithm in the network construction type direct drive fan, and control over the network construction type direct drive fan is completed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system control, and mainly relates to a method and system for self-adaptive adjustment of control parameters of grid-connected direct-drive wind turbines. BACKGROUND

[0002] With the wide access of high-proportion wind power in power systems, a large number of devices based on power electronic converters are connected to the power grid, which makes the power system show significant "power electronic" characteristics. Compared with traditional synchronous generators, such devices generally lack the inertia response capability provided by rotating mass, and when the system is disturbed by external disturbances, it is difficult to effectively suppress frequency deviation in time, thereby affecting the frequency stability and overall dynamic operation quality of the power grid.

[0003] To improve the above problems, grid-connected direct-drive wind turbines usually adopt a virtual synchronous generator (VSG) control strategy to simulate the electromagnetic and mechanical characteristics of synchronous generators. This type of control directly determines the system power angle through a power synchronization mechanism, eliminating the dependence of traditional phase-locked loops in transient processes, thereby having the ability to actively provide frequency support when the system is disturbed. However, the existing virtual synchronous control still has large frequency deviation and oscillation amplitude when facing severe operating conditions in actual operation, and its frequency support effect and transient stability performance still cannot fully meet the requirements of high-proportion new energy grid connection.

[0004] On the other hand, the enhanced direct current voltage control (eDVC) proposed by existing research couples the direct current voltage control loop with the virtual synchronous generator framework and introduces power angle-related quantities to participate in direct current side regulation, which improves the stability of the direct current link and the power exchange characteristics to some extent. However, in complex scenarios such as strong disturbance and weak grid, the control parameters are usually fixed, which makes it difficult to adapt to changes in system state in time, and frequency fluctuations and direct current voltage oscillations may still coexist.

[0005] Therefore, it is urgent to propose a parameter control method that can coordinate the frequency side and the direct current voltage side, significantly enhance the frequency support capability of grid-connected direct-drive wind turbines, and effectively suppress transient frequency fluctuations and direct current voltage fluctuations, in order to further improve the operation safety and reliability of power systems under high-proportion new energy access conditions. SUMMARY

[0006] In order to solve the problems existing in the prior art, the present application proposes a method and system for self-adaptive adjustment of control parameters of grid-connected direct-drive wind turbines.

[0007] The technical solution of the present application is as follows: On the one hand, the present application proposes a method for self-adaptive adjustment of control parameters of grid-connected direct-drive wind turbines, which comprises: Obtain the operating data of the grid-type direct-drive fan and initialize the control parameters, and obtain the adjustment parameters based on the operating data; The adjustment parameters are input into a BP neural network to obtain an output vector, where each vector value of the output vector corresponds to the change in the control parameter of a grid-type direct-drive fan. The adjustment parameters are mapped to a preset control parameter disturbance law table to obtain the update direction of the control parameters; The control parameters are updated based on the initial control parameters, the changes in the control parameters, and the update direction of the control parameters. The updated control parameters are input into the control equations of the virtual synchronous machine algorithm in the grid-type direct-drive wind turbine to complete the control of the grid-type direct-drive wind turbine.

[0008] Preferably, the grid-connected direct-drive wind turbine includes a wind turbine, a permanent magnet synchronous generator, and a back-to-back full-power converter, and is connected to the power grid through a grid connection point; The back-to-back converter includes a machine-side converter, a DC-side support capacitor, and a grid-side converter. The machine-side converter generates a power reference value through maximum power point tracking control; The grid-side converter generates its output power and voltage amplitude through the control equations of the virtual synchronous machine algorithm.

[0009] Preferably, the adjustment parameters include grid frequency deviation, grid frequency change rate, DC bus voltage deviation, and DC bus voltage change rate; Based on the current power grid operating frequency and the standard power frequency of the power grid The difference between them yields the power grid frequency deviation. Differentiating it yields the rate of change of the power grid frequency. ; Based on DC bus voltage and DC bus reference voltage The difference between them yields the DC bus voltage deviation. Differentiating it yields the rate of change of DC bus voltage. .

[0010] Preferably, the adjustment parameters are input into the BP neural network, and the specific steps are as follows: If the grid frequency deviation at the current sampling time is greater than the preset frequency deviation threshold or the DC bus voltage deviation is greater than the preset voltage deviation threshold, then forward calculation is performed. The BP neural network includes an input layer, a hidden layer, and an output layer; The input layer is used to receive the adjustment parameters; The hidden layer performs a non-linear transformation on the adjustment parameters to obtain the hidden layer output, which is calculated as follows: , ; ; ; In the formula, Indicates the first Hidden layer in the first layer The hidden layer input at the sampling time; Indicates the first Hidden layer in the first layer Hidden layer output at the sampling time; Represents the Sigmoid function; Indicates the first The first adjustment parameter and the first The weights between layers are implicit; Indicates the first Sampling time number One adjustment parameter; Indicates the first Hidden layer bias of hidden layers; This represents the input variables for the Sigmoid function; Indicates the number of hidden layers; The output layer linearly transforms the output of the hidden layer to obtain the output vector, which is calculated as follows: , ; ; ; In the formula, Indicates the first Layer output layer at the 1st Output layer input at sampling time; Indicates the first Hidden layers up to the first The layer outputs the weights between layers; Indicates the first The bias of the output layer; Indicates the first Layer output layer at the 1st The output at the sampling time; Representing a linear function Input variables; when At that time, the change in virtual inertia in the corresponding control parameters ; when At that time, the change in virtual damping in the corresponding control parameters ; when At that time, the change in the DC-side additional damping coefficient in the corresponding control parameters ; when At that time, the change in the acceleration damping coefficient in the corresponding control parameters .

[0011] Preferably, the method also includes updating the weights of the hidden layer and the output layer, with the following specific steps: Obtain the error of the power grid frequency deviation and the error of the DC bus voltage deviation; The performance index function is calculated based on the errors of grid frequency deviation and DC bus voltage deviation. The weights of the hidden and output layers are updated based on the performance metric function.

[0012] Preferably, the control parameters are updated based on the initial control parameters and the changes in the control parameters, and the calculation method is as follows: ; In the formula, Indicates the first Virtual inertia at the sampling time; Indicates the initial virtual inertia; Represents the minimum virtual inertia; Indicates the maximum virtual inertia; Indicates the first Virtual damping at the sampling time; Indicates the initial virtual damping; Indicates minimum virtual damping; Indicates the maximum virtual damping; Indicates the first Additional damping coefficient on the DC side at the sampling time; This represents the initial DC-side additional damping coefficient; Indicates the minimum DC-side additional damping coefficient; Indicates the maximum DC-side additional damping coefficient; Indicates the first The acceleration damping coefficient at the sampling time; Indicates the initial acceleration damping coefficient; Indicates the minimum acceleration damping coefficient; This represents the maximum acceleration damping coefficient.

[0013] Preferably, the updated control parameters are input into the control equations of the virtual synchronous machine algorithm in the grid-type direct-drive wind turbine, and the calculation method is as follows: ; ; ; In the formula, represents the active power reference value generated by the grid-side converter control; represents the output power of the grid-side converter; represents the rated angular frequency; represents the power angle change rate; represents the power angle acceleration; represents the DC voltage proportional control parameter; represents the DC voltage integral control parameter; represents the Laplace operator; represents the square of the DC bus voltage; represents the square of the DC bus voltage reference value; represents the DC side additional damping coefficient; represents the acceleration damping coefficient; represents the grid-side converter output voltage amplitude; represents the grid voltage amplitude; represents the grid angular frequency; represents the grid-side converter grid connection point equivalent inductance.

[0014] In another aspect, the application also provides a system for adaptive adjustment of control parameters of grid-connected direct-drive wind turbines, comprising: a data acquisition module, which acquires operation data of the grid-connected direct-drive wind turbines and initializes control parameters, and obtains adjustment parameters based on the operation data; a control parameter updating module, which inputs the adjustment parameters into a BP neural network to obtain an output vector, wherein each vector value of the output vector corresponds to a control parameter change amount of a grid-connected direct-drive wind turbine; maps the adjustment parameters to a preset control parameter disturbance law table to obtain an updating direction of the control parameters; and updates the control parameters based on the initialized control parameters, the control parameter change amount and the updating direction of the control parameters; an execution control module, which inputs the updated control parameters into a control equation of a virtual synchronous machine algorithm in the grid-connected direct-drive wind turbine to complete control of the grid-connected direct-drive wind turbine.

[0015] In still another aspect, the application also provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the method according to the application when executing the program.

[0016] In still another aspect, the application also provides a computer readable storage medium, which stores a computer program, and the program is executable on a processor to implement the method according to the application.

[0017] The application has the following beneficial effects: 1. The application provides a network type direct drive wind turbine control parameter adaptive adjustment method and system, which adopts the adaptive learning principle of BP neural network, takes the grid frequency deviation, grid frequency change rate, DC bus voltage deviation and DC bus voltage change rate as input, and outputs the change amount of virtual inertia, virtual damping, DC side additional damping coefficient and acceleration damping coefficient; at the same time, the performance index function is constructed based on the grid frequency deviation error and DC bus voltage deviation error, the neural network hidden layer and output layer weight are updated reversely, the continuous self-optimization of the control parameter is realized, the dynamic adaptation ability of the control parameter is improved, the parameter adjustment precision is improved, the collaborative control demand of the frequency side and the DC side is accurately matched, and the robustness of the system to complex working conditions is enhanced. 2. The application provides a network type direct drive wind turbine control parameter adaptive adjustment method and system, wherein the network type direct drive wind turbine adopts a multi-level architecture: the maximum power point tracking control is adopted on the machine side to capture the maximum wind energy, the VSG voltage / current double closed loop control is adopted on the grid side to realize accurate grid connection, and the adaptive parameter algorithm provides real-time optimized control parameters for the VSG of the grid side converter, thereby improving the energy utilization efficiency of wind power generation, improving the precision of the converter control, and enhancing the operation reliability of the unit under all working conditions. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is the specific flowchart of the embodiment of the application; Figure 2 is the system block diagram of the network type direct drive wind turbine of the embodiment of the application; Figure 3 is the maximum power tracking curve of the wind turbine of the embodiment of the application; Figure 4 is the control block diagram of the machine side converter of the network type direct drive wind turbine of the embodiment of the application; Figure 5 is the control block diagram of the grid side converter of the network type direct drive wind turbine of the embodiment of the application; Figure 6 is the grid frequency disturbance curve diagram of the embodiment of the application; Figure 7 is the DC voltage disturbance curve diagram of the embodiment of the application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the application will be described clearly and completely in combination with the drawings of the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0020] It should be understood that the step numbers used herein are only for the convenience of description, and are not intended to limit the execution sequence of the steps.

[0021] It should be understood that the terms used in the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application and the appended claims, "a", "an", and "the" in singular form are intended to include plural forms unless the context clearly indicates otherwise.

[0022] The terms "comprise" and "include" indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0023] The term "and / or" means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.

[0024] Embodiment one: Referring to Figure 1 , the present application provides a self-adaptive adjustment method for control parameters of a grid-forming direct-drive wind turbine, the method comprising: S1, obtaining operation data of the grid-forming direct-drive wind turbine and initializing control parameters, the control parameters including changes in virtual inertia, virtual damping, DC side additional damping coefficient, and acceleration damping coefficient ; S11, referring to Figure 2 , the grid-forming direct-drive wind turbine comprises a wind turbine, a permanent magnet synchronous generator, and a back-to-back full-power converter, which is connected to the power grid through a grid connection point; S111, the back-to-back converter comprises a machine-side converter, a DC side support capacitor, and a grid-side converter; Referring to Figure 3 , the machine-side converter is responsible for energy exchange between the generator electromagnetic energy and the DC side; by maximum power point tracking control, by adjusting the generator speed according to the real-time wind speed and operating conditions, the wind turbine operating point is always maintained on the characteristic curve corresponding to the best aerodynamic efficiency, generating a power reference value , referring to Figure 4 ; The actual output power of the machine-side converter is compared with the power error signal input proportional-integral regulator, and the current reference value is obtained ; After obtaining the current reference, further enter the current inner loop decoupling control stage; according to the mathematical model of the permanent magnet synchronous generator, the shaft and​ The shaft current is decoupled and closed-loop regulated, and a proportional-integral regulator generates a voltage command for the machine-side converter in a synchronous rotating coordinate system. and The calculation method is as follows: ; ; In the formula, The outer current loop (or power / DC voltage outer loop) indicates that... Shaft current reference value; This indicates the value given by the outer current loop (or the outer power / DC voltage loop). Shaft current reference value; Indicates that the machine-side converter is in Reference voltage of the shaft; Indicates that the machine-side converter is in Reference voltage of the shaft; express The proportionality factor of the shaft; express The integral coefficient of the axis; express The proportionality factor of the shaft; express The integral coefficient of the axis; Represents stator current Axial components; This indicates the electrical angular velocity of the motor rotor; Indicates synchronous motor Shaft inductance; Represents stator current Axial components; This refers to the magnetic flux generated by a permanent magnet; Through the , and By feeding forward compensation of the isocoupler and back EMF terms, the equivalent controlled object of the current inner loop is linearized, thereby achieving control over... , Fast, decoupled control; generally, to achieve unity power factor or maximum torque utilization, priority is given to using... This allows the excitation to be provided by permanent magnets, resulting in lower generator losses; Voltage command and After PWM processing, the electromagnetic torque is applied to the generator stator, and the electromagnetic torque follows the reference value in real time. Through the coordinated operation of the power outer loop, torque coupling loop and current inner loop, the generator-side converter can continuously convert the mechanical power captured by the wind turbine into electrical energy with the highest energy efficiency, and maintain the transient balance of DC link power by precisely adjusting the electromagnetic torque of the generator, thereby ensuring the stable operation and maximum power transmission capability of the grid-connected wind turbine under different operating conditions. S112, The DC support capacitor is used to stabilize the DC link voltage; S113, please refer to Figure 5 The grid-side converter is responsible for injecting electrical energy into the grid and providing the necessary voltage and frequency support under weak grid conditions; the output power and voltage amplitude of the grid-side converter are generated through the control equations of the virtual synchronous machine algorithm. The virtual synchronous machine algorithm calculates the voltage amplitude and phase angle reference values ​​in real time based on the DC-side power balance state and grid-side converter operation information, and generates the target output voltage command for the grid-side converter. In order to achieve accurate tracking of the command, the control system simultaneously collects grid-side output voltage and current, and constructs a dual closed-loop decoupled control structure that combines the voltage outer loop and the current inner loop. Among them, the inner loop current control loop is adjusted quickly. shaft and The shaft current enables rapid dynamic control of the injected current of the grid-side converter and provides necessary current limiting functions to ensure the system's rapid protection capability under disturbances such as short circuits and voltage drops. The outer voltage control loop uses the reference voltage generated by the virtual synchronous machine algorithm as the target. By adjusting the current reference value, it ensures that the output voltage amplitude and phase angle smoothly and accurately track the virtual synchronous machine command, thereby maintaining the stability of the converter terminal voltage. Finally, the voltage command signal obtained from the dual closed-loop regulation is converted into a drive signal by the pulse width modulation (PWM) module and applied to the grid-side converter to achieve precise synthesis of the output voltage waveform and flexible adjustment of active and reactive power. Therefore, the grid-type direct-drive wind turbine can enable the grid-type wind turbine to have inertial response and damping characteristics similar to synchronous generators, thereby improving its support capability and operational stability under weak power grid and fluctuating operating conditions. S2. Obtain adjustment parameters based on operational data; The adjustment parameters include grid frequency deviation, grid frequency change rate, DC bus voltage deviation, and DC bus voltage change rate; Based on the current power grid operating frequency and the standard power frequency of the power grid The difference between them yields the power grid frequency deviation. Differentiating it yields the rate of change of the power grid frequency. ; Based on DC bus voltage and DC bus reference voltage The difference between them yields the DC bus voltage deviation. Differentiating it yields the rate of change of DC bus voltage. ; S3. Input the adjustment parameters into the BP neural network to obtain the output vector, wherein each vector value of the output vector corresponds to the change in the control parameter of a grid-type direct-drive fan; S31. If the grid frequency deviation at the current sampling time is greater than the preset frequency deviation threshold or the DC bus voltage deviation is greater than the preset voltage deviation threshold, then forward calculation is performed, specifically as follows: like or ,in This indicates the preset power grid frequency deviation threshold. This indicates the preset DC bus voltage deviation threshold. Indicates the first DC bus voltage deviation at the sampling time Indicates the first If the power grid frequency deviation at the sampling time is true, then forward calculation is performed; otherwise, the change in virtual inertia is output. The change in virtual damping The change in the DC-side additional damping coefficient The change in acceleration damping coefficient ; S32. The BP neural network includes two parts: offline tuning and online tuning. Offline tuning uses a large amount of operating data under different working conditions to train the BP neural network, enabling it to have preliminary parameter adjustment capabilities. On the basis of this, online tuning continuously adjusts the control parameters according to dynamic information such as frequency deviation and DC voltage change measured in real time, so that the system can maintain optimal dynamic performance and stability under various disturbance conditions. The BP neural network includes an input layer, a hidden layer, and an output layer; The input layer is used to receive the adjustment parameters; The hidden layer performs a non-linear transformation on the adjustment parameters to obtain the hidden layer output, which is calculated as follows: , ; ; ; In the formula, Indicates the first Hidden layer in the first layer The hidden layer input at the sampling time; Indicates the first Hidden layer in the first layer Hidden layer output at the sampling time; Represents the Sigmoid function; Indicates the first The first adjustment parameter and the first The weights between layers are implicit; Indicates the first Sampling time number One adjustment parameter; Indicates the first Hidden layer bias of hidden layers; This represents the input variables for the Sigmoid function; Indicates the number of hidden layers; The output layer linearly transforms the output of the hidden layer to obtain the output vector, which is calculated as follows: , ; ; ; In the formula, Indicates the first Layer output layer at the 1st Output layer input at sampling time; Indicates the first Hidden layers up to the first The layer outputs the weights between layers; Indicates the first The bias of the output layer; Indicates the first Layer output layer at the 1st The output at the sampling time; Representing a linear function Input variables; when At that time, the change in virtual inertia in the corresponding control parameters ; when At that time, the change in virtual damping in the corresponding control parameters ; when At that time, the change in the DC-side additional damping coefficient in the corresponding control parameters ; when At that time, the change in the acceleration damping coefficient in the corresponding control parameters ; S4 also includes updating the weights of the hidden layer and the output layer. The specific steps are as follows: S41. Obtain the error of the power grid frequency deviation and the error of the DC bus voltage deviation; The error of the power grid frequency deviation is expressed as: ,in Indicates the first The error of the power grid frequency deviation at the sampling time. This represents the expected value of the power grid frequency deviation. The error of the DC bus voltage deviation is expressed as: ,in Indicates the first The error of the DC bus voltage deviation at the sampling time. This represents the expected value of the DC bus voltage deviation. S42. Based on the errors of grid frequency deviation and DC bus voltage deviation, the performance index function is calculated as follows: ; In the formula, Indicates the first Performance index function at sampling time; S43. Update the weights of the hidden layer and the output layer based on the performance index function; The output layer weights are updated, and the calculation method is as follows: ; ; ; ; achievable ; In the formula, This represents the gradient scaling factor, used to adjust the relative weights of the gradient terms. It can be set to 1 or absorbed by the learning rate. The update formula for obtaining the output layer weights is expressed as follows: ,in, This represents the inertia coefficient, which accelerates convergence and reduces oscillations; This represents the learning rate and controls the step size for weight updates. The hidden layer weights are updated and calculated as follows: ; ; ; ; achievable ; The formula for updating the hidden layer weights is expressed as follows: ; S5. Map the adjustment parameters to the preset control parameter disturbance law table to obtain the update direction of the control parameters; Please see Figure 6 After the system is disturbed, the power grid frequency deviation Typical dynamic trajectory within a response cycle; In the range where the absolute value of the frequency deviation increases and Increase virtual inertia This improves the system's ability to suppress the rate of frequency change, preventing the frequency deviation from increasing drastically; it also increases virtual damping. This enhances the system's oscillation dissipation capability and accelerates the decay of frequency oscillation energy; it also increases the DC-side additional damping coefficient. Enhance the additional damping based on the rate of change of power angle, enabling the DC voltage power control loop to provide more electromagnetic damping to the system; increase the acceleration damping parameter. This enhances the damping adjustment based on power angle acceleration, thereby increasing the controller's "feedforward" shock resistance. Within the recovery range where the absolute value of the frequency deviation decreases. and Reduce virtual inertia Reduce recovery resistance and accelerate the frequency's return to steady-state value; reduce virtual damping. To avoid excessive attenuation leading to system response lag and to ensure rapid frequency convergence; reduce the additional damping coefficient on the DC side. This reduces the reverse braking effect of the DC channel on frequency recovery. It also reduces the acceleration damping parameters. To reduce the amount of feedforward suppression, the system can naturally stabilize without generating unnecessary delay; as shown in Table 1. Table 1. Disturbance Laws of Control Parameters under Power Grid Frequency Disturbance

[0025] Please see Figure 7 The DC voltage after the disturbance is given The typical dynamic trajectory, within a recovery period after a disturbance, can also be divided into four typical stages. , , and ; During the stage of increasing absolute value of DC voltage deviation and Increase the additional damping coefficient on the DC side To increase the additional damping coefficient based on the rate of change of the power angle, the active power is allowed to release oscillating energy through the DC-AC energy channel, suppressing the expansion of DC voltage deviation; the acceleration damping parameter is increased. To improve the feedforward damping based on power angle acceleration, a stronger braking effect is achieved on abrupt components, reducing voltage peaks and valleys; details are shown in Table 2. Table 2. Disturbance Laws of Control Parameters under DC Voltage Disturbance

[0026] S6. Based on the initial control parameters, the changes in control parameters, and the update direction of control parameters, update the control parameters. The calculation method is as follows: ; In the formula, Indicates the first Virtual inertia at the sampling time; Indicates the initial virtual inertia; Represents the minimum virtual inertia; Indicates the maximum virtual inertia; Indicates the first Virtual damping at the sampling time; Indicates the initial virtual damping; Indicates minimum virtual damping; Indicates the maximum virtual damping; Indicates the first The DC-side additional damping coefficient at the sampling time; This represents the initial DC-side additional damping coefficient; Indicates the minimum DC-side additional damping coefficient; Indicates the maximum DC-side additional damping coefficient; Indicates the first The acceleration damping coefficient at the sampling time; Indicates the initial acceleration damping coefficient; Indicates the minimum acceleration damping coefficient; Indicates the maximum acceleration damping coefficient; S7. Input the updated control parameters into the control equations of the virtual synchronous machine algorithm in the grid-type direct-drive wind turbine to complete the control of the grid-type direct-drive wind turbine. The calculation method is as follows: ; ; ; In the formula, This represents the reference value of active power generated by the grid-side converter control. This indicates the output power of the grid-side converter; Indicates the rated angular frequency; Indicates the rate of change of the work angle; Indicates angular acceleration; Indicates the DC voltage proportional control parameters; Indicates the DC voltage integral control parameters; Represents the Laplace operator; Represents the square of the DC bus voltage; This represents the square of the DC bus voltage reference value; Indicates the additional damping coefficient on the DC side; Indicates the acceleration damping coefficient; Indicates the output voltage amplitude of the grid-side converter; Indicates the magnitude of the grid voltage; Indicates the angular frequency of the power grid; This represents the equivalent inductance at the grid connection point of the grid-side converter.

[0027] Example 2: This embodiment provides an adaptive adjustment system for the control parameters of a grid-type direct-drive wind turbine, the system comprising: The data acquisition module acquires the operating data of the grid-type direct-drive wind turbine and initializes the control parameters, and obtains the adjustment parameters based on the operating data; The control parameter update module inputs the adjustment parameters into a BP neural network to obtain an output vector, where each vector value of the output vector corresponds to a change in the control parameter of a grid-type direct-drive wind turbine; the adjustment parameters are mapped to a preset control parameter disturbance law table to obtain the update direction of the control parameters; and the control parameters are updated based on the initialized control parameters, the change in control parameters, and the update direction of the control parameters. The execution control module inputs the updated control parameters into the control equations of the virtual synchronous machine algorithm in the grid-type direct-drive wind turbine, thereby completing the control of the grid-type direct-drive wind turbine.

[0028] Example 3: This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements an adaptive adjustment method for control parameters of a grid-type direct-drive fan as described in any one of Embodiment 1.

[0029] Example 4: This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements an adaptive adjustment method for control parameters of a grid-type direct-drive fan as described in any one of Embodiment 1.

[0030] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0031] Those skilled in the art will recognize that the units and algorithm steps described in the embodiments disclosed herein can be implemented using electronic hardware, computer software, or a combination of electronic hardware and software. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0032] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0033] In the several embodiments provided in this application, any function, if implemented as a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0034] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for adaptive adjustment of control parameters of a grid-connected direct drive wind turbine, characterized in that, The method comprises: obtaining operation data of the grid-connected direct-drive wind turbine and initializing control parameters, and obtaining adjustment parameters based on the operation data; inputting the adjustment parameters into a BP neural network to obtain an output vector, wherein each vector value of the output vector corresponds to a control parameter change amount of the grid-connected direct-drive wind turbine; mapping the adjustment parameters to a preset control parameter perturbation law table to obtain an update direction of the control parameters; updating the control parameters based on the initialized control parameters, the control parameter change amount and the update direction of the control parameters; inputting the updated control parameters into a control equation of a virtual synchronous machine algorithm of the grid-connected direct-drive wind turbine to complete control of the grid-connected direct-drive wind turbine.

2. The method according to claim 1, wherein, The grid-connected direct-drive wind turbine comprises a wind turbine, a permanent magnet synchronous generator and a back-to-back full-power converter, and is connected to a power grid through a grid connection point; The back-to-back converter comprises a machine-side converter, a DC side support capacitor and a grid-side converter; The machine-side converter generates a power reference value through maximum power point tracking control; The grid-side converter generates a grid-side converter output power and a voltage amplitude through a control equation of a virtual synchronous machine algorithm.

3. The method according to claim 2, wherein, The adjustment parameters comprise a power grid frequency deviation, a power grid frequency change rate, a DC bus voltage deviation and a DC bus voltage change rate; based on a difference between a current operating frequency of the power grid and a standard operating frequency of the power grid to obtain a power grid frequency deviation deriving a power grid frequency change rate ​ based on a difference between the dc bus voltage and a dc bus reference voltage and the dc bus reference voltage to obtain a dc bus voltage deviation deriving the dc bus voltage deviation .

4. The method according to claim 3, wherein, The specific steps of inputting the adjustment parameters into the BP neural network are as follows: forward calculation is performed if the power grid frequency deviation at the current sampling time is greater than a preset frequency deviation threshold or the DC bus voltage deviation is greater than a preset voltage deviation threshold; The BP neural network comprises an input layer, a hidden layer and an output layer; The input layer is used to receive the adjustment parameters; The hidden layer performs nonlinear transformation on the adjustment parameters to obtain hidden layer output, and the calculation method is as follows: ， ； ； ； wherein, represents the layer hidden layer at the sampling time; represents the layer hidden layer at the sampling time; represents a Sigmoid function; represents the adjustment parameter and the weight between the hidden layers; represents the adjustment parameter at the sampling time; represents the hidden layer bias of the hidden layer; represents an input variable of the Sigmoid function; represents the number of hidden layers; The output layer performs linear transformation on the hidden layer output to obtain an output vector, and the calculation method is as follows: ， ； ； ； wherein represents the layer output layer at the sampling time instant; represents the layer hidden layer to the layer output layer; represents the bias of the layer output layer; represents the layer output layer at the sampling time instant; represents the input variable of the linear function . When the amount of change in the virtual inertia in the corresponding control parameter ; When the amount of change in the virtual damping in the corresponding control parameter ; When the variation of the additional damping coefficient on the DC side in the corresponding control parameter ; When the amount of change in the acceleration damping coefficient in the corresponding control parameter .

5. The method according to claim 4, wherein the method is characterized by, The weights of the hidden layer and the output layer are also updated, and the specific steps are as follows: obtaining errors of the power grid frequency deviation and the DC bus voltage deviation; calculating a performance index function based on the errors of the power grid frequency deviation and the DC bus voltage deviation; updating the weights of the hidden layer and the output layer based on the performance index function.

6. The method according to claim 5, wherein, The control parameters are updated based on the initial control parameters and the control parameter change amount, and the calculation method is as follows: ； In the formula, Indicates the first Virtual inertia at the sampling time; Indicates the initial virtual inertia; Represents the minimum virtual inertia; Indicates the maximum virtual inertia; Indicates the first Virtual damping at the sampling time; Indicates the initial virtual damping; Indicates minimum virtual damping; Indicates the maximum virtual damping; Indicates the first The DC-side additional damping coefficient at the sampling time; This represents the initial DC-side additional damping coefficient; Indicates the minimum DC-side additional damping coefficient; Indicates the maximum DC-side additional damping coefficient; Indicates the first The acceleration damping coefficient at the sampling time; Indicates the initial acceleration damping coefficient; Indicates the minimum acceleration damping coefficient; This represents the maximum acceleration damping coefficient.

7. The method according to claim 6, wherein the method is characterized by, The updated control parameters are input into the control equation of the virtual synchronous machine algorithm of the grid-connected direct-drive wind turbine, and the calculation method is as follows: ； ； ； In the formula, represents the active power reference value generated by the grid-side converter control; represents the output power of the grid-side converter; represents the rated angular frequency; represents the power angle change rate; represents the power angle acceleration; represents the DC voltage proportional control parameter; represents the DC voltage integral control parameter; represents the Laplace operator; represents the square of the DC bus voltage; represents the square of the DC bus voltage reference value; represents the DC side additional damping coefficient; represents the acceleration damping coefficient; represents the grid-side converter output voltage amplitude; represents the grid voltage amplitude; represents the grid angular frequency; represents the grid-side converter grid connection point equivalent inductance.

8. A system for adaptive adjustment of control parameters of a grid-connected direct drive wind turbine, characterized in that The system comprises: a data acquisition module that obtains operation data of the grid-connected direct-drive wind turbine and initializes control parameters, and obtains adjustment parameters based on the operation data; a control parameter updating module that inputs the adjustment parameters into a BP neural network to obtain an output vector, wherein each vector value of the output vector corresponds to a control parameter change amount of the grid-connected direct-drive wind turbine; maps the adjustment parameters to a preset control parameter perturbation law table to obtain an update direction of the control parameters; and updates the control parameters based on the initialized control parameters, the control parameter change amount and the update direction of the control parameters; an execution control module that inputs the updated control parameters into a control equation of a virtual synchronous machine algorithm of the grid-connected direct-drive wind turbine to complete control of the grid-connected direct-drive wind turbine.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the method of any one of claims 1 to 7 when executing the program.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by a processor to implement the method of any one of claims 1 to 7.

Images