Double-fed wind turbine generator rotating speed recovery method and related device

Abstract

The invention belongs to the technical field of wind power generation, and discloses a doubly-fed wind turbine generator rotating speed recovery method and a related device. A rotating speed recovery model constructed based on a model prediction control strategy is adopted, and the wind speed, the generator rotating speed, the output power and the power grid frequency at the current moment of a doubly-fed wind turbine generator are combined; and according to a preset reference frequency and a target generator rotating speed of the doubly-fed wind turbine generator, obtaining an active power load shedding amount of the doubly-fed wind turbine generator at the current moment, thereby realizing rotating speed recovery control of the doubly-fed wind turbine generator. A data driving model is adopted to improve the control precision; the target function considers the frequency deviation and the generator speed deviation to ensure that the control strategy can effectively suppress the secondary drop of the frequency and quickly recover the generator speed in practical application; and a model prediction control strategy is combined to adapt to continuously changing wind speed conditions. The problems that in an existing method, a physical model is not accurate and the control performance is not enough are effectively solved, and the overall performance and reliability of a wind power generation system are improved.

Description

Technical Field

[0001] This invention belongs to the field of wind power generation technology and relates to a method and related device for restoring the speed of a doubly fed wind turbine. Background Technology

[0002] Wind power, as a clean and renewable energy source, has been widely used globally. Doubly-fed induction generators (DFIGs) have become the mainstream model in wind farms due to their efficient and flexible power regulation capabilities. However, in actual operation, drastic changes in wind speed can cause fluctuations in generator speed, thus affecting power generation efficiency and grid stability. Therefore, achieving rapid and stable speed recovery of wind turbines under varying wind speed conditions is one of the key technologies for improving the performance of wind power systems.

[0003] Currently, existing DFIG speed control methods mainly rely on precise physical models to achieve speed control by adjusting the excitation current and rotor current. However, the actual operating environment of DFIGs is complex and variable, and physical models cannot accurately describe the system's state characteristics, resulting in unsatisfactory control performance. Furthermore, existing control methods often suffer from slow response speed and low control accuracy when dealing with sudden changes in wind speed, failing to fully utilize wind energy resources and consequently affecting power generation efficiency and equipment lifespan. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a method and related device for restoring the speed of a doubly fed wind turbine.

[0005] To achieve the above objectives, the present invention employs the following technical solution:

[0006] In a first aspect, the present invention provides a method for restoring the operating speed of a doubly-fed induction generator (DFIG) wind turbine, comprising: acquiring the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine at the current moment; based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine at the current moment, combined with a preset reference frequency and a target generator speed of the DFIG wind turbine, obtaining the active power load reduction of the DFIG wind turbine at the current moment through a pre-trained operating speed restoration model; wherein, the operating speed restoration model is constructed based on a model predictive control strategy, the prediction model in the operating speed restoration model adopts a data-driven model, used to predict the generator speed and output power of the DFIG wind turbine at the next moment based on the wind speed, generator speed, and output power of the DFIG wind turbine at the current moment; the objective function of the operating speed restoration model is to minimize the weighted sum of the frequency deviation and the generator speed deviation within the control cycle, and the constraints are output power limitation and generator speed limitation.

[0007] Optionally, the preset reference frequency of the doubly-fed wind turbine is obtained by the following formula:

[0008] f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ]

[0009] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G (k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN This is the rated power of the power grid.

[0010] The preset target generator speed of the doubly fed wind turbine is the optimal generator speed of the doubly fed wind turbine under maximum power point tracking control.

[0011] Optionally, the speed recovery model is constructed based on a model predictive control strategy, including: the speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

[0012] Optionally, the prediction model in the speed recovery model is constructed by: obtaining historical operating data of the doubly-fed wind turbine; establishing an initial prediction model based on a machine learning algorithm, and training the initial prediction model based on the historical operating data of the doubly-fed wind turbine to obtain the prediction model in the speed recovery model; wherein, the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

[0013] Optionally, the objective function of the speed recovery model is:

[0014]

[0015] Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δω r (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine.

[0016] The output power limit is:

[0017] PDFIG Min ≤PDFIG≤PDFIG Max

[0018] Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This represents the maximum output power of the doubly-fed wind turbine.

[0019] The generator speed limit is:

[0020] ω rmin ≤ω r ≤ω rmax

[0021] Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

[0022] Optionally, the step of obtaining the active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, through a pre-trained speed recovery model includes: solving the pre-trained speed recovery model based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, to obtain the active power load reduction sequence of the DFIG wind turbine within the control cycle; and extracting the first active power load reduction of the active power load reduction sequence as the active power load reduction of the DFIG wind turbine at the current moment.

[0023] Optionally, the step of obtaining the active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and through a pre-trained speed recovery model includes: constructing a simulation model of the DFIG wind turbine, and based on the simulation model, converting the pre-trained speed recovery model into a piecewise affine function through offline calculation; determining the control domain of the piecewise affine function of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and determining the linear control law based on the piecewise affine function; and obtaining the active power load reduction of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and combined with the determined linear control law.

[0024] In a second aspect, the present invention provides a doubly-fed induction generator (DFIG) wind turbine speed recovery system, comprising: a data acquisition module for acquiring the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine at the current moment; and a predictive control module for obtaining the active power load reduction of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with a preset reference frequency and a target generator speed of the DFIG wind turbine, through a pre-trained speed recovery model; wherein the speed recovery model is constructed based on a model predictive control strategy, and the predictive model in the speed recovery model adopts a data-driven model for predicting the generator speed and output power of the DFIG wind turbine at the next moment based on the wind speed, generator speed, and output power of the DFIG wind turbine at the current moment; the objective function of the speed recovery model is to minimize the weighted sum of the frequency deviation and generator speed deviation within the control period, and the constraints are output power limitation and generator speed limitation.

[0025] Optionally, the preset reference frequency of the doubly-fed wind turbine is obtained by the following formula:

[0026] f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ]

[0027] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G(k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN This is the rated power of the power grid.

[0028] The preset target generator speed of the doubly fed wind turbine is the optimal generator speed of the doubly fed wind turbine under maximum power point tracking control.

[0029] Optionally, the speed recovery model is constructed based on a model predictive control strategy, including: the speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

[0030] Optionally, the prediction model in the speed recovery model is constructed by: obtaining historical operating data of the doubly-fed wind turbine; establishing an initial prediction model based on a machine learning algorithm, and training the initial prediction model based on the historical operating data of the doubly-fed wind turbine to obtain the prediction model in the speed recovery model; wherein, the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

[0031] Optionally, the objective function of the speed recovery model is:

[0032]

[0033] Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δω r (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine.

[0034] The output power limit is:

[0035] PDFIG Min ≤PDFIG≤PDFIG Max

[0036] Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This represents the maximum output power of the doubly-fed wind turbine.

[0037] The generator speed limit is:

[0038] ω rmin ≤ω r≤ω rmax

[0039] Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

[0040] Optionally, the predictive control module is specifically used to: solve a pre-trained speed recovery model based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed induction generator (DFIG) wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, to obtain the active power load reduction sequence of the DFIG wind turbine within the control period; and extract the first active power load reduction of the active power load reduction sequence as the active power load reduction of the DFIG wind turbine at the current moment.

[0041] Optionally, the predictive control module is specifically used for: constructing a simulation model of a doubly-fed induction generator (DFIG) wind turbine, and based on the simulation model, converting a pre-trained speed recovery model into a piecewise affine function through offline calculation; determining the control domain of the piecewise affine function of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and determining the linear control law based on the piecewise affine function; and obtaining the active power load reduction of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and combined with the determined linear control law.

[0042] In a third aspect, the present invention provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the doubly-fed wind turbine speed recovery method described above.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described doubly-fed wind turbine speed recovery method.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] This invention discloses a speed recovery method for doubly-fed induction generator (DFIG) wind turbines. It employs a speed recovery model constructed based on a model predictive control (MMC) strategy. By combining the current wind speed, generator speed, output power, and grid frequency of the DFIG, along with the preset reference frequency and target generator speed, the active power load reduction of the DFIG at the current moment is obtained, thereby achieving speed recovery control of the DFIG. Specifically, the predictive model in the speed recovery model uses a data-driven approach to accurately predict the state of the DFIG, thus improving control accuracy. Simultaneously, the objective function is set to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, ensuring that the control strategy can effectively suppress secondary frequency drops and quickly restore the generator speed of the DFIG in practical applications. Furthermore, the rolling optimization of the MMC strategy adapts to constantly changing wind speed conditions, achieving rapid and stable speed recovery of the DFIG, improving power generation efficiency and equipment lifespan. This effectively solves the problems of inaccurate physical models and insufficient control performance in existing methods, improving the overall performance and reliability of wind power generation systems. Attached Figure Description

[0046] Figure 1 This is a flowchart of the speed recovery method for a doubly fed wind turbine according to an embodiment of the present invention.

[0047] Figure 2 This is a block diagram of the speed recovery system for a doubly fed wind turbine according to an embodiment of the present invention. Detailed Implementation

[0048] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0049] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0050] The present invention will now be described in further detail with reference to the accompanying drawings:

[0051] See Figure 1 In one embodiment of the present invention, a method for restoring the speed of a doubly fed wind turbine is provided, which aims to improve the speed recovery performance of the doubly fed wind turbine under wind speed fluctuations, thereby improving the operating efficiency and reliability of the doubly fed wind turbine. It is applicable to various wind power generation scenarios and has important application value and promotion prospects.

[0052] Specifically, the doubly-fed wind turbine speed recovery method of the present invention includes the following steps:

[0053] S1: Obtain the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine.

[0054] S2: Based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine, combined with the preset reference frequency and target generator speed of the doubly-fed wind turbine, the active power load reduction of the doubly-fed wind turbine at the current moment is obtained through a pre-trained speed recovery model.

[0055] The speed recovery model is constructed based on a model predictive control strategy. The prediction model in the speed recovery model adopts a data-driven model, which is used to predict the generator speed and output power of the doubly-fed wind turbine at the next moment based on the wind speed, generator speed and output power of the doubly-fed wind turbine at the current moment. The objective function of the speed recovery model is to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, and the constraints are output power limit and generator speed limit.

[0056] This invention discloses a speed recovery method for doubly-fed induction generator (DFIG) wind turbines. It employs a speed recovery model constructed based on a model predictive control (MMC) strategy. By combining the current wind speed, generator speed, output power, and grid frequency of the DFIG, along with the preset reference frequency and target generator speed, the active power load reduction of the DFIG at the current moment is obtained, thereby achieving speed recovery control of the DFIG. Specifically, the predictive model in the speed recovery model uses a data-driven approach to accurately predict the state of the DFIG, thus improving control accuracy. Simultaneously, the objective function is set to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, ensuring that the control strategy can effectively suppress secondary frequency drops and quickly restore the generator speed of the DFIG in practical applications. Furthermore, the rolling optimization of the MMC strategy adapts to constantly changing wind speed conditions, achieving rapid and stable speed recovery of the DFIG, improving power generation efficiency and equipment lifespan. This effectively solves the problems of inaccurate physical models and insufficient control performance in existing methods, improving the overall performance and reliability of wind power generation systems.

[0057] Explanatory model predictive control (MMC) is a forward-looking advanced control method based on a system dynamic model. It optimizes the system behavior over a finite future time period by using the current system state and the predictive model in each control cycle, generating an optimal control sequence that minimizes the preset performance index. It applies only the first step of the sequence control and repeats the optimization process in the next cycle. With its ability to perform rolling optimization, feedback correction, and handle multivariate constraints, it exhibits excellent dynamic response performance and robustness in multiple fields.

[0058] Interpretive, data-driven models, through the analysis of extensive historical operational data, can establish more accurate dynamic models of the system. Combined with predictive control technology, control strategies can be optimized in real time as wind speed changes, achieving precise control of the generator speed of doubly-fed wind turbines.

[0059] In one possible implementation, the preset reference frequency of the doubly-fed wind turbine is obtained by the following formula:

[0060] f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ]

[0061] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G (k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN This is the rated power of the power grid.

[0062] Explanatoryly, the preset reference frequency of the doubly-fed induction generator (DFIG) is set according to the droop curve, which allows the DFIG to dynamically adjust the reference frequency based on the real-time power conditions and rated parameters of the power grid. This helps to maintain the stability of the power grid frequency and enhances the adaptability and stability of the power grid during power fluctuations.

[0063] Optionally, the preset target generator speed of the doubly-fed wind turbine is the optimal generator speed of the doubly-fed wind turbine under maximum power point tracking (MPPT) control. Explanatoryly, MPPT is a widely used technology in renewable energy power generation systems designed to ensure that power generation equipment always operates near its maximum power point, thereby maximizing energy conversion efficiency.

[0064] In one possible implementation, the speed recovery model is constructed based on a model predictive control strategy, including: the speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

[0065] Interpretive, explicit model predictive control (EMC) is a powerful method for designing optimal control strategies, featuring a rapid online evaluation process. It utilizes control laws obtained offline for online predictive control. By pre-compiling and storing optimal control strategies for various possible states, EMC can quickly find and apply appropriate control inputs during runtime, thereby improving the real-time performance of the control system. Approximate model predictive control simplifies the dynamic model of the system through approximation methods, reducing computational complexity. In the control of doubly-fed induction generator (DFIG) wind turbines, simplified physical models or linearized models can be used for predictive control, ensuring improved computational efficiency while maintaining a certain level of control accuracy. Neural network-based model predictive control uses neural networks to build the dynamic model of the system. By training on a large amount of historical operating data, the neural network can capture the complex nonlinear relationships of the system, achieving accurate dynamic predictions.

[0066] In one possible implementation, the prediction model is constructed by: acquiring historical operating data of the doubly-fed wind turbine; establishing an initial prediction model based on a machine learning algorithm; and training the initial prediction model based on the historical operating data of the doubly-fed wind turbine to obtain the prediction model in the speed recovery model; wherein the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

[0067] Interpretive historical operating data of the doubly-fed induction generator (DFIG) wind turbines are collected, including wind speed, generator speed, and output power. The collected historical operating data undergoes preprocessing, such as noise reduction, missing value imputation, and normalization, to ensure data quality and consistency. Then, using the preprocessed data, a dynamic model of the DFIG wind turbines, built using machine learning algorithms (such as deep learning, support vector machines, or random forests), is trained to predict the impact of wind speed changes on generator speed and output power.

[0068] In one possible implementation, the objective function of the speed recovery model is:

[0069]

[0070] Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δωr (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine.

[0071] The explanatory objective function of the speed recovery model is designed to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, aiming to ensure the recovery of generator speed in the doubly-fed induction generator (DFIG) while suppressing secondary frequency drops. The specific values ​​of q and r can be adjusted during pre-training to account for the relative importance of frequency deviation and generator speed deviation in the optimization process.

[0072] Optionally, the output power limitation is:

[0073] PDFIG Min ≤PDFIG≤PDFIG Max

[0074] Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This represents the maximum output power of the doubly-fed wind turbine.

[0075] The generator speed limit is:

[0076] ω rmin ≤ω r ≤ω rmax

[0077] Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

[0078] Explanatoryly, by setting two constraints—output power limit and generator speed limit—the control strategy is ensured to conform to the actual operating conditions of the doubly-fed wind turbine, thus guaranteeing the availability of the control strategy.

[0079] In one possible implementation, the step of obtaining the active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and through a pre-trained speed recovery model, includes: solving the pre-trained speed recovery model based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine at the current moment, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, to obtain the active power load reduction sequence of the DFIG wind turbine within the control period; and extracting the first active power load reduction of the active power load reduction sequence as the active power load reduction of the DFIG wind turbine at the current moment.

[0080] Interpretively, the latest data is input into a pre-established speed recovery model to predict the response of the doubly-fed induction generator (DFIG) wind turbine under different control inputs over a future period. Based on the prediction results, the speed recovery model solves an optimization problem to determine the optimal control input sequence in the control time domain, i.e., the active power load reduction sequence of the DFIG wind turbine within the control cycle. Then, the first active power load reduction in the active power load reduction sequence is executed, and the process is advanced by one sampling cycle for a new round of data acquisition. This entire prediction, optimization, and execution process is then repeated continuously, forming a dynamic, real-time adjusted closed-loop control, enabling the DFIG wind turbine to continuously adapt to changes in operating conditions and maintain optimized performance.

[0081] Explaining this, in actual operation, the speed recovery model, through its data-driven model and rolling optimization mechanism, is designed to adapt to constantly changing wind speed conditions, addressing various wind speed variations. The data-driven model learns from historical data to predict the impact of wind speed changes on generator speed and output power. In each control cycle, it uses the latest real-time data to resolve the optimization problem and adjusts the active power load reduction in real time—this is its core adaptive and real-time adjustment capability. Therefore, for regular wind speed changes, the speed recovery model can achieve dynamic adaptation and performance maintenance through its inherent prediction and optimization mechanisms.

[0082] For example, the speed recovery model can be further optimized or updated, including continuously monitoring key performance indicators such as speed recovery time, frequency support effect, and constraint satisfaction of the doubly-fed induction generator (DFIG) under different wind speed variations. If performance significantly degrades in a specific wind speed range or under extreme wind speed changes, the parameters of the speed recovery model (such as the weighting coefficients of the objective function) need to be periodically recalibrated and optimized to improve its response capability under these conditions. Furthermore, when a large amount of new operating data accumulates or the system operating environment changes significantly, the new data can be used to retrain or incrementally learn the data-driven model to improve its prediction accuracy and adaptability to new operating conditions. After the improvements are completed, the updated speed recovery model is deployed to the actual system and rigorously validated in a controlled environment to ensure its stability and performance improvement, ultimately achieving continuous optimization and continuous improvement in system control performance.

[0083] In one possible implementation, the step of obtaining the active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and through a pre-trained speed recovery model includes: constructing a simulation model of the DFIG wind turbine, and based on the simulation model, converting the pre-trained speed recovery model into a piecewise affine function through offline calculation; determining the control domain of the piecewise affine function of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and determining the linear control law based on the piecewise affine function; and obtaining the active power load reduction of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and combined with the determined linear control law.

[0084] An interpretive simulation model of a doubly-fed induction generator (DFIG) wind turbine, incorporating a synchronous generator and a DFIG, was constructed in the Matlab / Simulink environment. The software system consists of Matlab / Simulink and RT-LAB simulation software. RT-LAB, as a multi-domain real-time simulation platform, seamlessly integrates with Matlab / Simulink, enabling real-time interaction between the Simulink model and the real-world environment. Matlab / Simulink is used for building the controller simulation model; specifically, it utilizes Simulink programming to construct the speed recovery model based on the model predictive control strategy.

[0085] Interpretive analysis was conducted by testing a speed recovery model on a doubly-fed induction generator (DFIG) wind turbine simulation model. By analyzing control strategies under different conditions, a linear control law was obtained through offline calculation. Then, predictive control was implemented online based on this linear control law. For example, the linear control law can typically be expressed as a piecewise affine function:

[0086] u k =F j *θ k +g j

[0087] Among them, u k Let F be the active power load reduction at time k. j Let g be the control law vector of control domain j. j Let θ be the control law matrix for control domain j. k This represents the system state of the doubly-fed wind turbine at time k.

[0088] For different system states, a pre-calculated linear control law is selected based on the piecewise affine function control domain to which the system state belongs to generate the control input, effectively improving the efficiency of control strategy generation.

[0089] In one possible implementation, based on a simulation model of a doubly fed wind turbine, the speed recovery method of the doubly fed wind turbine of the present invention and the existing methods are analyzed to determine the speed recovery and frequency drop.

[0090] At 50 seconds, a 200MW load is introduced at a certain node as a system power disturbance event. At this time, the system frequency decreases, and a comprehensive inertial control strategy is adopted during the frequency support phase, initiating a speed recovery strategy. In the traditional speed recovery strategy, the electromagnetic power of the doubly-fed induction generator (DFIG) is directly reduced to the corresponding power point on the maximum power point tracking curve. Two calculation examples are set up with wind power output levels of 30% and 50% of rated power.

[0091] By setting simulation experimental conditions, a comprehensive inertial control strategy is adopted to initiate a speed recovery strategy when a system power disturbance event occurs. This strategy consists of two stages: Inertial Response Stage: The doubly-fed induction generator (DFIG) responds to the system frequency change by increasing its output power to provide frequency support. At this time, the output power is increased from the power corresponding to the maximum power tracking point curve to the reference value. Speed ​​Recovery Control: After the inertial response ends, the output power of the DFIG is reduced, recovering to the optimal speed along the maximum power tracking point curve. In the speed recovery stage, based on the established speed recovery model, the active power load reduction of the DFIG is optimized in real time to minimize the secondary frequency drop and ensure the recovery of rotor speed.

[0092] Finally, by comparing the traditional speed recovery strategy, the improved speed recovery strategy and the method of the present invention, it was found that in simulation experiments conducted under different wind power output levels, the speed recovery method of the doubly fed wind turbine of the present invention can effectively suppress the secondary frequency drop while ensuring the recovery performance of the generator speed.

[0093] The following are embodiments of the apparatus of the present invention, which can be used to execute embodiments of the method of the present invention. For details not disclosed in the apparatus embodiments, please refer to the embodiments of the method of the present invention.

[0094] See Figure 2 In another embodiment of the present invention, a doubly fed wind turbine speed recovery system is provided, which can be used to implement the above-mentioned doubly fed wind turbine speed recovery method. Specifically, the doubly fed wind turbine speed recovery system includes a data acquisition module and a predictive control module.

[0095] The data acquisition module acquires the current wind speed, generator speed, output power, and grid frequency of the doubly-fed induction generator (DFIG). The predictive control module, based on the current wind speed, generator speed, output power, and grid frequency of the DFIG, combined with the preset reference frequency and target generator speed of the DFIG, uses a pre-trained speed recovery model to obtain the active power load reduction of the DFIG at the current moment. The speed recovery model is constructed based on a model predictive control strategy. The predictive model in the speed recovery model is a data-driven model used to predict the generator speed and output power of the DFIG at the next moment based on the current wind speed, generator speed, and output power. The objective function of the speed recovery model is to minimize the weighted sum of the frequency deviation and generator speed deviation within the control cycle, with constraints including output power limits and generator speed limits.

[0096] In one possible implementation, the preset reference frequency of the doubly-fed wind turbine is obtained by the following formula:

[0097] f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ]

[0098] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G (k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN This is the rated power of the power grid.

[0099] The preset target generator speed of the doubly fed wind turbine is the optimal generator speed of the doubly fed wind turbine under maximum power point tracking control.

[0100] In one possible implementation, the speed recovery model is constructed based on a model predictive control strategy, including: the speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

[0101] In one possible implementation, the prediction model in the speed recovery model is constructed by: acquiring historical operating data of the doubly-fed wind turbine; establishing an initial prediction model based on a machine learning algorithm; and training the initial prediction model based on the historical operating data of the doubly-fed wind turbine to obtain the prediction model in the speed recovery model; wherein the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

[0102] In one possible implementation, the objective function of the speed recovery model is:

[0103]

[0104] Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δω r (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine.

[0105] The output power limit is:

[0106] PDFIG Min ≤PDFIG≤PDFIG Max

[0107] Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This represents the maximum output power of the doubly-fed wind turbine.

[0108] The generator speed limit is:

[0109] ω rmin ≤ω r ≤ω rmax

[0110] Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

[0111] In one possible implementation, the predictive control module is specifically used to: solve a pre-trained speed recovery model based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed induction generator (DFIG) wind turbine, combined with a preset reference frequency and target generator speed of the DFIG wind turbine, to obtain the active power load reduction sequence of the DFIG wind turbine within the control period; and extract the first active power load reduction of the active power load reduction sequence as the active power load reduction of the DFIG wind turbine at the current moment.

[0112] In one possible implementation, the predictive control module is specifically used for: constructing a simulation model of a doubly-fed induction generator (DFIG) wind turbine, and based on the simulation model, converting a pre-trained speed recovery model into a piecewise affine function through offline calculation; determining the control domain of the piecewise affine function of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and determining a linear control law based on the piecewise affine function; and obtaining the active power load reduction of the DFIG wind turbine at the current moment based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and combined with the determined linear control law.

[0113] All relevant content of each step involved in the aforementioned embodiments of the doubly fed wind turbine speed recovery method can be referenced to the functional description of the corresponding functional module of the doubly fed wind turbine speed recovery system in the embodiments of the present invention, and will not be repeated here.

[0114] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0115] In another embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a doubly-fed wind turbine speed recovery method.

[0116] In another embodiment of the present invention, a storage medium is provided, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, the storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the doubly-fed wind turbine speed recovery method in the above embodiments.

[0117] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0118] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0119] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0120] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for restoring the speed of a doubly-fed wind turbine, characterized in that, include: Obtain the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine. Based on the current wind speed, generator speed, output power and grid frequency of the doubly-fed wind turbine, combined with the preset reference frequency and target generator speed of the doubly-fed wind turbine, the active power load reduction of the doubly-fed wind turbine at the current moment is obtained through a pre-trained speed recovery model. The speed recovery model is constructed based on a model predictive control strategy. The prediction model in the speed recovery model adopts a data-driven model, which is used to predict the generator speed and output power of the doubly-fed wind turbine at the next moment based on the wind speed, generator speed and output power of the doubly-fed wind turbine at the current moment. The objective function of the speed recovery model is to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, and the constraints are output power limit and generator speed limit.

2. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The preset reference frequency of the doubly fed wind turbine is obtained by the following formula: f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G (k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN Rated power of the power grid; The preset target generator speed of the doubly fed wind turbine is the optimal generator speed of the doubly fed wind turbine under maximum power point tracking control.

3. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The speed recovery model is constructed based on a model predictive control strategy and includes: The speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

4. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The prediction model in the speed recovery model is constructed using the following formula: Obtain historical operating data of double-fed wind turbine units; An initial prediction model is established based on a machine learning algorithm, and the initial prediction model is trained based on the historical operating data of the doubly fed wind turbine to obtain the prediction model in the speed recovery model; wherein, the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

5. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The objective function of the speed recovery model is: Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δω r (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine. The output power limit is: PDFIG Min ≤PDFIG≤PDFIG Max Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This is the maximum output power of the doubly-fed wind turbine. The generator speed limit is: oh rmin ≤ω r ≤ω rmax Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

6. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment, based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, is obtained through a pre-trained speed recovery model. This includes: Based on the current wind speed, generator speed, output power and grid frequency of the doubly fed wind turbine, combined with the preset reference frequency and target generator speed of the doubly fed wind turbine, the pre-trained speed recovery model is solved to obtain the active power load reduction sequence of the doubly fed wind turbine within the control cycle. Extract the first active power load reduction from the active power load reduction sequence and use it as the active power load reduction of the doubly-fed wind turbine at the current moment.

7. The method for restoring the speed of a doubly-fed wind turbine according to claim 1, characterized in that, The active power load reduction of the doubly-fed induction generator (DFIG) wind turbine at the current moment, based on the wind speed, generator speed, output power, and grid frequency of the DFIG wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, is obtained through a pre-trained speed recovery model. This includes: A simulation model of a doubly fed wind turbine is constructed, and based on the simulation model, the pre-trained speed recovery model is converted into a piecewise affine function through offline calculation. Based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine, and in conjunction with the preset reference frequency and target generator speed of the doubly-fed wind turbine, the piecewise affine function control domain of the doubly-fed wind turbine at the current moment is determined, and the linear control law is determined in conjunction with the piecewise affine function. Based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed induction generator (DFIG) wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and the determined linear control law, the current active power load reduction of the DFIG wind turbine is obtained.

8. A speed recovery system for a doubly-fed wind turbine, characterized in that, include: The data acquisition module is used to acquire the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine. The predictive control module is used to obtain the active power load reduction of the doubly-fed wind turbine at the current moment based on the wind speed, generator speed, output power and grid frequency of the doubly-fed wind turbine, combined with the preset reference frequency and target generator speed of the doubly-fed wind turbine, through a pre-trained speed recovery model. The speed recovery model is constructed based on a model predictive control strategy. The prediction model in the speed recovery model adopts a data-driven model, which is used to predict the generator speed and output power of the doubly-fed wind turbine at the next moment based on the wind speed, generator speed and output power of the doubly-fed wind turbine at the current moment. The objective function of the speed recovery model is to minimize the weighted sum of frequency deviation and generator speed deviation within the control cycle, and the constraints are output power limit and generator speed limit.

9. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The preset reference frequency of the doubly fed wind turbine is obtained by the following formula: f ref =f n -[-K R (P G (k)-P L (k)+P W (k)) / P LN ] Among them, f ref The preset reference frequency for the doubly fed wind turbine, f n K is the rated frequency of the power grid. R P is the droop coefficient. G (k) represents the power of conventional generating units in the power grid at time k, P L (k) represents the load power of the power grid at time k, P W (k) represents the power of the doubly-fed wind turbine at time k, P LN Rated power of the power grid; The preset target generator speed of the doubly fed wind turbine is the optimal generator speed of the doubly fed wind turbine under maximum power point tracking control.

10. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The speed recovery model is constructed based on a model predictive control strategy and includes: The speed recovery model is constructed based on an explicit model predictive control strategy, an approximate model predictive control strategy, or a neural network-based model predictive control strategy.

11. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The prediction model in the speed recovery model is constructed using the following formula: Obtain historical operating data of double-fed wind turbine units; An initial prediction model is established based on a machine learning algorithm, and the initial prediction model is trained based on the historical operating data of the doubly fed wind turbine to obtain the prediction model in the speed recovery model; wherein, the machine learning algorithm is a deep learning algorithm, a support vector machine algorithm, or a random forest algorithm.

12. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The objective function of the speed recovery model is: Where q is the frequency deviation weighting coefficient, and N p To control the total number of time points in the cycle, f(k) is the grid frequency of the doubly-fed induction generator at time k. ref The preset reference frequency for the doubly-fed induction generator (DFIG) wind turbine, r is the generator speed deviation weighting coefficient, and Δω r (k) represents the difference between the target generator speed and the generator speed at time k of the doubly fed wind turbine. The output power limit is: PDFIG Min ≤PDFIG≤PDFIG Max Where PDFIG represents the output power of the doubly-fed wind turbine generator. Min PDFIG is the minimum output power of a doubly-fed wind turbine. Max This is the maximum output power of the doubly-fed wind turbine. The generator speed limit is: oh rmin ≤ω r ≤ω rmax Where, ω r ω is the generator speed of the doubly-fed wind turbine. rmin ω is the minimum generator speed of the doubly-fed wind turbine. rmax This is the maximum generator speed of the doubly fed wind turbine.

13. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The predictive control module is specifically used for: Based on the current wind speed, generator speed, output power and grid frequency of the doubly fed wind turbine, combined with the preset reference frequency and target generator speed of the doubly fed wind turbine, the pre-trained speed recovery model is solved to obtain the active power load reduction sequence of the doubly fed wind turbine within the control cycle. Extract the first active power load reduction from the active power load reduction sequence and use it as the active power load reduction of the doubly-fed wind turbine at the current moment.

14. The doubly-fed wind turbine speed recovery system according to claim 8, characterized in that, The predictive control module is specifically used for: A simulation model of a doubly fed wind turbine is constructed, and based on the simulation model, the pre-trained speed recovery model is converted into a piecewise affine function through offline calculation. Based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed wind turbine, and in conjunction with the preset reference frequency and target generator speed of the doubly-fed wind turbine, the piecewise affine function control domain of the doubly-fed wind turbine at the current moment is determined, and the linear control law is determined in conjunction with the piecewise affine function. Based on the current wind speed, generator speed, output power, and grid frequency of the doubly-fed induction generator (DFIG) wind turbine, combined with the preset reference frequency and target generator speed of the DFIG wind turbine, and the determined linear control law, the current active power load reduction of the DFIG wind turbine is obtained.

15. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the doubly fed wind turbine speed recovery method as described in any one of claims 1 to 7.

16. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the doubly fed wind turbine speed recovery method as described in any one of claims 1 to 7.

Images