Wind power optimization control system and method based on multi-target coupling enhancement

Abstract

The embodiment of the invention discloses a wind power optimization control system and method based on multi-target coupling strengthening, and relates to the technical field of wind driven generator control, and the system comprises a fan state generation module which is used for obtaining the current estimated state of a fan by using a state estimator according to the wind speed, wind direction and fan setting parameters at the current moment; the fan prediction module is used for predicting the wind speed, the fan response and the fan power output within the preset time based on the current prediction state and a wind speed prediction model; the control parameter generation module is used for searching an optimal control parameter input by using an optimization algorithm in a prediction time domain so as to minimize a target function; and the control parameter sending module is used for sending the control parameters generated by the control parameter generation module to the fan controller. The system can give consideration to multiple optimization objectives of the wind power plant, maximizes the power output of the wind power plant, reduces the wear of equipment as much as possible, reduces the operation and maintenance cost, and improves the overall operation benefit and reliability of the wind power plant.

Description

Technical Field

[0001] The present disclosure relates to the technical field of wind turbine control, and in particular to a wind power optimization control system and method based on multi-objective coupling enhancement. Background Art

[0002] Wind power generation is a technology that converts wind energy into electrical energy, and its core equipment is a wind turbine. The generator set is mainly composed of a wind rotor, a transmission system, a generator, a tower and a control system. The wind rotor is composed of multiple blades. Through aerodynamic design, the blades rotate under the action of wind, converting the kinetic energy of wind into rotational mechanical energy. The transmission system usually includes a speed increaser to increase the low-speed rotation of the blades to a high speed suitable for generator power generation. The generator converts mechanical energy into electrical energy through the principle of electromagnetic induction. The control system is responsible for monitoring wind speed, wind direction and the operating status of the generator to ensure that the wind turbine operates efficiently and safely under different wind conditions. With the advancement of technology, new technologies such as variable pitch control, doubly fed induction generators and direct-drive permanent magnet synchronous generators are constantly being applied, which significantly improves the efficiency and reliability of wind power generation and can better adapt to different wind energy resource conditions.

[0003] At present, wind turbines generally adopt a single-objective control strategy, usually with maximizing output power or minimizing load as the main control target. In the control strategy of maximizing output power, the control system adjusts the angle of the blades (pitch control) or the speed of the generator so that the wind turbine can output maximum power under different wind speed conditions. In the control strategy of minimizing load, the focus is on reducing the mechanical stress of the wind turbine during operation, thereby extending the life of the equipment.

[0004] The main disadvantage of single-objective control is its limitation and singleness. When the control system focuses on one goal, other important factors are often ignored. For example, in the pursuit of maximizing power output, it may cause the unit to bear greater mechanical loads, increase equipment wear and failure risks; conversely, when too much attention is paid to reducing loads, the available wind energy resources may not be fully utilized, resulting in reduced power generation efficiency.

[0005] In summary, how to perform multi-objective control on wind turbines and automatically adjust them before equipment wear and power generation efficiency is a problem that needs to be solved. Summary of the invention

[0006] In view of this, the embodiments of the present disclosure provide a wind power optimization control system and method based on multi-objective coupling enhancement to achieve multi-objective control of wind turbines.

[0007] In a first aspect, the embodiments of the present disclosure provide a wind power optimization control method based on multi-objective coupling enhancement, which adopts the following technical solutions:

[0008] A wind power optimization control system based on multi-objective coupling enhancement, comprising:

[0009] The wind turbine state generation module is used to obtain the current estimated state of the wind turbine according to the wind speed, wind direction and wind turbine setting parameters at the current moment, and use the state estimator;

[0010] The wind turbine prediction module predicts the wind speed, wind turbine response and wind turbine power output at a preset time based on the current estimated state and wind speed prediction model;

[0011] The control parameter generation module uses an optimization algorithm to find the optimal control parameter input to minimize the objective function and satisfy all objective constraints in the prediction time domain;

[0012] The control parameter sending module sends the control parameters generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameters to the fan.

[0013] As an optional implementation, the objective function J is specifically expressed as:

[0014]

[0015] Where θ represents the pitch angle, T g represents electromagnetic torque, w1, w2, w3 represent weight coefficients,

[0016] p out (t) represents the fan output power, C wear (t) represents the fan wear cost, C maint (t) represents the fan maintenance cost.

[0017] As an optional implementation, the fan wear cost C wear (t) is determined by the following formula:

[0018]

[0019] Among them, C replace represents the equipment replacement cost, L is the expected life span, and D(t) represents the degree of wear at time t.

[0020] As an optional embodiment, the wear degree D(t) at time t is determined by the following method:

[0021] Obtain wind turbine operating parameters and meteorological parameters at time t;

[0022] The fan operating parameters and meteorological parameters at time t are input into the wear prediction model to obtain the wear degree D(t) at time t.

[0023] As an optional implementation, the fan maintenance cost C maint (t) determined by:

[0024]

[0025] Among them, P i represents the probability of the i-th fault occurring within time t, C i represents the cost of repairing the i-th fault, and n represents the number of all possible faults.

[0026] As an optional implementation, it also includes: a weight adjustment module, which uses a reinforcement learning algorithm to adjust the weight value.

[0027] As an optional implementation, the optimization algorithm is a gradient descent method, and the change in the objective function value, the norm of the gradient, the maximum number of iterations, and the change in parameter update are set as convergence conditions.

[0028] In a second aspect, an embodiment of the present disclosure further provides a wind power optimization control method based on multi-objective coupling enhancement, based on the above-mentioned wind power optimization control system based on multi-objective coupling enhancement, including:

[0029] Using the wind turbine state generation module, according to the current wind speed, wind direction, and wind turbine setting parameters, and using the state estimator to obtain the current estimated state of the wind turbine;

[0030] Using the wind turbine prediction module, based on the current estimated state and wind speed prediction model, the wind speed, wind turbine response and wind turbine power output at a preset time are predicted;

[0031] Using the control parameter generation module, in the prediction time domain, an optimization algorithm is used to find the optimal control parameter input to minimize the objective function while satisfying all objective constraints;

[0032] The control parameter sending module is used to send the control parameter generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameter to the fan.

[0033] As an optional implementation, it also includes: using a weight adjustment module to adjust the weight value using a reinforcement learning algorithm.

[0034] As an optional implementation, the optimization algorithm is a gradient descent method, and the change in the objective function value, the norm of the gradient, the maximum number of iterations, and the change in parameter update are set as convergence conditions.

[0035] In summary, the technical effects of the wind power optimization control system based on multi-objective coupling enhancement provided by the present disclosure are:

[0036] The wind power optimization control system based on multi-objective coupling enhancement provided by the embodiment of the present disclosure can effectively take into account multiple optimization objectives of the wind farm. Through the combination of real-time prediction and strategy learning, it can maximize the power output of the wind farm while minimizing equipment wear and reducing operation and maintenance costs, thereby improving the overall operation efficiency and reliability of the wind farm.

[0037] On the other hand, the weight adjustment module dynamically adjusts the weights in the system through the reinforcement learning algorithm, which can achieve automatic optimization of multiple objectives. In practical applications, it can effectively control complex multi-objective systems. By applying this system, it can adapt to changes in the environment, automatically adjust weights to meet different needs, and improve the operating efficiency and effect of the overall system.

[0038] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the following preferred embodiments are specifically cited and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 A schematic diagram of the structure of a wind power optimization control system based on multi-objective coupling enhancement provided in an embodiment of the present disclosure;

[0041] Figure 2 A schematic flow chart of a wind power optimization control method based on multi-objective coupling enhancement provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0042] The embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

[0043] It should be clear that the following embodiments of the present disclosure are described by specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present disclosure from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. The present disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present disclosure. It should be noted that the following embodiments and features in the embodiments can be combined with each other in the absence of conflict. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present disclosure.

[0044] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present disclosure, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0045] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present disclosure. The drawings only show components related to the present disclosure rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0046] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0047] Reference Figure 1 The first aspect of the present invention provides a wind power optimization control system based on multi-objective coupling enhancement, comprising:

[0048] The wind turbine state generation module is used to obtain the current estimated state of the wind turbine according to the wind speed, wind direction and wind turbine setting parameters at the current moment, and use the state estimator;

[0049] For example, the state estimator can be implemented using a Kalman filter, and the current estimated state usually includes the following key aspects, which are critical to the operation and control of the wind turbine: speed, pitch angle, yaw angle, generator torque, wind speed, wind direction, bearing temperature, stress and vibration state, power output, aerodynamic load, voltage and current state, etc. Accurate estimation of these state parameters can help the wind turbine control system make more optimized and safe decisions, thereby improving the operating efficiency of the wind turbine, extending its service life, and ensuring reliable operation under various wind conditions.

[0050] These parameters are usually set by the control system. However, even though they are set actively, there are several reasons why state estimation is needed for these parameters: 1. The actual measured values ​​of speed, pitch angle, yaw angle, and generator torque are usually affected by noise and measurement errors. Even if these parameters are set actively, estimating the actual state through a state estimator (such as a Kalman filter) can reduce the impact of noise and obtain more accurate parameter values. 2. The wind turbine is a complex dynamic system, and actively set control inputs do not take effect immediately. For example, when you adjust the pitch angle, it may take some time for the actual pitch angle to reach the set value due to mechanical and aerodynamic lags. Therefore, state estimation can help predict how the actual state evolves over time, so as to more accurately understand the current wind turbine state. 3. The control system needs real-time and accurate feedback information to adjust the control strategy. State estimation can provide more accurate state information than relying solely on sensor measurement data, allowing the control system to better optimize the control input, thereby achieving better performance and stability. In summary, even if the speed, pitch angle, yaw angle and generator torque are set actively, state estimation is still essential because it can provide an accurate understanding of the actual state and help the control system make better decisions. The goal of state estimation is to enable the control system to operate more efficiently and safely in a real complex environment.

[0051] The wind turbine prediction module predicts the wind speed, wind turbine response and wind turbine power output at a preset time based on the current estimated state and wind speed prediction model;

[0052] The wind turbine prediction module is mainly implemented through the wind speed prediction model and the wind turbine dynamics model. The specific implementation method is as follows:

[0053] First, a wind speed prediction model is used to predict the wind speed and direction for a period of time in the future. Usually, time series analysis, machine learning models, or physical models are used to predict wind speed. These models take into account the current wind speed, historical wind speed data, and possible weather forecast information.

[0054] By taking the current estimated state as the initial condition and combining it with the wind speed forecast, the wind turbine dynamic model can be used to predict the state and output of the wind turbine in the future. This includes parameters such as wind turbine speed, pitch angle, and power generation at multiple time points in the future.

[0055] The dynamic model of a wind turbine describes how the wind turbine responds to control inputs (such as pitch angle, speed, etc.) and changes in wind speed. This model can be a physics-based model, such as a controlled equation or a numerical simulation model, or a data-driven model, such as a linear or nonlinear regression model.

[0056] In the prediction process, various physical and operational constraints also need to be considered, such as the physical limit of the pitch angle, the safe range of the rotation speed, the upper limit of the power output, etc. These constraints will affect future dynamic predictions.

[0057] The control parameter generation module uses an optimization algorithm to find the optimal control parameter input to minimize the objective function and satisfy all objective constraints in the prediction time domain;

[0058] The objective function J is specifically expressed as:

[0059]

[0060] Where θ represents the pitch angle, T g represents electromagnetic torque, w1, w2, w3 represent weight coefficients,

[0061] p out (t) represents the fan output power, C wear (t) represents the fan wear cost, C maint (t) represents the fan maintenance cost.

[0062] Specifically, the fan wear cost C wear (t) is determined by the following formula:

[0063]

[0064] Among them, C replace represents the equipment replacement cost, L is the expected life span, and D(t) represents the degree of wear at time t.

[0065] The degree of wear D(t) at time t is determined by the following method:

[0066] Obtain wind turbine operating parameters and meteorological parameters at time t;

[0067] The fan operating parameters and meteorological parameters at time t are input into the wear prediction model to obtain the wear degree D(t) at time t. The wear prediction model generally adopts a neural network model, which is trained with historical data to obtain training data, which generally includes fan operating parameters, meteorological parameters and fan wear conditions.

[0068] Generally speaking, the maintenance cost of a fan is C maint (t) determined by:

[0069]

[0070] Among them, P i represents the probability of the i-th fault occurring within time t, C i represents the cost of repairing the i-th fault, and n represents the number of all possible faults.

[0071] Specifically, P i It is determined by the fault probability determination model. The neural network model is trained through historical fault data, meteorological data when the fault occurs, and fan operation data. Each probability corresponds to a model. When in use, the current meteorological data and fan operation data are input into the fault probability determination model to obtain the probability of each fault respectively.

[0072] The optimization algorithm is a gradient descent method, which sets the objective function value change, the norm of the gradient, the maximum number of iterations, and the parameter update change as convergence conditions. Specifically, the following steps are used for optimization:

[0073] When using the gradient descent method to optimize the control of the fan, we need to clarify the specific optimization goal, select the appropriate learning rate, and determine the gradient calculation method. The following is a specific solution:

[0074] 1. Define the objective function J

[0075] First, define the objective function J, which can comprehensively consider power output, equipment wear cost and maintenance cost. The definition of the objective function is the same as above and will not be repeated here.

[0076] 2.1 Initialization

[0077] Initialize control parameters θ and T g The initial values ​​of θ0 and T g0 .

[0078] Set the learning rate α (i.e. step size).

[0079] Determine the maximum number of iterations N and the convergence condition (such as the change in the objective function J is less than a certain threshold).

[0080] 2.2 Calculate the gradient of the objective function

[0081] In the gradient descent method, we need to calculate the objective function J for the control parameters θ and T g The partial derivative of :

[0082]

[0083] These derivatives can be computed either by numerical differentiation (finite differences) or analytical methods.

[0084] 2.3 Update control parameters

[0085] In each iteration, the control parameters θ and T are updated g :

[0086]

[0087] Where k represents the kth iteration.

[0088] 2.4 Checking the convergence conditions

[0089] After each iteration, check the change of the objective function J. If the change of J is less than the set threshold or the maximum number of iterations is reached, the iteration is stopped.

[0090] 3. Specific implementation plan

[0091] 3.1 Initial Setup:

[0092] Set the initial pitch angle and electromagnetic torque to θ0 and T g0 .

[0093] Choose a learning rate α, for example, a value between 0.01 and 0.1.

[0094] Set the maximum number of iterations, such as 1000, and set the convergence threshold, such as 10 -6 .

[0095] 3.2 Run the gradient descent method:

[0096] Read the current wind speed, wind rotor speed, generator speed and other data, and calculate the current objective function value J(θ k ,T g,k ).

[0097] Calculate the objective function for θ and T g gradient.

[0098] Update θ and T g The value of .

[0099] Repeat the above steps until the convergence condition is met.

[0100] 3.3 Feedback and Adjustment:

[0101] If you find that the learning rate α is too large and causes no convergence, or the convergence speed is too slow, you can adjust the learning rate appropriately.

[0102] Monitor the operating status of the wind turbine in real time and adjust the weight coefficients (w1, w2, w3) of the objective function according to the operating conditions to better balance the output power and maintenance cost.

[0103] 3.4 Application of the objective function during operation

[0104] In actual operation of the wind turbine, the gradient descent method is applied once every certain period of time (such as every second or every minute) to adjust the pitch angle and electromagnetic torque. By continuously adjusting these control parameters, it is possible to ensure that the wind turbine maintains the best operating state under different wind conditions, maximizing the output power while minimizing equipment wear and maintenance costs.

[0105] In actual use, the convergence conditions can be determined according to the following methods: 1. The change in the objective function value, 2. The norm of the gradient, 3. The maximum number of iterations, 4. The change in parameter update, etc.

[0106] The control parameter generation module uses the gradient descent method to optimize the pitch angle and electromagnetic torque of the wind turbine in real time, with the goal of minimizing the overall cost and maximizing the power output. In practical applications, flexible adjustment of the learning rate and weight coefficient, and sufficient simulation verification will help achieve efficient operation of the wind turbine.

[0107] The control parameter sending module sends the control parameters generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameters to the fan.

[0108] This system can effectively take into account the multiple optimization objectives of the wind farm. By combining real-time prediction with strategy learning, it can maximize the power output of the wind farm while minimizing equipment wear and reducing operation and maintenance costs, thereby improving the overall operation efficiency and reliability of the wind farm.

[0109] As an optional implementation, the system further includes: a weight adjustment module, which uses a reinforcement learning algorithm to adjust the weight value. The reinforcement learning algorithm is obtained by training with a Q-learning algorithm.

[0110] Specifically, the complete method of automatically adjusting weights using reinforcement learning algorithms is as follows, which combines problem modeling, reinforcement learning algorithm selection, Q-value updating, weight adjustment strategy, and continuous optimization steps.

[0111] 1. Problem Modeling

[0112] 1.1 State Space

[0113] Define the state space S, each state s includes the following parameters:

[0114] The current weight combination (w1, w2, w3).

[0115] Current power output P out .

[0116] Current equipment wear C wear .

[0117] Current maintenance cost C maint .

[0118] 1.2 Action Space

[0119] Define the action space A. Each action a is a small adjustment to the weights (w1, w2, w3), including:

[0120] Increase or decrease w1;

[0121] Increase or decrease w2;

[0122] Increase or decrease w3.

[0123] 1.3 Reward Function

[0124] Define the reward function R(s,a) to measure the policy performance under the current weight combination. A possible reward function form is:

[0125] R(s,a)=P out (s,a)-λ1C wear (s,a)-λ2C maint (s,a)

[0126] Among them, λ1 and λ2 are constant coefficients used to balance the relative importance of different objectives.

[0127] 2. Environment construction and data generation

[0128] 2.1 Simulation Environment

[0129] Build a simulation environment that can return the new state s and the corresponding reward R(s,a) based on the input weights (w1, w2, w3) and action a.

[0130] 2.2 Data Generation

[0131] Run the initial policy in the simulation environment to collect enough state-action-reward triplets as training data for the reinforcement learning model.

[0132] 3. Reinforcement Learning Algorithm Selection

[0133] Choose a suitable reinforcement learning algorithm to learn the optimal weight adjustment strategy:

[0134] Q-learning: suitable for small-scale discrete state space and action space.

[0135] Deep Q Network (DQN): Suitable for continuous state space or action space and can handle high-dimensional input.

[0136] Policy Gradient Methods: such as PPO, A3C, etc., suitable for complex strategies and continuous action spaces.

[0137] Here we choose Q-learning network.

[0138] 4. Q value update and weight adjustment

[0139] 4.1Q value update

[0140] For each state-action pair, update the Q value according to the following rule:

[0141] Q(s,a)←Q(s,a)+a[R(s,a)+γmaxQ(s',a')-Q(s,a)]

[0142] Where a is the learning rate, γ is the discount factor, and s' is the next state after executing action a.

[0143] 4.2 Weight Adjustment

[0144] According to the learned Q value, choose the best action a to adjust the weight:

[0145] The rules for selecting actions are:

[0146]

[0147] After executing the action, adjust the corresponding weight. For example, if action a1 is "increase w1", then:

[0148] w1←w1+Δw

[0149] Here, Δw is a preset adjustment step (eg, 0.01).

[0150] The weight w2 or w3 can be increased or decreased accordingly depending on the action type.

[0151] 5. Continuous optimization and training

[0152] 5.1 Initial Training

[0153] In a simulation environment, the reinforcement learning model is trained through multiple iterations to learn the optimal weight adjustment strategy.

[0154] 5.2 Real-time Adjustment

[0155] In actual operation, the trained model is used for real-time weight adjustment. The model automatically selects the optimal weight adjustment strategy based on the current system status.

[0156] 5.3 Balance between exploration and exploitation

[0157] Use the ∈-greedy strategy: choose the best action in most cases, and randomly choose actions in a few cases to avoid local optimal solutions.

[0158] 6. Evaluation and Improvement

[0159] 6.1 Regularly evaluate model performance

[0160] Regularly evaluate the effectiveness of weight adjustment strategies and collect new data for analysis and improvement.

[0161] 6.2 Adjusting learning parameters

[0162] Adjust the parameters of the reinforcement learning algorithm according to the actual performance, such as learning rate a, discount factor γ, step size Δw, etc.

[0163] 6.3 Online Learning and Transfer Learning

[0164] Use online learning techniques to continuously optimize model performance based on new data or transfer learning experience between different operating environments.

[0165] The weight adjustment module dynamically adjusts the weights in the system through the reinforcement learning algorithm, which can achieve automatic optimization of multiple objectives. Through reasonable modeling, strategy selection and continuous optimization, effective control of complex multi-objective systems can be achieved in practical applications. By applying this system, it can adapt to changes in the environment, automatically adjust weights to meet different needs, and improve the operating efficiency and effectiveness of the overall system.

[0166] On the other hand, refer to Figure 2 The present invention provides a wind power optimization control method based on multi-objective coupling enhancement, based on the wind power optimization control system based on multi-objective coupling enhancement as mentioned above, comprising:

[0167] Step S1, using a wind turbine state generation module, according to the current wind speed, wind direction, and wind turbine setting parameters, and using a state estimator to obtain the current estimated state of the wind turbine;

[0168] Step S2, using the wind turbine prediction module to predict the wind speed, wind turbine response and wind turbine power output at a preset time based on the current estimated state and the wind speed prediction model;

[0169] Step S3, using the control parameter generation module, in the prediction time domain, using an optimization algorithm to find the optimal control parameter input to minimize the objective function while satisfying all objective constraints;

[0170] Step S4: using the control parameter sending module to send the control parameters generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameters to the fan.

[0171] This method can effectively take into account the multiple optimization objectives of the wind farm. By combining real-time prediction with strategy learning, it can maximize the power output of the wind farm while minimizing equipment wear and reducing operation and maintenance costs, thereby improving the overall operation efficiency and reliability of the wind farm.

[0172] As an optional implementation, the method further includes: using a weight adjustment module to adjust the weight value using a reinforcement learning algorithm.

[0173] As an optional implementation, the optimization algorithm is a gradient descent method, and the change in the objective function value, the norm of the gradient, the maximum number of iterations, and the change in parameter update are set as convergence conditions.

[0174] For detailed description of this embodiment, reference may be made to the corresponding descriptions in the aforementioned embodiments, which will not be repeated here.

[0175] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, and are not limitations. The above details do not limit the present disclosure to the necessity of adopting the above specific details to be implemented.

[0176] In the present disclosure, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. The block diagrams of the devices, devices, equipment, and systems involved in the present disclosure are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, devices, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0177] Additionally, as used herein, "or" used in a list of items beginning with "at least one" indicates a separate list, so that, for example, a list of "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not mean that the example described is preferred or better than other examples.

[0178] It should also be noted that in the system and method of the present disclosure, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present disclosure.

[0179] Various changes, substitutions, and modifications of the techniques described herein may be made without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of the present disclosure is not limited to the specific aspects of the processes, machines, manufactures, compositions of events, means, methods, and actions described above. Currently existing or later to be developed processes, machines, manufactures, compositions of events, means, methods, or actions that perform substantially the same functions or achieve substantially the same results as the corresponding aspects described herein may be utilized. Thus, the appended claims include such processes, machines, manufactures, compositions of events, means, methods, or actions within their scope.

[0180] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Therefore, the present disclosure is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0181] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present disclosure to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A wind power optimization control system based on multi-objective coupling enhancement, characterized in that: include: The wind turbine state generation module is used to obtain the current estimated state of the wind turbine according to the wind speed, wind direction and wind turbine setting parameters at the current moment, and use the state estimator; The wind turbine prediction module predicts the wind speed, wind turbine response and wind turbine power output at a preset time based on the current estimated state and wind speed prediction model; The control parameter generation module uses an optimization algorithm to find the optimal control parameter input to minimize the objective function and satisfy all objective constraints in the prediction time domain; The control parameter sending module sends the control parameters generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameters to the fan.

2. The wind power optimization control system based on multi-objective coupling enhancement according to claim 1 is characterized in that: The objective function J is specifically expressed as: Where θ represents the pitch angle, T g represents electromagnetic torque, w1, w2, w3 represent weight coefficients, p out (t) represents the fan output power, C wear (t) represents the fan wear cost, C maint (t) represents the fan maintenance cost.

3. The wind power optimization control system based on multi-objective coupling enhancement according to claim 2 is characterized in that: Fan wear cost C wear (t) is determined by the following formula: Among them, C replace represents the equipment replacement cost, L is the expected life span, and D(t) represents the degree of wear at time t.

4. The wind power optimization control system based on multi-objective coupling enhancement according to claim 3 is characterized in that: The degree of wear D(t) at time t is determined by the following method: Obtain wind turbine operating parameters and meteorological parameters at time t; The fan operating parameters and meteorological parameters at time t are input into the wear prediction model to obtain the wear degree D(t) at time t.

5. The wind power optimization control system based on multi-objective coupling enhancement according to claim 4 is characterized in that: Fan maintenance cost C maint (t) determined by: Among them, P i represents the probability of the i-th fault occurring within time t, C i represents the cost of repairing the i-th fault, and n represents the number of all possible faults.

6. The wind power optimization control system based on multi-objective coupling enhancement according to claim 5 is characterized in that: Also includes: The weight adjustment module uses the reinforcement learning algorithm to adjust the weight value.

7. The wind power optimization control system based on multi-objective coupling enhancement according to claim 6 is characterized in that: The optimization algorithm is a gradient descent method, which sets the change in the objective function value, the norm of the gradient, the maximum number of iterations, and the change in parameter update as convergence conditions.

8. A wind power optimization control method based on multi-objective coupling enhancement, based on the wind power optimization control system based on multi-objective coupling enhancement as claimed in claim 1, characterized in that: include: Using the wind turbine state generation module, according to the current wind speed, wind direction, and wind turbine setting parameters, and using the state estimator to obtain the current estimated state of the wind turbine; Using the wind turbine prediction module, based on the current estimated state and wind speed prediction model, the wind speed, wind turbine response and wind turbine power output at a preset time are predicted; Using the control parameter generation module, in the prediction time domain, an optimization algorithm is used to find the optimal control parameter input to minimize the objective function while satisfying all objective constraints; The control parameter sending module is used to send the control parameter generated by the control parameter generating module to the fan controller, so that the fan controller applies the control parameter to the fan.

9. The wind power optimization control method based on multi-objective coupling enhancement according to claim 8 is characterized in that: Also includes: The weight adjustment module uses the reinforcement learning algorithm to adjust the weight value.

10. The wind power optimization control method based on multi-objective coupling enhancement according to claim 8, characterized in that: The optimization algorithm is a gradient descent method, which sets the change in the objective function value, the norm of the gradient, the maximum number of iterations, and the change in parameter update as convergence conditions.

Images