Power grid frequency regulation and control method and system based on disturbance dynamic characteristics and medium

Abstract

The invention provides a power grid frequency regulation and control method and system based on disturbance dynamic characteristics and a medium, and relates to the technical field of power system operation and maintenance, and the power grid frequency regulation and control method based on disturbance dynamic characteristics comprises the steps: obtaining transient data in a preset time period before and after a power grid generates a fault; inputting the transient data into a trained neural network model, and generating key disturbance dynamic characteristics of the fault; constructing a corresponding objective function according to the key disturbance dynamic characteristics; based on a model predictive control MPC algorithm, planning a regulation and control task corresponding to the fault into a plurality of different control stages, and generating a corresponding control instruction according to the target function and the plurality of different control stages; and according to the control instruction, the wind turbine generator is controlled to carry out power grid frequency regulation and control, optimal and targeted intervention is carried out at the initial stage of disturbance generation, the regulation and control effect is enhanced, and safe operation of a power system is ensured.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the technical field of power system operation and maintenance. More specifically, the present application relates to a power grid frequency regulation method and system based on disturbance dynamic characteristics, a medium and equipment. BACKGROUND

[0002] With the transformation of global energy structure to low-carbon and clean, the penetration rate of renewable energy represented by wind power in the power system is rapidly increasing through power converters. However, there are essential differences between the physical characteristics of such new energy generation units and traditional synchronous generators. The decoupling characteristics of the new energy generation units with the power grid result in a significant decrease in the equivalent rotational inertia of the power system. The decrease in the equivalent rotational inertia makes the frequency stability of the power grid deteriorate dramatically when it suffers from a large power disturbance (such as large wind turbine generator unit fault de-connection, important transmission line trip, etc.), which specifically manifests as an increase in the frequency change rate and a deeper frequency minimum point, posing an unprecedented severe challenge to the safe and stable operation of the power grid.

[0003] To address this challenge, the academic and industrial communities have proposed virtual inertia control technology. The core idea is to control the power converter of the wind turbine generator to quickly adjust its active power output when the frequency of the power grid changes, and to use the rotational kinetic energy contained in the large blades and transmission chain of the wind turbine generator to simulate the inertia response characteristics of the traditional synchronous generator, thereby providing rapid frequency support for the power grid. Although this virtual inertia control technology alleviates the low inertia problem to some extent, it still has the following shortcomings in practical application: (1) The traditional virtual inertia control is essentially a passive response mechanism based on error feedback (such as proportional-differential control). Its control instruction completely depends on the real-time measurement of the frequency deviation or its change rate of the power grid that has already occurred. This "after-the-fact compensation" mode lacks consideration of the future evolution trend of the disturbance event, resulting in a blind and lagging response, which cannot optimally and specifically intervene in the initial stage of the disturbance, leading to poor regulation effect; (2) The over-power generation of the wind turbine generator for providing inertia support is at the expense of consuming its own rotor kinetic energy. After the frequency event ends, the wind turbine generator must absorb power from the power grid to restore the rotational speed to the optimal working point to repay the "energy debt". The existing control method lacks fine management of the energy recovery process. If its power absorption behavior coincides with the end of the primary frequency modulation of the power grid and the stage of fragile frequency recovery, it will form a secondary power shortage, causing a more severe secondary frequency drop than the first drop, which may cause protection devices to malfunction, and even trigger a cascading failure; (3) Most of them use fixed control parameters or single control logic, which cannot adaptively adjust the support strategy according to the actual severity of the disturbance event. This leads to an over-reaction when facing a slight disturbance, unnecessarily affecting the service life and power generation economy of the wind turbine; while facing a serious fault, it may not be able to meet the most urgent safety needs of the power grid.

[0004] Therefore, there is an urgent need for an advanced control method that can break through the passive response limitation, prospectively recognize the disturbance, and intelligently plan the optimal frequency regulation strategy throughout the process, so as to reduce the risk of secondary drop while efficiently supporting the power grid. SUMMARY

[0005] In order to at least solve one or more technical problems as mentioned above, the present application proposes, in a plurality of aspects, a power grid frequency regulation method based on disturbance dynamic characteristics, a system, a medium and equipment.

[0006] In a first aspect, the power grid frequency regulation method based on disturbance dynamic characteristics provided by the present application comprises the following steps: Obtaining transient data in a preset time period before and after the power grid generates this fault, wherein the transient data at least includes power grid frequency, power grid voltage amplitude, wind turbine active power, wind turbine rotor speed and power grid tie-line exchange power; Inputting the transient data into a trained neural network model to generate key disturbance dynamic characteristics of this fault; According to the key disturbance dynamic characteristics, a corresponding target function is constructed; Based on the model predictive control (MPC) algorithm, the regulation task corresponding to this fault is planned into a plurality of different control stages, and according to the target function and a plurality of different control stages, a corresponding control instruction is generated, wherein the control instruction includes an optimal active power sequence; According to the control instruction, the wind turbine is controlled to perform power grid frequency regulation.

[0007] In some examples, the key disturbance dynamic characteristics include: the lowest point of the power grid frequency, the time to reach the lowest point of the power grid frequency, and the maximum power grid frequency change rate.

[0008] In some examples, a plurality of different control stages include a power grid frequency containment stage, a power grid frequency recovery stage, and a power grid frequency secondary drop suppression stage.

[0009] In some examples, based on the model predictive control (MPC) algorithm, the regulation task corresponding to this fault is planned into a plurality of control stages on the time axis, and according to the target function and a plurality of control stages, a corresponding control instruction is generated, which includes: The MPC algorithm is used to identify a control phase in which the regulation task currently locates and dynamically adjust a first weight corresponding to a grid frequency deviation penalty term and a second weight corresponding to a wind turbine rotor speed deviation penalty term in a loss function of the target function pair according to the control phase in which the regulation task currently locates; According to the first weight and the second weight, a corresponding control instruction is generated.

[0010] In some examples, the loss function J corresponding to the target function is: wherein N P is a prediction step, N c is a control step, W1 is the first weight, W2 is the second weight, W3 is a third weight corresponding to an optimal power penalty term output by the MPC algorithm, f ref is a rated grid frequency, is a target grid frequency at k+i, is a predicted grid frequency at k+i at k, is a predicted power output by the wind turbine at k+i at k, is a predicted rotor speed of the wind turbine at k+i at k, is an optimal rotor speed of the wind turbine at k, is a predicted power output by the wind turbine at k+i at k, and || ||2 is a two-norm symbol.

[0011] In some examples, dynamically adjusting the first weight corresponding to the grid frequency deviation penalty term and the second weight corresponding to the wind turbine rotor speed deviation penalty term in the loss function of the target function according to the control phase in which the regulation task currently locates comprises: In a case where the control phase in which the regulation task currently locates is a grid frequency containment phase, W1>>W2 is set; In a case where the control phase in which the regulation task currently locates is a grid frequency recovery phase, W1>W2 is set; In a case where the control phase in which the regulation task currently locates is a grid frequency secondary drop suppression phase, W1<<W2 is set.

[0012] In some examples, the constraint condition of the loss function of the target function is: The rotor speed of the wind turbine is not lower than a lower limit of safe operation and not higher than an upper limit of safe operation; The total active power output by the wind turbine converter does not exceed its rated capacity; The change rate of the optimal active power does not exceed a preset power climb rate limit.

[0013] In a second aspect, the grid frequency regulation system based on disturbance dynamic characteristics provided by the present application comprises: An acquisition module configured to acquire transient data of a preset time period before and after the power grid generates the current fault, wherein the transient data at least includes grid frequency, grid voltage amplitude, active power of the wind turbine generator and power exchange of the grid tie line; A generation module configured to input the transient data into a trained neural network model to generate key disturbance dynamic characteristics of the current fault; A construction module configured to construct a corresponding target function according to the key disturbance dynamic characteristics; A planning module configured to plan a regulation task corresponding to the current fault into a plurality of different control stages based on a model predictive control (MPC) algorithm and generate corresponding control instructions according to the target function and the plurality of different control stages, wherein the control instructions include an optimal active power sequence; A regulation module configured to control the wind turbine generator to perform grid frequency regulation according to the control instructions.

[0014] In a third aspect, the computer readable storage medium provided by the present application contains program instructions, when the program instructions are executed by a processor, so that the method disclosed in the first aspect is realized.

[0015] In a fourth aspect, the electronic device provided by the present application comprises: A processor; and A memory storing computer instructions, when the computer instructions are executed by the processor, so that the electronic device executes the method disclosed in the first aspect.

[0016] The grid frequency regulation method, system, medium and device based on disturbance dynamic characteristics provided by the present application, based on the model predictive control (MPC) algorithm, plan the regulation task corresponding to the current fault into a plurality of different control stages and generate corresponding control instructions according to the target function and the plurality of different control stages, realize the best and targeted intervention in the initial stage of disturbance, enhance the regulation effect and ensure the safe operation of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0017] The above and other objects, features and advantages of the exemplary embodiments of the present application will be readily understood through reading the detailed description of the exemplary embodiments of the present application below, with reference to the accompanying drawings. In the drawings, several embodiments of the present application are shown by way of example and not limitation, and the same or corresponding reference numbers indicate the same or corresponding parts, in which: Figure 1 An exemplary flowchart of the grid frequency regulation method based on disturbance dynamic characteristics provided by the embodiments of the present application is shown. Figure 2 Fig. 1 shows a schematic structural diagram of an exemplary target clustering-based unmanned aerial vehicle intelligent safety monitoring system according to an embodiment of the present application; Figure 3 Fig. 2 shows an exemplary structural block diagram of an electronic device according to some embodiments of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the scope of the present application.

[0019] It should be understood that the terms "comprise" and "include" used in the specification and claims of the present application indicate the presence of the described features, integers, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0020] It should also be understood that the terms used in the specification of 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 specification and claims of the present application, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should be further understood that the term "and / or" used in the specification and claims of the present application means any combination of one or more of the associated listed items and all possible combinations thereof, and includes these combinations.

[0021] As used in the specification and claims of the present application, the term "if" can be interpreted as "when" or "upon" or "in response to a determination" or "in response to detecting" depending on the context. Similarly, the phrase "if it is determined" or "if [a described condition or event] is detected" can be interpreted as meaning "upon determining" or "in response to determining" or "upon detecting [a described condition or event]" or "in response to detecting [a described condition or event]" depending on the context.

[0022] The specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0023] As Figure 1 shown, the power grid frequency regulation method based on disturbance dynamic characteristics provided by the present application includes the following steps: S101, obtain transient data in a preset time period before and after the power grid generates this fault, wherein the transient data at least includes grid point power grid frequency, grid point voltage amplitude, active power of the wind turbine generator, and grid point tie line exchange power.

[0024] This step is in the "golden window" of the first few hundred milliseconds of the power grid disturbance, and the quantitative evaluation of the future evolution trend of the event is completed. When the absolute value of the grid frequency change rate is monitored to exceed the preset sensitivity threshold, this step is immediately activated. It will obtain a high-resolution grid transient data window (transient data) containing the trigger point before and after from the data stream of the synchronous phasor measurement unit, which is a multi-dimensional electrical quantity time sequence matrix.

[0025] S102, input the transient data into the trained neural network model to generate the key disturbance dynamic characteristics of this fault.

[0026] Specifically, the neural network model is a Transformer model. The self-attention mechanism of the Transformer model can effectively capture the key transient patterns from the input sequence. This module can be formally represented as a nonlinear mapping function in function, which accurately maps the initial transient sequence to the key disturbance dynamic characteristics representing the final severity of the event.

[0027] S103, construct a corresponding target function according to the key disturbance dynamic characteristics.

[0028] In some examples, the key disturbance dynamic characteristics include the grid frequency minimum point, the time to reach the grid frequency minimum point, and the maximum grid frequency change rate.

[0029] Wherein, the grid frequency minimum point is to predict how many hertz the grid frequency will drop to the lowest point caused by this disturbance. The time to reach the grid frequency minimum point is to predict how long it will take for the frequency to reach this minimum point, and the maximum grid frequency change rate is used to quantify how fast the initial impact speed of this disturbance is.

[0030] Specifically, this step is a key logical bridge connecting discrete prediction characteristics and time domain optimization control, which converts "static" information into "dynamic" time domain instructions that can be understood and tracked by the MPC algorithm, thereby completely solving the problem of information not being closed loop. This step receives the key disturbance dynamic characteristics output by the previous step and uses them as a set of boundary conditions to substitute into a preset parameterized function model that can simulate frequency dynamics flexibly. In this embodiment, a double exponential function model that can well represent the response of a second-order system is used to generate the target function, which has the form: (1) In equation (1), the grid nominal frequency, four unknown parameters , , are solved by solving a nonlinear equation system determined by the estimated characteristics. The constraints of the equation system include: the initial value constraint determined by the grid nominal frequency , the initial slope constraint determined by the maximum frequency rate of change, and the minimum point amplitude and slope constraint jointly defined by the expected grid frequency minimum point and its arrival time . By numerically solving the equation system, a smooth target frequency that accurately reflects the characteristics of this disturbance is obtained. The target frequency is then discretized to form a vector that covers the time domain of the MPC algorithm as the core reference of the MPC algorithm.

[0031] S104, based on the model predictive control MPC algorithm, the control task corresponding to the disturbance is planned into multiple different control stages, and the corresponding control instruction is generated according to the target function and multiple different control stages, wherein the control instruction includes an optimal active power sequence.

[0032] It can be understood that the multiple different control stages include a grid frequency suppression stage, a grid frequency recovery stage, and a grid frequency secondary drop suppression stage, which realizes the best and targeted intervention in the initial stage of the disturbance, enhances the control effect, and can guarantee the safe operation of the power system.

[0033] In some examples, step S104 specifically includes the following steps: using the MPC algorithm, the control stage where the control task is currently located is identified in real time, and the first weight corresponding to the grid frequency deviation penalty term and the second weight corresponding to the wind turbine rotor speed deviation penalty term in the loss function of the target function pair are dynamically adjusted according to the control stage where the control task is currently located; generate the corresponding control instruction according to the first weight and the second weight.

[0034] wherein the loss function J of the target function pair is: (2) In formula (2), N P is the prediction step, N c is the control step, W1 is the first weight, W2 is the second weight, W3 is the third weight corresponding to the optimal power penalty term output by the Mpc algorithm, f ref is the grid nominal frequency, is the target grid frequency at time k+i, is the grid frequency at time k+i predicted at time k, Pk+i-1 is the predicted output power of the wind turbine at time k+i-1, ωk+i is the predicted rotor speed of the wind turbine at time k+i, ωk is the optimal rotor speed of the wind turbine at time k, Pk+i is the predicted output power of the wind turbine at time k+i, || ||2 is the two-norm symbol. Specifically, the constraint condition of the loss function of the objective function is: the rotor speed of the wind turbine is not lower than the lower limit of safe operation and not higher than the upper limit of safe operation, the total active power output by the converter of the wind turbine does not exceed the rated capacity thereof, and the change rate of the optimal active power does not exceed the preset power climbing rate limit value.

[0035] In some examples, according to the control phase in which the regulation task currently locates, dynamically adjusting the first weight corresponding to the grid frequency deviation penalty term and the second weight corresponding to the wind turbine rotor speed deviation penalty term in the loss function of the objective function pair includes: In the case where the control phase in which the regulation task currently locates is the grid frequency containment phase, W1>>W2 is set; In the case where the control phase in which the regulation task currently locates is the grid frequency recovery phase, W1>W2 is set; In the case where the control phase in which the regulation task currently locates is the grid frequency secondary drop suppression phase, W1<<W2 is set.

[0036] wherein W1 and W2 are dynamically changed. W1 is used to penalize the deviation of the predicted frequency from the target frequency, and W2 is used to penalize the deviation of the predicted rotor speed from the optimal rotor speed. The switching of the phases is managed by a hybrid trigger unit. Phase one (inertial response and frequency containment) is triggered by the time when the disturbance is monitored, at which time the weight is set as W1>>W2 to prioritize the stability of the grid frequency, forcing the MPC algorithm to prioritize how to make the predicted frequency closely follow the target frequency when performing rolling optimization, even if the rotor speed of the wind turbine will significantly decrease. Phase two (frequency bottom-up and recovery support) is started by a prospective time trigger based on the estimated time T nadir to the lowest point of the grid frequency, at which time the weight is adjusted as W1>W2 to take into account both sides, that is, to effectively support the grid frequency while also starting to focus on the safety of the wind turbine itself to avoid excessively low rotor speed. Phase three (active recovery of rotor speed and secondary drop suppression) is started by a robust trigger based on the real-time state, which confirms that the current time has exceeded T nadirAnd the continuously measured frequency change rate remains positive. At this time, the weights are completely reversed to W1 << W2 to preferentially ensure the smooth recovery of the wind turbine speed, enabling the wind turbine speed to safely and smoothly recover to the optimal state, while strictly preventing secondary impacts on the power grid during this process. Adaptively adjusting the support strategy according to the actual severity of the disturbance event can meet the most urgent safety requirements of the power grid.

[0037] In summary, in each control cycle, the MPC algorithm solves a complete quadratic programming problem with constraints, that is, it minimizes the above loss function while satisfying the system model constraints, rotor speed safety constraints, power capacity constraints, and power change rate constraints.

[0038] Specifically, step S104 is the core step of intelligent decision-making and execution, and the MPC algorithm is used for continuous rolling optimization. Its "phase-adaptive" characteristic is reflected in that the internal loss function will dynamically adjust its optimization focus in different stages according to a clear hybrid triggering mechanism. The optimization of the MPC algorithm is based on a discretized system state space model, which describes the dynamic response of the system.

[0039] Define the state vector x(k) of the system as: x(k)=[f(k)−f ref ,ω(k)−ω opt (k)] T (3) In equation (3), f(k) is the grid frequency at time k, f ref is the grid rated frequency, ω(k) is the rotor speed of the wind turbine at time k, and ω opt (k) is the optimal speed of the wind turbine rotor at time k.

[0040] The control input u(k) is: u(k)=ΔP mpc (k)=P(k+i|k)-P(k+i-1|k) (4) In equation (4), ΔP mpc (k) is the power increment output by the MPC algorithm.

[0041] The measurable disturbance d(k) is d(k)=P(k)−P opt (k) (5) In equation (5), P(k) is the actual power output by the wind turbine at time k, and P opt (k) is the optimal power of the wind turbine at time k.

[0042] This model can be expressed as: x(k+1)=Ax(k)+Bu(k)+B dd(k) (6) In equation (6), A is the state space matrix, B and B d The results are obtained by linearizing and discretizing the wind turbine rotor motion equation and the power grid frequency response characteristic equation, respectively. In each control cycle, the MPC algorithm aims to solve an optimization problem that minimizes the system cost over a finite future time domain.

[0043] S105, according to the control command, control the wind turbine to perform grid frequency regulation.

[0044] In some examples, the first element of the optimal control sequence corresponding to the generated control command is sent to the converter of the wind turbine for execution, and then the next control cycle is entered, repeating this rolling optimization process.

[0045] It is understood that the preferred implementation platform for the method provided in this application is a station-level or regional-level centralized controller deployed in a wind farm or regional power grid dispatch center. This controller is a high-performance computing platform, and its hardware architecture includes at least: Data Interface Unit: Responsible for receiving high-speed, synchronous phasor measurement unit (PMU) data streams from external sources, with a sampling rate of no less than 50Hz to ensure the capture of transient data from the power grid. Simultaneously, this unit collects routine operating data from the on-site monitoring and data acquisition (SCADA) system via industrial Ethernet, including real-time wind speed, rotational speed, pitch angle, and operating status of each wind turbine.

[0046] High-performance computing units: These can be servers based on multi-core central processing units (CPUs), graphics processing units (GPUs), or dedicated artificial intelligence (AI) chips. GPUs are primarily used to accelerate online inference computations of Transformer models, while CPUs are responsible for real-time solutions to large-scale quadratic programming problems in model predictive control algorithms.

[0047] Command issuing unit: responsible for sending the optimal active power command calculated by the high-performance computing unit to the central power controller in the wind farm through industrial communication protocols such as IEC61850, with a control cycle of no more than 50 milliseconds, safely and reliably. The central power controller then controls the converters of each wind turbine to perform grid frequency regulation.

[0048] like Figure 2 As shown, the power grid frequency control system based on disturbance dynamic characteristics provided in this application includes: The acquisition module is configured to acquire transient data within a preset time period before and after the power grid fault occurs. The transient data includes at least the power grid frequency, power grid voltage amplitude, active power of wind turbine generators, and power exchange power of power grid interconnection lines. The generation module is configured to input the transient data into a trained neural network model to generate key disturbance dynamic features of the current fault. The construction module is configured to construct a corresponding objective function based on the key disturbance dynamic characteristics; The planning module is configured to use the Model Predictive Control (MPC) algorithm to plan the control task corresponding to the current fault into multiple different control stages and generate corresponding control instructions based on the objective function and the multiple different control stages. The control instructions include the optimal active power sequence. The control module is configured to control the wind turbine to perform grid frequency regulation according to the control command.

[0049] In another aspect, embodiments of this application also provide an electronic device, see [link to relevant documentation]. Figure 3 , Figure 3 This is an exemplary structural block diagram of an electronic device according to an embodiment of this application, such as... Figure 3 As shown, the electronic device includes a processor and a memory, the memory storing computer instructions, and the processor executing the computer instructions to perform the method provided in this application.

[0050] Specifically, processor 601 may include a central processing unit (CPU) or a graphics processing unit (GPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application. Memory 602 may include memory for data or instructions. For example, memory 602 may be at least one of the following: a hard disk drive (HDD), read-only memory (ROM), random access memory (RAM), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, universal serial bus (USB) drive, or other physical / tangible memory storage device. Alternatively, memory 602 may include removable or non-removable (or fixed) media. Furthermore, memory 602 may be internal or external to the integrated gateway disaster recovery device. Memory 602 may be non-volatile solid-state memory. In other words, typically memory 602 includes a tangible (non-transitory) computer-readable storage medium (such as a memory device) encoded with executable instructions, wherein the stored executable instructions, when executed by processor 601 (e.g., by one or more processors), can implement the methods in the embodiments of this application.

[0051] In one example Figure 3The illustrated electronic device may also include a communication interface 603 and a bus 610. The processor 601, memory 602, and communication interface 603 are connected via bus 610 and communicate with each other. Communication interface 603 is primarily used to enable communication between modules, devices, units, and / or equipment within the electronic device. Bus 610, including hardware, software, or both, couples components of the online data flow metering device together. For example, the bus may include at least one of the following: Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) interconnect, Industry Standard Architecture (ISA) bus, Infinite Bandwidth Interconnect, Low Pin Count (LPC) bus, memory bus, Microchannel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus, or other suitable buses. Bus 610 may include one or more buses. Although specific buses are described or illustrated in the embodiments of this application, any suitable bus or interconnection method may be considered in the embodiments of this application.

[0052] In another aspect, embodiments of this application also provide a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the aforementioned method. The computer-readable storage medium may be, for example, a classic computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), disk storage media, optical storage media, flash memory, or other electrical, optical, or other physical / tangible memory storage devices.

[0053] In another aspect, embodiments of this application also provide a computer program product, which includes computer program instructions that, when executed by a processor, implement the method provided in embodiments of this application. This computer program product may be, for example, a software installation package, a plug-in compatible with a related software system, etc.

[0054] The flowcharts and / or block diagrams of the methods and systems of embodiments of this application have been described above by way of example, and related aspects have been described. It should be understood that each block or combination thereof in the flowcharts and / or block diagrams can be implemented by computer program instructions, by dedicated hardware performing a specified function or action, or by a combination of dedicated hardware and computer instructions. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc.; when implemented in software, it is a program or code segment used to perform the required task. The program or code segment can be stored in memory or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0055] While this application has shown and described numerous embodiments, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will arise for those skilled in the art without departing from the spirit and intent of this application. It should be understood that various alternatives to the embodiments of this application described herein may be employed in the practice of this application. The appended claims are intended to define the scope of protection of this application and therefore cover equivalents or alternatives within the scope of these claims.

Claims

1. A power grid frequency regulation method based on disturbance dynamic characteristics, comprising: Acquire transient data within a preset time period before and after the power grid fault occurs. The transient data includes at least the power grid frequency, power grid voltage amplitude, active power of wind turbine generators, rotor speed of wind turbine generators, and switching power of power grid interconnection lines. The transient data is input into a trained neural network model to generate key dynamic features of the current fault. Based on the key dynamic characteristics of the disturbance, a corresponding objective function is constructed; Based on the Model Predictive Control (MPC) algorithm, the control task corresponding to this fault is planned into multiple different control stages, and corresponding control commands are generated according to the objective function and the multiple different control stages. The control commands include the optimal active power sequence. According to the control command, the wind turbine is controlled to perform grid frequency regulation.

2. The power grid frequency regulation method according to claim 1, characterized in that, The key dynamic characteristics of the disturbance include: the lowest point of the grid frequency, the time to reach the lowest point of the grid frequency, and the maximum rate of change of the grid frequency.

3. The power grid frequency regulation method according to claim 1, characterized in that, The various control phases include a grid frequency suppression phase, a grid frequency recovery phase, and a grid frequency secondary drop suppression phase.

4. The power grid frequency regulation method according to claim 3, characterized in that, Based on the Model Predictive Control (MPC) algorithm, the control task corresponding to this fault is planned into multiple control stages on the time axis, and corresponding control commands are generated according to the objective function and the multiple control stages, including: Using the MPC algorithm, the current control stage of the regulation task is identified in real time, and the first weight corresponding to the grid frequency deviation penalty term and the second weight corresponding to the wind turbine rotor speed deviation penalty term in the loss function of the objective function pair are dynamically adjusted according to the current control stage of the regulation task. Based on the first weight and the second weight, corresponding control commands are generated.

5. The power grid frequency regulation method according to claim 4, characterized in that, The loss function J corresponding to the objective function is: Where, N P To predict the step size, N c To control the step size, W1 is the first weight, W2 is the second weight, and W3 is the third weight corresponding to the optimal power penalty term output by the MPC algorithm. ref The rated frequency of the power grid. Let k+i be the target grid frequency. Let k be the predicted power grid frequency at time k+i. This represents the predicted output power of the wind turbine at time k+i-1, obtained at time k. Let k be the rotor speed of the wind turbine at time k+i, predicted at time k. Let k be the optimal rotor speed of the wind turbine at time k. Let |||2| represent the power output of the wind turbine at time k+i predicted at time k, where |||2| represents the L2 norm.

6. The power grid frequency regulation method according to claim 5, characterized in that, Based on the current control stage of the control task, the first weight corresponding to the grid frequency deviation penalty term and the second weight corresponding to the wind turbine rotor speed deviation penalty term in the loss function of the objective function are dynamically adjusted, including: If the current control phase of the regulation task is the power grid frequency suppression phase, set W1>>W2; When the current control phase of the regulation task is the power grid frequency recovery phase, set W1>W2; When the current control phase of the regulation task is the grid frequency second sag suppression phase, set W1< <W2。 7. The power grid frequency regulation method according to claim 6, characterized in that, The constraints of the loss function of the objective function are as follows: The rotor speed of the wind turbine shall not be lower than the lower limit of safe operation and not higher than the upper limit of safe operation. The total active power output of the wind turbine converter shall not exceed its rated capacity; The rate of change of the optimal active power does not exceed the preset power ramp-up rate limit.

8. A power grid frequency control system based on disturbance dynamic characteristics, comprising: The acquisition module is configured to acquire transient data within a preset time period before and after the power grid fault occurs. The transient data includes at least the power grid frequency, power grid voltage amplitude, active power of wind turbine generators, and power exchange power of power grid interconnection lines. The generation module is configured to input the transient data into a trained neural network model to generate key disturbance dynamic features of the current fault. The construction module is configured to construct a corresponding objective function based on the key disturbance dynamic characteristics; The planning module is configured to use the Model Predictive Control (MPC) algorithm to plan the control task corresponding to the current fault into multiple different control stages and generate corresponding control instructions based on the objective function and the multiple different control stages. The control instructions include the optimal active power sequence. The control module is configured to control the wind turbine to perform grid frequency regulation according to the control command.

9. A computer-readable storage medium, characterized in that, It includes program instructions that, when executed by a processor, cause the method according to any one of claims 1-7 to be implemented.

10. An electronic device, characterized in that, include: processor; as well as A memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method according to any one of claims 1-7.

Images