A wind turbine yaw control dynamic optimization method, system, terminal and medium

Abstract

The present application relates to the technical field of wind turbine control, and discloses a wind turbine yaw control dynamic optimization method, system, terminal and medium, according to historical operation data of the wind turbine, optimization is carried out through an intelligent algorithm, and yaw threshold and delay time are updated and iterated in a fixed period, so that a higher-precision yaw control process is realized, the cost of optimization and reconstruction is reduced, and the adaptability of yaw control is improved. According to the actual operation environment of the wind turbine, the yaw controller parameter update iteration can be carried out in a certain period. By deploying the optimization solution process in the wind farm to realize periodic automatic start calculation, the yaw control optimization can collect historical operation data in a short time every certain period, the optimization result is updated and iterated, and the adaptability of the yaw control of the unit is greatly improved.

Description

A dynamic optimization method, system, terminal and medium for yaw control of wind turbine units Technical Field

[0001] This invention relates to the field of wind turbine control technology, specifically to a dynamic optimization method, system, terminal, and medium for wind turbine yaw control. Background Technology

[0002] While wind resources in nature are continuous, they also exhibit fluctuations and randomness. Therefore, it is necessary to implement certain controls on wind turbine and wind farm scheduling to improve the power generation efficiency of the units and achieve efficient utilization of wind resources. The control of wind turbine units aims to improve the performance of individual units, including yaw control to achieve maximum wind energy conversion efficiency. Typically, yaw control of wind turbine units is controlled by a specific set of parameters: the yaw control threshold and the yaw control delay time. The control principle is that when the deviation angle between the inflow wind direction and the current nacelle position is greater than the yaw control threshold and lasts for more than the yaw control delay time, the wind turbine unit initiates yaw control to align with the wind. This is the yaw control strategy of the vast majority of wind turbine units currently on the market. Yaw control in wind turbines is a typical open-loop control. Once the yaw error of the wind turbine meets the start-up conditions for yaw control, the yaw controller sends a command to the yaw actuator system, and the yaw actuator motor begins to perform its action. Yaw control continues until the stop conditions are met, at which point it ends. During this period, the main control system cannot influence the yaw execution process; therefore, there is no dynamic adjustment process in yaw control. This characteristic leads to existing wind turbine yaw control strategies often suffering from poor control accuracy, poor robustness, and weak adaptability. These shortcomings have become a key research focus for yaw control optimization.

[0003] There are numerous optimization methods for wind turbine yaw control systems, all ultimately aiming to increase power generation by improving the accuracy of wind response. Common yaw control optimization methods include yaw correction, which improves the accuracy of wind speed and direction measurements by replacing wind condition measurement devices with high-precision ones, such as nacelle-mounted lidar. Simultaneously, based on historical operating data of the wind turbine, the inherent yaw error of the turbine is identified and corrected, thus improving wind response accuracy. Secondly, incorporating wind direction prediction into yaw control optimization is also an effective method. By predicting wind direction, advance yaw can be achieved, reducing power generation losses due to yaw errors. Furthermore, optimizing and tuning yaw control parameters, namely the yaw threshold and yaw delay time, according to different wind speed ranges can also improve the turbine's yaw wind response efficiency and reduce wind curtailment rates to some extent.

[0004] The aforementioned methods, which utilize high-precision wind measuring instruments such as lidar for yaw correction, are effective but typically expensive. It's difficult to install lidar or similar devices on every wind turbine in a wind farm, so in most cases, only typical turbines are selected for optimization and correction before being applied to other turbines. This can lead to several problems: firstly, the correction values ​​may not be well-suited for other turbines, resulting in poor yaw correction or even adverse effects; secondly, correction based on historical data usually involves mining yaw errors from a period of turbine operation, without updating the correction values, leading to decreased control accuracy over longer timescales. While yaw control optimization incorporating wind speed and direction prediction is effective, its performance is heavily dependent on the accuracy of these predictions. Current wind speed and direction prediction technologies cannot achieve high accuracy within specific ranges, making this yaw optimization method largely theoretical and difficult to apply in practical engineering. The yaw control of wind turbines is affected by the complex and variable inflow wind during actual operation, which often results in poor adaptability. In addition, the wind conditions vary greatly in different regions, and the yaw control parameters set when the turbine is manufactured will have different control effects in different operating environments. Therefore, targeted optimization of the yaw control of the turbine is of engineering significance. Summary of the Invention

[0005] In order to overcome the shortcomings of the existing technology, the present invention aims to provide a dynamic optimization method, system, terminal and medium for yaw control of wind turbine units, so as to solve the technical problems of existing technology being unable to optimize and correct each wind turbine, having low adaptability, poor yaw correction effect and errors in historical data correction that affect the accuracy of correction.

[0006] This invention is achieved through the following technical solution:

[0007] A dynamic optimization method for yaw control of wind turbine units includes the following steps:

[0008] Step 1: Establish a linearized model of the wind turbine and obtain the linearized state-space expression of the turbine based on the structural parameters of the wind turbine. Take wind speed, pitch angle and generator rated torque as inputs, and power and generator torque as outputs.

[0009] Step 2: Fit the coefficient of the influence of wind direction deviation on power using the linearized model of the wind turbine, and correct the power output of the linearized model of the wind turbine.

[0010] Step 3: Calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0011] Step 4: Establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw operations. Construct a minimum optimization problem and use Pareto optimality theory to seek the Pareto optimal solution model.

[0012] Step 5: The server stores several Pareto optimal solution models, each corresponding to a wind turbine in the wind farm. The data optimization cycle is set, and communication is established between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system. This allows the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0013] Preferably, in step 1, the linearized state-space expression of the wind turbine is obtained based on the structural parameters of the wind turbine, as shown in the following formula:

[0014]

[0015] Where A is the system state coefficient matrix, B is the system control coefficient matrix, C is the output state coefficient matrix, and D is the output control coefficient matrix. x, u, and y are all vectors, where u is the input vector, including wind speed, rated torque, and pitch angle; y is the output vector, including power and generator speed; and x and x' are the current state and the next state of the system, respectively.

[0016] Preferably, in step 2, the linearization model of the wind turbine fits the coefficients of the influence of wind direction deviation on power. The specific process is as follows:

[0017] Let the wind energy capture efficiency of the wind turbine be P, and the yaw error angle of the wind turbine be θ. Then the formula for the wind energy capture efficiency is as follows:

[0018]

[0019] Where: P is the wind energy captured by the unit, in W; ρ is the air density, in kg / m3; R is the radius of the wind turbine rotor, in m; V is the inflow wind speed, in m / s; θ is the yaw error angle, in rad; n is an undetermined coefficient;

[0020] The energy loss generated during the conversion of wind energy captured by the wind turbine into active power is denoted as P. 损 Then the actual active power generated by the unit is: P 有功 =PP 损 ;

[0021] Where the yaw error is θ:

[0022] P 有功 =P0·Cos n θ

[0023] Wherein, P0 is the active power of the wind turbine when the yaw error is 0°;

[0024] Based on the yaw error sequence and active power recorded in the SCADA system of the wind turbine, the curve is fitted using the least squares method or the Fourier series approximation method to determine the value of the undetermined coefficient n.

[0025] Preferably, in step 3, a yaw control algorithm is written using Python or C++ to calculate the yaw control process and power output of the wind turbine under the input wind conditions based on the power output of the corrected linearized model. The yaw control process is represented by the absolute azimuth angle of the wind turbine nacelle and the yaw flag. The yaw flag is a series of digital signals. When the yaw motor is not running, the yaw flag is set to 0. When the yaw motor performs a yaw control action, the yaw flag is set to 1. The number of times the yaw flag is triggered is counted to determine the number of yaw executions within the time period.

[0026] Preferably, in step 4, the formula for the minimum value optimization problem is constructed as follows:

[0027]

[0028] Where f1 and f2 represent the mapping relationship between negative equivalent power generation and the number of yaw executions relative to the threshold and delay time; x is the controller parameter, i.e. the threshold and the delay time; X represents the value range of the threshold and the delay time, with the threshold being 5-20° and the delay time being 20s-210s, both being natural numbers. In the minimum value optimization problem, the maximum number of yaw control executions is set as the boundary condition.

[0029] Furthermore, in step 4, a multi-objective optimization problem is established to improve the equivalent power generation while limiting the significant increase in the number of yaw operations. The Pareto optimal solution is sought using Pareto optimality theory. The specific process is as follows:

[0030] Based on the initial yaw threshold and delay time, a genetic algorithm is used to optimize the solution. The yaw threshold and delay time corresponding to the optimal solution on the Pareto front are calculated and compared with the initial value. If the solution on the front is better than the initial solution, the middle point on the front is selected as the optimization solution for iterative optimization of the threshold and delay time. If the initial value is also on the front, no update is made, and one cycle of optimization is completed. Finally, all algorithms are packaged.

[0031] Preferably, in step 5, the specific process of updating the new threshold and delay time calculated by the optimized solution algorithm to the main control system of the corresponding unit is as follows:

[0032] The data optimization cycle is set to x hours. Every x hours, wind speed, wind direction, power and yaw execution process data within x hours are collected. With the current threshold, delay time and wind condition information as input, the optimization algorithm corresponding to the unit is started in the server to solve for the new optimal value and compare it with the current value. If it is better than the current value, it is iteratively written into the main control system with the new threshold and delay time.

[0033] A dynamic optimization system for yaw control of a wind turbine includes:

[0034] The model building module is used to build a linearized model of the wind turbine and obtain the linearized state-space expression of the unit based on the structural parameters of the wind turbine. The wind speed, pitch angle and generator rated torque are used as inputs, and the power and generator torque are used as outputs.

[0035] The model correction module is used to fit the coefficient of the influence of wind direction deviation on power through the linearization model of the wind turbine, and to correct the power output of the linearization model of the wind turbine.

[0036] The first data processing module is used to calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0037] The second data processing module is used to establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw executions. It constructs a minimum optimization problem, uses Pareto optimality theory to seek the Pareto optimal solution, and completes the dynamic optimization of wind turbine yaw control.

[0038] The third data processing module is used to store several Pareto optimal solution models in the server, where each Pareto optimal solution model corresponds to a wind turbine in the wind farm. It sets a data optimization cycle, establishes communication between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system, and uses the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0039] A mobile terminal includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the dynamic optimization method for yaw control of a wind turbine as described above.

[0040] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the dynamic optimization method for yaw control of a wind turbine as described above.

[0041] Compared with the prior art, the present invention has the following beneficial technical effects:

[0042] This invention provides a dynamic optimization method for yaw control of wind turbines. Based on historical operating data of the wind turbines, it uses intelligent algorithms to optimize and update the yaw threshold and delay time within a fixed period to achieve a more precise yaw control process, reduce the cost of optimization and retrofitting, and improve the adaptability of yaw control. This invention can update and iterate the yaw controller parameters within a certain period according to the actual operating environment of the wind turbines. By deploying the optimization solution process at the wind farm to achieve periodic automated start-up calculations, yaw control optimization can collect historical operating data for a short period at regular intervals, update the optimization results, and greatly improve the adaptability of the turbine's yaw control.

[0043] This invention also provides a dynamic optimization system for yaw control of wind turbine generators. The main control algorithm of wind turbine generators is generally deployed in the main control PLC of the wind turbine generator, and the languages ​​that can be used are very limited. The dynamic optimization device for yaw control of wind turbine generators of this invention encapsulates the optimization method into a module to achieve high-speed calculation, ensuring the speed of optimization solution and yaw control parameter update.

[0044] This invention also provides a mobile terminal that enables dynamic optimization of yaw control for different wind turbines when the computer program for each wind turbine is executed by a processor. Each computer program is independent, and the yaw control parameters for each wind turbine are optimized in parallel without affecting each other. Typically, yaw control uses the same control parameters across different turbines, but this invention allows different turbines to use different yaw control parameters, achieving a more ideal control effect. Attached Figure Description

[0045] Figure 1 is a flowchart of the dynamic optimization method for yaw control of wind turbines in this invention.

[0046] Figure 2 is a flowchart of the dynamic optimization method for wind turbine yaw control in an embodiment of the present invention;

[0047] Figure 3 is a model diagram of a wind turbine in an embodiment of the present invention;

[0048] Figure 4 is a graph showing the relationship between yaw control parameters and the number of yaw executions in an embodiment of the present invention.

[0049] Figure 5 is a Pareto front diagram in an embodiment of the present invention;

[0050] Figure 6 is a diagram illustrating the process of establishing the yaw control optimization solution model in an embodiment of the present invention. Detailed Implementation

[0051] 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. 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.

[0052] The present invention will now be described in further detail with reference to the accompanying drawings:

[0053] The purpose of this invention is to provide a dynamic optimization method, system, terminal, and medium for yaw control of wind turbine units, in order to solve the technical problems of existing technologies that cannot optimize and correct each wind turbine, have low adaptability, poor yaw correction effect, and have errors in historical data correction that affect the accuracy of correction.

[0054] Specifically, as shown in Figure 1, the dynamic optimization method for yaw control of this wind turbine includes the following steps:

[0055] Step 1: Establish a linearized model of the wind turbine and obtain the linearized state-space expression of the turbine based on the structural parameters of the wind turbine. Take wind speed, pitch angle and generator rated torque as inputs, and power and generator torque as outputs.

[0056] Specifically, the linearized state-space expression of the wind turbine is obtained based on the structural parameters of the wind turbine, as shown in the following formula:

[0057]

[0058] Where A is the system state coefficient matrix, B is the system control coefficient matrix, C is the output state coefficient matrix, and D is the output control coefficient matrix. x, u, and y are all vectors, where u is the input vector, including wind speed, rated torque, and blade pitch angle, and y is the output vector, including power and generator speed; x and These represent the current state of the system and the state at the next moment, respectively.

[0059] Step 2: Fit the coefficient of the influence of wind direction deviation on power using the linearized model of the wind turbine, and correct the power output of the linearized model of the wind turbine.

[0060] Specifically, the linearization model of the wind turbine is fitted with the coefficients of the influence of wind direction deviation on power. The specific process is as follows:

[0061] Let the wind energy capture efficiency of the wind turbine be P, and the yaw error angle of the wind turbine be θ. Then the formula for the wind energy capture efficiency is as follows:

[0062]

[0063] Where: P is the wind energy captured by the unit, in W; ρ is the air density, in kg / m3; R is the radius of the wind turbine rotor, in m; V is the inflow wind speed, in m / s; θ is the yaw error angle, in rad; n is an undetermined coefficient.

[0064] The energy loss generated during the conversion of wind energy captured by the wind turbine into active power is denoted as P. 损 Then the actual active power generated by the unit is: P 有功 =PP 损 ;

[0065] Where the yaw error is θ:

[0066] P 有功 =P0·cos n θ

[0067] Wherein, P0 is the active power of the wind turbine when the yaw error is 0°;

[0068] Based on the yaw error sequence and active power recorded in the SCADA system of the wind turbine, the curve is fitted using the least squares method or the Fourier series approximation method to determine the value of the undetermined coefficient n.

[0069] Step 3: Calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0070] Specifically, a yaw control algorithm is written using Python or C++ to calculate the yaw control process and power output of the wind turbine under the input wind conditions, based on the power output of the corrected linearized model. The yaw control process is represented by the absolute azimuth of the wind turbine nacelle and the yaw flag. The yaw flag is a series of digital signals. When the yaw motor is not running, the yaw flag is set to 0. When the yaw motor performs a yaw control action, the yaw flag is set to 1. The number of times the yaw flag is triggered is counted to determine the number of yaw executions within that time period.

[0071] Step 4: Establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw executions. Construct a minimum optimization problem and use Pareto optimality theory to find the Pareto optimal solution, thus completing the dynamic optimization of wind turbine yaw control.

[0072] Specifically, the formula for constructing the minimum value optimization problem is as follows:

[0073]

[0074] Where f1 and f2 represent the mapping relationship between negative equivalent power generation and the number of yaw executions relative to the threshold and delay time; x is the controller parameter, i.e. the threshold and delay time; X represents the value range of the threshold and delay time, with the threshold ranging from 5 to 20° and the delay time ranging from 20s to 210s, both being natural numbers.

[0075] In the minimum optimization problem, the maximum number of yaw control executions is set as the boundary condition.

[0076] Specifically, a multi-objective optimization problem is established to improve the equivalent power generation while limiting the number of yaw operations. The Pareto optimal solution is sought using Pareto optimality theory. The specific process is as follows:

[0077] Based on the initial yaw threshold and delay time, a genetic algorithm is used to optimize the solution at the Pareto front, calculating the yaw threshold and delay time corresponding to the optimal solution and comparing it with the initial value. If the solution at the front is better than the initial solution, the intermediate point at the front is selected as the optimized solution for iterative optimization of the threshold and delay time. If the initial value is also at the front, no update is made, completing one cycle of optimization. Finally, all algorithms are encapsulated and packaged.

[0078] Step 5: The server stores several Pareto optimal solution models, each corresponding to a wind turbine in the wind farm. The data optimization cycle is set, and communication is established between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system. This allows the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0079] The specific process of updating the new threshold and delay time calculated by the optimized solution algorithm to the main control system of the corresponding unit is as follows:

[0080] The data optimization cycle is set to x hours. Every x hours, wind speed, wind direction, power and yaw execution process data within x hours are collected. With the current threshold, delay time and wind condition information as input, the optimization algorithm corresponding to the unit is started in the server to solve for the new optimal value and compare it with the current value. If it is better than the current value, it is iteratively written into the main control system with the new threshold and delay time.

[0081] Example

[0082] This embodiment presents a yaw control optimization method based on dynamically adjusting the yaw control threshold and yaw control delay time. To implement this dynamic optimization method, as shown in Figures 2 and 6, it is first necessary to establish a multi-objective optimization problem solution model for dynamic yaw control optimization, encapsulate the model in a server, and set an execution cycle to perform optimization every 3 hours. Detailed steps are as follows:

[0083] Step 1: Establish a linearized model of the wind turbine, as shown in Figure 3. Using wind speed, pitch angle, and generator rated torque as inputs, and power and generator torque as outputs, establish the linearized state-space expression of the turbine based on its structural parameters, etc.

[0084]

[0085] Where A is the system state coefficient matrix, B is the system control coefficient matrix, C is the output state coefficient matrix, and D is the output control coefficient matrix.

[0086] Step 2: Based on the linearized model of the wind turbine, fit the coefficients of the influence of wind direction deviation on power. Assuming the wind energy capture efficiency of the turbine is P, and at a certain moment, the yaw error angle of the turbine is θ, then:

[0087]

[0088] Where: P represents the wind energy captured by the unit, in W;

[0089] ρ is the density of air, in kg / m³. 3 ;

[0090] R is the radius of the wind turbine rotor, in meters (m).

[0091] V is the inflow velocity, in m / s;

[0092] θ is the yaw error angle, in rad;

[0093] n is an undetermined coefficient.

[0094] The energy loss generated during the conversion of wind energy captured by a wind turbine into active power is denoted as P. 损 The actual active power P generated by the unit 有功 =PP 损 When the yaw error is θ:

[0095] P 有功 =P0·cos n θ

[0096] Where P0 is the active power of the train when the yaw error is 0°.

[0097] Based on the yaw error sequence and active power recorded in the SCADA system of the wind turbine, the curve is fitted using the least squares method or the Fourier series approximation method to determine the value of the undetermined coefficient n.

[0098] Step 3: Linearization Model Correction. Since the original unit linearization model does not include the input of the yaw error angle, according to Step 2, when the angle between the inflow wind direction angle and the nacelle position is 0, the active power output of the model is corrected.

[0099] Step 4: Write the yaw control algorithm using Python or C++. The initial control algorithm sets the threshold and delay time to the parameter values ​​in the original control strategy of the generator set. The control algorithm needs to discretize the generator set model from Step 3 and incorporate it into the program. The algorithm should be able to calculate the yaw control process and active power output of the generator set under given wind speed and direction sequences. The yaw control process is represented by the absolute azimuth angle of the generator set's nacelle (relative to true north) and the yaw flag. The yaw flag is a series of digital signals; when the yaw motor starts, the flag is set to 0; when the yaw motor performs a yaw control action, the flag is set to 1. Finally, the algorithm counts the number of yaw executions within a given time period by counting the number of times the yaw flag is triggered, as shown in Figure 4.

[0100] Step 5: Establish a multi-objective optimization problem. In the yaw control dynamic optimization method proposed in this invention, increasing power generation is not the only control objective. According to the simulation results of the yaw control algorithm, setting the threshold and delay time too small will lead to an exponential increase in the number of yaw control operations, which is very detrimental to the healthy operation of the unit.

[0101] Construct a minimum optimization problem:

[0102]

[0103] Where f1 and f2 represent the mapping relationship between negative equivalent power generation and the number of yaw executions relative to the threshold and delay time, and x represents the controller parameters, namely the threshold and delay time. X represents the value range of the threshold and delay time, with the threshold ranging from 5 to 20° and the delay time ranging from 20s to 210s, both taken as natural numbers. In the optimization problem, the maximum number of yaw control executions is set as a boundary condition.

[0104] Step Six: Based on Pareto optimization theory, use a genetic algorithm to solve the multi-objective optimization problem in Step Five. The optimal solution to a multi-objective optimization problem is often not unique, but Pareto optimality theory provides a solution approach. According to Pareto optimality theory, taking the minimum multi-objective optimization problem as an example, the feasible region of its solution has a Pareto front. Solutions on the Pareto front can be called Pareto optimal solutions, as shown in Figure 5.

[0105] Specifically, for the multi-objective optimization problem proposed in step five, based on the solutions corresponding to the initial yaw threshold and delay time, a genetic algorithm is used to find the optimal solution on the Pareto front, calculating the yaw threshold and delay time corresponding to the best solution and comparing it with the initial value. If the solution on the front is better than the initial solution, the intermediate point on the front is selected as the optimal solution for iterative optimization of the threshold and delay time. If the initial value is also on the front, no update is made, completing one cycle of optimization. Finally, all algorithms are encapsulated and packaged.

[0106] Step 7: Install a dedicated yaw control dynamic optimization solution server at the wind farm site. The server stores several optimization solution models constructed in Step 6, ensuring the independence between models. Each model corresponds to one unit, and the optimization calculation between models adopts a parallel mode.

[0107] The model takes wind speed and direction and the current controller parameters as input and outputs optimized controller parameters. The optimization algorithm is set to execute once every 3 hours. It should be noted that if the unit is currently in yaw control when the next optimization cycle needs to begin after 3 hours, the optimization algorithm will not execute temporarily, but will wait for 1 minute after the current yaw action is completed before executing the optimization algorithm.

[0108] Step 8: Establish communication between the optimization solution server, the wind farm SCADA database, and the wind turbine main control system to ensure that the server can collect historical data from the SCADA database and that the new thresholds and delay times calculated by the optimization solution algorithm can be updated to the main control system of the corresponding unit.

[0109] Step 9: Every 3 hours, the unit collects wind speed, wind direction, power, and yaw execution process data for the past 3 hours. Using the current threshold, delay time, and wind condition information as input, the corresponding optimization algorithm for the unit is started on the server to solve for the new optimal value and compare it with the current value. If it is better than the current value, the new threshold and delay time are iteratively written into the main control.

[0110] Based on common yaw control optimization methods, this invention proposes a dynamic yaw control optimization method that can update and iterate controller parameters within a certain period according to the actual operating environment of the wind turbine. Through optimization problem modeling and algorithm writing and encapsulation, it enables field deployment. The key technical points and protected aspects of this invention are as follows:

[0111] (1) Dynamic optimization technology for yaw controller parameters. Existing yaw control optimization methods mostly involve analyzing problems in the current yaw control of the wind turbine or introducing new predictive technologies to perform a one-time optimization or correction of the original yaw control strategy. While such optimization methods can achieve good results, the operating environment of wind turbines is complex and variable. Optimization results based on a single set of historical operating data, although achieving ideal results, are not always optimal over a longer time scale. The dynamic optimization method proposed in this invention deploys the optimization solution process within the wind farm to achieve periodic automated startup calculations. This allows yaw control optimization to collect historical operating data over a short period at regular intervals, updating and iterating the optimization results, greatly improving the adaptability of the wind turbine's yaw control.

[0112] (2) Encapsulation and field deployment of yaw control optimization algorithm. The main control algorithm of wind turbine is generally deployed in the main control PLC of the unit, and the languages ​​that can be used are very limited. The optimization solution algorithm proposed in this invention is deployed to a high-performance server, which can use higher-level languages, such as Python and C++, to solve the optimization problem, and can also achieve high-speed computing, ensuring the speed of optimization solution and update iteration.

[0113] (3) Independent computation of yaw control optimization solution. First, different models of turbines in the wind farm need to be modeled separately, and different linearized models of the turbines need to be established. Second, the optimization solution algorithms are deployed independently on the server and are computed in parallel during operation. Usually, yaw control uses the same control parameters across different turbines, but the independent optimization computation method proposed in this invention can enable different turbines to use different yaw control parameters, achieving a more ideal control effect.

[0114] The present invention also provides a dynamic optimization system for yaw control of wind turbine generators, including a model building module, a model correction module, a first data processing module, a second data processing module and a third data processing module;

[0115] The model building module is used to build a linearized model of the wind turbine and obtain the linearized state-space expression of the unit based on the structural parameters of the wind turbine. The wind speed, pitch angle and generator rated torque are used as inputs, and the power and generator torque are used as outputs.

[0116] The model correction module is used to fit the coefficient of the influence of wind direction deviation on power through the linearization model of the wind turbine, and to correct the power output of the linearization model of the wind turbine.

[0117] The first data processing module is used to calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0118] The second data processing module is used to establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw executions. It constructs a minimum optimization problem, uses Pareto optimality theory to seek the Pareto optimal solution, and completes the dynamic optimization of wind turbine yaw control.

[0119] The third data processing module is used to store several Pareto optimal solution models in the server, where each Pareto optimal solution model corresponds to a wind turbine in the wind farm. It sets a data optimization cycle, establishes communication between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system, and uses the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0120] The present invention also provides a mobile terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, such as a dynamic optimization program for wind turbine yaw control.

[0121] When the processor executes the computer program, it implements the steps of the above-described dynamic optimization method for wind turbine yaw control, for example:

[0122] A linearized model of the wind turbine is established, and the linearized state-space expression of the turbine is obtained based on the structural parameters of the wind turbine. The wind speed, pitch angle and generator rated torque are taken as inputs, and the power and generator torque are taken as outputs.

[0123] The coefficient of the influence of wind direction deviation on power is fitted by a linearized model of the wind turbine, and the power output of the linearized model of the wind turbine is corrected.

[0124] The yaw control process and power output of the wind turbine under the given wind conditions are calculated by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0125] A multi-objective optimization problem is established to improve the equivalent power generation and limit the number of yaw operations. A minimum optimization problem is constructed, and Pareto optimality theory is used to seek the Pareto optimal solution model.

[0126] The server stores several Pareto optimal solution models, each corresponding to a wind turbine in the wind farm. A data optimization cycle is set, and communication is established between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system. This allows the server to collect historical data from the SCADA database, optimize the new thresholds and delay times calculated by the solution algorithm, and update them to the corresponding wind turbine's main control system, thus completing the dynamic optimization of the wind turbine's yaw control.

[0127] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above system, for example:

[0128] The model building module is used to build a linearized model of the wind turbine and obtain the linearized state-space expression of the unit based on the structural parameters of the wind turbine. The wind speed, pitch angle and generator rated torque are used as inputs, and the power and generator torque are used as outputs.

[0129] The model correction module is used to fit the coefficient of the influence of wind direction deviation on power through the linearization model of the wind turbine, and to correct the power output of the linearization model of the wind turbine.

[0130] The first data processing module is used to calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0131] The second data processing module is used to establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw executions. It constructs a minimum optimization problem, uses Pareto optimality theory to seek the Pareto optimal solution, and completes the dynamic optimization of wind turbine yaw control.

[0132] The third data processing module is used to store several Pareto optimal solution models in the server, where each Pareto optimal solution model corresponds to a wind turbine in the wind farm. It sets a data optimization cycle, establishes communication between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system, and uses the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0133] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the mobile terminal. For example, the computer program can be divided into a model building module, a model correction module, a first data processing module, a second data processing module, and a third data processing module;

[0134] The model building module is used to build a linearized model of the wind turbine and obtain the linearized state-space expression of the unit based on the structural parameters of the wind turbine. The wind speed, pitch angle and generator rated torque are used as inputs, and the power and generator torque are used as outputs.

[0135] The model correction module is used to fit the coefficient of the influence of wind direction deviation on power through the linearization model of the wind turbine, and to correct the power output of the linearization model of the wind turbine.

[0136] The first data processing module is used to calculate the yaw control process and power output of the unit under the wind condition input by using the yaw control algorithm to calculate the power output of the corrected linearized model of the wind turbine.

[0137] The second data processing module is used to establish a multi-objective optimization problem to improve the equivalent power generation and limit the number of yaw executions. It constructs a minimum optimization problem, uses Pareto optimality theory to seek the Pareto optimal solution, and completes the dynamic optimization of wind turbine yaw control.

[0138] The third data processing module is used to store several Pareto optimal solution models in the server, where each Pareto optimal solution model corresponds to a wind turbine in the wind farm. It sets a data optimization cycle, establishes communication between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system, and uses the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the main control system of the corresponding wind turbine, thus completing the dynamic optimization of the wind turbine's yaw control.

[0139] The mobile terminal can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. The mobile terminal may include, but is not limited to, a processor and a memory.

[0140] The processor can 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. A general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the mobile terminal, connecting all parts of the mobile terminal via various interfaces and lines.

[0141] The memory can be used to store the computer program and / or module. The processor implements various functions of the mobile terminal by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory.

[0142] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a given function (such as sound playback or image playback). The data storage area may store data created based on the use of the phone (such as audio data or a phonebook). Furthermore, the memory may include high-speed random access memory (RAM) and non-volatile memory, such as hard disks, RAM, plug-in hard disks, SmartMediaCards (SMC), Secure Digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0143] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the dynamic optimization method for yaw control of a wind turbine.

[0144] If the modules / units integrated in the mobile terminal are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0145] Based on this understanding, all or part of the processes in the above-described method can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the above-described dynamic optimization method for wind turbine yaw control. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or some intermediate form.

[0146] The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0147] It should be noted that the content contained in the computer-readable medium may be appropriately added to or subtracted from the content as required by the legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable medium may not include electrical carrier signals and telecommunication signals.

[0148] 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 dynamic optimization method for yaw control of a wind turbine generator, characterized in that, The process includes the following steps: Step 1, establishing a linearized model of the wind turbine and obtaining its linearized state-space expression based on the structural parameters, with wind speed, pitch angle, and generator rated torque as inputs, and power and generator torque as outputs; Step 2, fitting the coefficient of wind direction deviation on power using the linearized model of the wind turbine and correcting the power output of the linearized model; Step 3, using a yaw control algorithm to calculate the yaw control process and power output of the turbine under the given wind conditions; Step 4, establishing a multi-objective optimization problem to improve equivalent power generation while limiting a significant increase in the number of yaw executions, constructing a minimum optimization problem, and using Pareto optimality theory to seek Pareto optimality. Pareto optimal solution model; Step 5: Several Pareto optimal solution models are stored in the server, where each Pareto optimal solution model corresponds to a wind turbine in the wind farm. A data optimization cycle is set, and communication is established between the optimization solution server, the wind farm's SCADA database, and the wind turbine's main control system. This is used for the server to collect historical data from the SCADA database and update the new thresholds and delay times calculated by the optimization solution algorithm to the corresponding wind turbine's main control system, completing the dynamic optimization of the wind turbine's yaw control. The linearization model of the wind turbine fits the coefficient of wind direction deviation on power. The specific process is as follows: Let the wind energy captured by the wind turbine be P, and the yaw error angle of the wind turbine be θ. Then the formula for the wind energy captured by the turbine is as follows: Where: P is the wind energy captured by the unit, in W; ρ is the air density, in kg / m³. 3 R is the radius of the wind turbine rotor, in meters; V is the inflow wind speed, in meters per second; θ is the yaw error angle, in rads; n is an undetermined coefficient; where P is the energy loss generated during the conversion of wind energy captured by the wind turbine into active power. 损 Then the actual active power generated by the unit is: P 有功 = P - P 损 Where, when the yaw error is θ: in, Let n be the active power of the wind turbine when the yaw error is 0°. Based on the yaw error sequence and active power recorded in the SCADA system of the wind turbine, the least squares method or Fourier series approximation method is used to fit the curve to determine the value of the undetermined coefficient n. The formula for the minimum value optimization problem is as follows: in, This represents the mapping relationship between negative equivalent power generation and the number of yaw executions relative to the threshold and delay time; x represents the controller parameters, i.e., the threshold and delay time; X represents the value range of the threshold and delay time, with the threshold ranging from 5° to 20° and the delay time ranging from 20s to 210s, both taken as natural numbers. In the minimum optimization problem, the maximum number of yaw control executions is set as the boundary condition. A multi-objective optimization problem is established to improve equivalent power generation while limiting a significant increase in the number of yaw executions. Pareto optimality theory is used to seek the Pareto optimal solution. The specific process is as follows: Based on the solution corresponding to the initial yaw threshold and delay time, a genetic algorithm is used to optimize and calculate the yaw threshold and delay time corresponding to the optimal solution on the Pareto front, and compare it with the initial value. If the solution on the front is better than the initial solution, the middle point on the front is selected as the optimization solution for iterative optimization of the threshold and delay time. If the initial value is also on the front, no update is made, and one cycle of optimization is completed. Finally, all algorithms are encapsulated and packaged.

2. The dynamic optimization method for yaw control of a wind turbine generator according to claim 1, characterized in that, In step 1, the linearized state-space expression of the wind turbine is obtained based on the structural parameters of the wind turbine, as shown in the following formula: Where A is the system state coefficient matrix, B is the system control coefficient matrix, C is the output state coefficient matrix, and D is the output control coefficient matrix. x, u, and y are all vectors, where u is the input vector, including wind speed, rated torque, and blade pitch angle; y is the output vector, including power and generator speed; x and... These represent the current state of the system and the state at the next moment, respectively.

3. The dynamic optimization method for yaw control of a wind turbine generator according to claim 1, characterized in that, In step 3, a yaw control algorithm is written using Python or C++ to calculate the yaw control process and power output of the wind turbine under the input wind conditions, based on the power output of the corrected linearized model. The yaw control process is represented by the absolute azimuth of the wind turbine nacelle and the yaw flag. The yaw flag is a series of digital signals. When the yaw motor is not started, the yaw flag is set to 0. When the yaw motor performs a yaw control action, the yaw flag is set to 1. The number of times the yaw flag is triggered is counted to determine the number of yaw executions within that time period.

4. The dynamic optimization method for yaw control of a wind turbine generator according to claim 1, characterized in that, In step 5, the specific process of updating the new threshold and delay time calculated by the optimization algorithm to the main control system of the corresponding wind turbine is as follows: Set the data optimization cycle to x hours, collect the wind speed, wind direction, power and yaw execution process data within x hours every x hours, take the current threshold and delay time and wind condition information as input, start the optimization algorithm corresponding to the wind turbine in the server, solve the new optimal value and compare it with the current value. If it is better than the current value, iteratively write the new threshold and delay time into the main control system.

5. A dynamic optimization system for yaw control of a wind turbine generator, characterized in that, A dynamic optimization method for yaw control of a wind turbine generator according to any one of claims 1-4, comprising: a model building module for building a linearized model of the wind turbine generator and obtaining a linearized state-space expression of the generator based on the structural parameters of the wind turbine generator, with wind speed, pitch angle, and rated generator torque as inputs, and power and generator torque as outputs; a model correction module for fitting the coefficient of wind direction deviation on power through the linearized model of the wind turbine generator and correcting the power output of the linearized model of the wind turbine generator; a first data processing module for calculating the yaw control process and power output of the generator under the wind condition input using the yaw control algorithm on the power output of the corrected linearized model of the wind turbine generator; and a second data processing module for establishing a method to improve... The system addresses a multi-objective optimization problem that aims to achieve equivalent power generation while limiting the number of yaw operations. It constructs a minimum optimization problem and uses Pareto optimality theory to find the Pareto optimal solution, thus completing the dynamic optimization of wind turbine yaw control. The third data processing module stores several Pareto optimal solution models on the server, each corresponding to a wind turbine in the wind farm. It sets a data optimization cycle and establishes communication between the optimization server, the wind farm's SCADA database, and the wind turbine's main control system. This allows the server to collect historical data from the SCADA database and update the corresponding wind turbine's main control system with new thresholds and delay times calculated by the optimization algorithm, completing the dynamic optimization of wind turbine yaw control.

6. A mobile terminal, 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 dynamic optimization method for yaw control of a wind turbine as described in any one of claims 1 to 4.

7. 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 dynamic optimization method for yaw control of a wind turbine as described in any one of claims 1-4.

Images