A Wind Farm Dynamic Optimization Method and System Based on Adaptive Regulation of Real-Time Data

Abstract

The present invention discloses a method and system for dynamic optimization of a wind farm based on real-time data adaptive regulation, which relates to the technical field of wind farm optimization. The method includes: acquiring wind farm environmental data and wind turbine status data, mapping them to a state space model, and aligning them within the same time step; constructing a wind farm dynamic optimization objective, calculating the uncertainty degree, dynamic importance degree, and target collaborative contribution rate, constructing a fuzzy analytic hierarchy process matrix, solving to obtain the weight of the wind farm dynamic optimization objective, and constructing a dynamic optimization objective function; constructing wind farm constraints, solving the dynamic optimization objective function to generate an optimal control parameter set, inputting it into a safety barrier function for verification, and after passing the verification, performing wind farm dynamic optimization. By constructing a wind farm dynamic optimization objective, power coordination optimization within the entire field is achieved, and wake damage is reduced; by introducing a safety barrier function and a fault tolerance control mechanism, the overall optimization efficiency and reliability of the wind farm are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind farm optimization, and specifically to a wind farm dynamic optimization method and system based on real-time data adaptive control. Background Art

[0002] As an important renewable energy source, the development of wind power is limited by the uncertainty and complexity of wind farm operation. The operating efficiency of wind turbines depends not only on environmental factors such as wind speed and wind direction, but also on the state of the turbines themselves. To achieve the overall optimal power generation efficiency of a wind farm, dynamic optimization control needs to be carried out for the wind turbine group. However, the current wind farm optimization control methods mainly have the following problems: Lack of coordination in single-machine optimization. Traditional wind turbine control mainly relies on single-machine control strategies, that is, each single wind turbine is optimized and adjusted based on its own sensor data, such as adjusting the pitch angle and yaw angle to maximize power output. However, this method ignores the interaction between wind turbines, especially the wake effect of wind turbines, which may lead to a decrease in the power generation efficiency of downstream wind turbines and even exacerbate the fatigue loss of wind turbine components. Uncertainty of wind farm operating conditions. Environmental factors such as wind speed, wind direction, and turbulence intensity have significant time-varying characteristics, and the state variables of wind turbines (such as bearing loads and drivetrain torques) also change with the operating conditions. Current optimization methods often rely on static models and are difficult to adapt to the changes in wind farm operating conditions in real time, resulting in the optimization scheme being difficult to be effective in the long term. Difficulty in balancing optimization objectives. The optimization of a wind farm is not only about increasing power output, but also involves reducing the fatigue load of wind turbines and reducing wake interference. However, there are conflicts between different objectives. For example, increasing power output may exacerbate the load on wind turbine blades and affect the lifespan of wind turbines; excessive reduction of the wake effect may affect the power generation efficiency of wind turbines. Current optimization methods are difficult to effectively balance these optimization objectives, resulting in unstable or difficult-to-implement optimization results. Summary of the Invention

[0003] Based on the above-mentioned disadvantages of the prior art, the object of the present invention is to provide a wind farm dynamic optimization method and system based on real-time data adaptive control to solve the above technical problems.

[0004] To achieve the above object, the present invention provides the following technical solution: A wind farm dynamic optimization method based on real-time data adaptive control, including:

[0005] Obtain wind farm environmental data and wind turbine status data, map the data to a state space model, and align them within the same time step;

[0006] Construct the dynamic optimization objective of the wind farm, calculate the uncertainty degree, dynamic importance degree and target collaborative contribution rate based on the dynamic optimization objective of the wind farm, construct a fuzzy analytic hierarchy process matrix according to the target collaborative contribution rate, solve to obtain the weight of the dynamic optimization objective of the wind farm, and construct a dynamic optimization objective function according to the dynamic optimization objective of the wind farm and the weight;

[0007] Construct the wind farm constraints, and solve the dynamic optimization objective function under the condition of meeting the wind farm constraint conditions to generate an optimal control parameter set;

[0008] Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set to perform dynamic optimization of the wind farm.

[0009] The present invention is further configured that the wind farm environment data includes wind speed, wind direction, wind shear index, turbulence intensity, wake wind speed and wake diffusion coefficient; the wind turbine state data includes wind turbine power output, rotational speed, tip speed ratio, pitch angle, yaw angle, bearing load and drive train torque.

[0010] The present invention is further configured to map the data to a state space model and align it within the same time step, including:

[0011] Construct an initial state vector and covariance matrix according to the wind farm environment data and wind turbine state data;

[0012] Perform state prediction according to the initial state vector and output a predicted state vector;

[0013] Calculate an adaptive state correction coefficient according to the covariance matrix;

[0014] Update the data according to the predicted state vector and the adaptive state correction coefficient to obtain the wind farm environment data and wind turbine state data aligned within the same time step.

[0015] The present invention is further configured that the calculation logic of the predicted state vector is: , is the predicted state vector of the data at time, is the state transition matrix, is the data at time prediction update vector, is the control input matrix, is the control input vector;

[0016] The calculation logic of the adaptive state correction coefficient is: , is the adaptive state correction coefficient, is the predicted error covariance matrix, is the observation matrix, is the transpose of the observation matrix, is the measurement noise covariance matrix;

[0017] The calculation logic for updating the data is: , is the data At time, the predicted update vector, is the data At time, the measurement value.

[0018] The present invention is further configured such that the wind farm dynamic optimization objectives include maximizing the wind turbine power, minimizing the fatigue load, and wake interference;

[0019] Converting the maximization of the wind turbine power into the minimization of the negative wind turbine power, the calculation logic for maximizing the wind turbine power is: , is the number of wind turbines, is the wind turbine power mapping function, , is the air density, is the swept area of the wind turbine, is the power coefficient, is the tip speed ratio, is the pitch angle, is the wind speed;

[0020] The calculation logic for minimizing the fatigue load is: , is the fatigue load mapping function, , and are empirical parameters, is the wind turbine acceleration, is the torque fluctuation of the drive train;

[0021] The calculation logic for minimizing the wake interference is: , is the wake effect mapping function, , is the induction factor, is the wake diffusion coefficient, is the wind turbine spacing, is the wind turbine diameter.

[0022] The present invention is further configured such that the calculation logic for the uncertainty degree is: , is the uncertainty degree of the th objective function, is the number of data groups, is the probability density, , is the value of the th objective function for the th set of data;

[0023] The calculation logic of the dynamic importance is: , is the dynamic importance of the th objective function;

[0024] The calculation logic of the target collaborative contribution rate is: , is the th and th target collaborative contribution rates of the objective functions, and are the values of the th and th objective functions, and are the preset thresholds.

[0025] The present invention is further configured to construct a fuzzy analytic hierarchy process matrix according to the target collaborative contribution rate , , and according to the fuzzy analytic hierarchy process matrix , solve the eigenvector corresponding to its maximum eigenvalue : , is the maximum eigenvalue of the fuzzy analytic hierarchy process matrix, is the final target weight vector, , , and are the weights of the objective functions, and after normalization, construct a dynamic optimization objective function: , is the dynamic optimization objective function.

[0026] The present invention is further configured to include control input constraints and wind turbine state variable constraints in the wind field constraints. The control input constraints include pitch angle control constraints and yaw angle control constraints. The wind turbine state variable constraints include wind turbine speed constraints, fatigue load constraints, and wake interference constraints. By controlling the pitch angle and yaw angle within the constraints, solve the dynamic optimization objective function to obtain an optimal control parameter set.

[0027] The present invention is further configured to have the construction logic of the safety barrier function as: , is the safety barrier function, is the maximum eigenvalue of the wind turbine, is the influence of the control input on the wind turbine state. When When verified, execute the optimal control parameter set to perform wind farm dynamic optimization;

[0028] When happens, enter the fault-tolerant control mechanism for adjustment, , is the adjusted control parameter set, is the current control parameter set, is the optimal control parameter set, is the constraint condition. In the feasible region, use the projected gradient descent method to solve the adjusted control parameter set, and input the adjusted control parameter set into the safety barrier function for verification. If the verification passes, execute.

[0029] The present invention also provides a wind farm dynamic optimization system based on real-time data adaptive regulation for implementing the above-mentioned wind farm dynamic optimization method based on real-time data adaptive regulation. The system includes:

[0030] Data acquisition module: Acquire wind farm environment data and wind turbine status data, map the data to the state space model, and align them within the same time step;

[0031] Objective construction module: Construct the wind farm dynamic optimization objective, calculate the uncertainty degree, dynamic importance degree, and objective collaborative contribution rate based on the wind farm dynamic optimization objective, construct a fuzzy analytic hierarchy process matrix according to the objective collaborative contribution rate, solve to obtain the weight of the wind farm dynamic optimization objective, and construct a dynamic optimization objective function based on the wind farm dynamic optimization objective and the weight;

[0032] Optimal generation module: Construct wind farm constraints, solve the dynamic optimization objective function under the condition of satisfying the wind farm constraint conditions, and generate an optimal control parameter set;

[0033] Safety verification module: Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set to perform wind farm dynamic optimization.

[0034] The present invention provides a wind farm dynamic optimization method and system based on real-time data adaptive regulation. The method includes acquiring wind farm environment data and wind turbine status data, mapping the data to the state space model, and aligning them within the same time step; constructing the wind farm dynamic optimization objective, calculating the uncertainty degree, dynamic importance degree, and objective collaborative contribution rate based on the wind farm dynamic optimization objective, constructing a fuzzy analytic hierarchy process matrix according to the objective collaborative contribution rate, solving to obtain the weight of the wind farm dynamic optimization objective, and constructing a dynamic optimization objective function based on the wind farm dynamic optimization objective and the weight; constructing wind farm constraints, solving the dynamic optimization objective function under the condition of satisfying the wind farm constraint conditions, and generating an optimal control parameter set; inputting the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set to perform wind farm dynamic optimization. The beneficial effects produced include:

[0035] 1. Improve the overall power generation efficiency of the wind farm: By constructing a dynamic optimization objective for the wind farm, comprehensively considering the fan power, fatigue load, and wake effect, ensure that the fan group generates electricity under the optimal operating conditions; adopt state-space modeling and real-time data adaptive alignment methods to enable the fan control to accurately match the real-time changes in the wind farm environment, thereby enhancing the power output; through the optimization objective function that maximizes the fan power, avoid the local optimum problem caused by the traditional single-machine optimization method, and achieve coordinated power optimization within the entire field.

[0036] 2. Solve the wake effect and improve the power generation capacity of downstream fans: The wake effect between fans will affect the wind energy utilization efficiency of downstream fans. The present invention reduces wake damage and improves the power generation capacity of downstream fans during the collaborative optimization of the fan group through the optimization objective function that minimizes wake interference; adopts a wake diffusion model, considers the wind speed attenuation and fan spacing during the optimization process, reasonably adjusts the control parameters of the fans, makes the wind energy utilization more balanced, and enhances the overall efficiency of the wind farm.

[0037] 3. Intelligent collaborative calculation for objective optimization to improve the optimization efficiency: By calculating the objective collaborative contribution rate and dynamically adjusting the optimization objective weights, ensure that the optimization direction automatically adjusts with the environmental changes, and improve the calculation efficiency; adopt an uncertainty calculation method to evaluate the importance of each optimization objective under different wind farm environments, avoid the optimization failure problem caused by fixed weights, and improve the intelligent level of the optimization strategy.

[0038] 4. Introduce a safety barrier function to improve the system stability and security: While optimizing the control parameters, introduce a safety barrier function to ensure that the optimization scheme meets the safe operating conditions of the fans, and avoid fan failures or abnormalities caused by optimization mistakes; adopt a fault-tolerant control mechanism. When the optimization scheme does not meet the safety requirements, automatically adjust the control input to make it adjust within the feasible region, and improve the reliability of the fan operation.

[0039] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically gives the specific implementation manners of this application. Brief Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings:

[0041] Figure 1 A flowchart of a wind farm dynamic optimization method based on real-time data adaptive regulation shown in an exemplary embodiment of the present invention;

[0042] Figure 2 A structural schematic diagram of a wind farm dynamic optimization system based on real-time data adaptive regulation shown in an exemplary embodiment of the present invention. Detailed implementation manners

[0043] The following will describe the implementation manners of the present invention with reference to the drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for explaining the present invention, rather than for limiting the protection scope of the present invention.

[0044] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, number, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0045] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0046] Embodiment 1

[0047] A wind farm dynamic optimization method based on real-time data adaptive regulation, as Figure 1 shown, includes:

[0048] Obtain wind farm environmental data and wind turbine status data, map the data to a state space model, and align them within the same time step;

[0049] Construct a wind farm dynamic optimization target, calculate the uncertainty degree, dynamic importance degree, and target collaborative contribution rate based on the wind farm dynamic optimization target, construct a fuzzy analytic hierarchy process matrix according to the target collaborative contribution rate, solve to obtain the weight of the wind farm dynamic optimization target, and construct a dynamic optimization objective function according to the wind farm dynamic optimization target and the weight;

[0050] Construct a wind farm constraint, solve the dynamic optimization objective function under the condition of meeting the wind farm constraint conditions, and generate an optimal control parameter set;

[0051] Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set for wind farm dynamic optimization.

[0052] The present invention is further configured such that the wind farm environmental data includes wind speed, wind direction, wind shear index, turbulence intensity, wake wind speed, and wake diffusion coefficient; the wind turbine status data includes wind turbine power output, rotational speed, tip speed ratio, pitch angle, yaw angle, bearing load, and drivetrain torque. Specifically, the wind speed refers to the air flow velocity within the wind farm. The wind speed directly affects the power output of the wind turbine. The higher the wind speed, the greater the wind energy, but too high a wind speed may cause the wind turbine to enter the protection mode; the wind direction refers to the direction of wind flow, and the wind direction affects the yaw control of the wind turbine to ensure that the wind turbine always faces the best wind energy direction; the wind shear index describes the rate of change of wind speed with height and is represented by the logarithmic wind speed distribution formula: , is the wind speed at height , is the reference height at the wind speed is the wind shear index, and the value range is [0.1, 0.3]; the turbulence intensity measures the fluctuation degree of the wind speed, , is the standard deviation of the wind speed, is the average wind speed; the wake wind speed refers to the wind speed behind the wind turbine blades. Since the wind turbine extracts energy from the wind, the wind speed in the wake area is usually lower than the incoming flow wind speed, , is the induction factor, is the wake diffusion coefficient, is the wind turbine spacing, is the wind turbine diameter; the wake diffusion coefficient represents the diffusion degree of the wake in the wind farm and is related to the terrain, wind speed, and turbulence intensity. The wake diffusion trend is calculated using historical data; the wind turbine power output refers to the actual output power of the wind turbine at a certain moment; the rotational speed refers to the rotational speed of the wind turbine impeller; the tip speed ratio refers to the ratio of the blade tip speed to the incoming flow wind speed; the pitch angle refers to the rotation angle of the wind turbine blade; the yaw angle refers to the deflection angle of the wind turbine nacelle relative to the wind direction; the bearing load refers to the torque borne by the main bearing of the wind turbine; the drivetrain torque refers to the torque borne by the drivetrain.

[0053] The present invention is further configured to map the data to a state space model and align it within the same time step, including:

[0054] Construct an initial state vector and a covariance matrix based on wind farm environmental data and wind turbine status data; specifically, the state vector is used to describe the current key operating parameters of the wind turbine and is jointly composed of wind farm environmental data and wind turbine status data; the covariance matrix is used to describe the uncertainty and correlation between state variables. The size of the covariance matrix determines the sensitivity of the system to measurement errors; in the covariance matrix, the diagonal elements represent the variances of the state variables, measuring the uncertainty of the variables, and the non-diagonal elements represent the covariances between the variables, reflecting the correlation between the variables;

[0055] Perform state prediction based on the initial state vector and output the predicted state vector; the calculation logic of the predicted state vector is: , is the data at time of the predicted state vector, is the state transition matrix, is the data at time of the predicted update vector, is the control input matrix, is the control input vector; specifically, based on the initial state vector, use the state transition matrix to predict the state of the wind turbine to obtain the predicted state vector, thereby providing an estimate of the future state of the wind turbine. This process allows the wind farm optimization control system to take measures in advance to improve the wind energy utilization efficiency, reduce wake interference, and optimize the wind turbine fatigue load; starting from the initial state vector , obtain the current wind speed, power, rotational speed and other state information of the wind turbine; based on the state transition matrix , calculate the predicted state of the wind turbine at the next moment, consider the control input , and combine the control input matrix to adjust the predicted state. Finally, obtain the predicted update vector , providing input for subsequent optimization calculations;

[0056] Calculate the adaptive state correction coefficient according to the covariance matrix; the calculation logic of the adaptive state correction coefficient is: , is the adaptive state correction coefficient, is the predicted error covariance matrix, is the observation matrix, is the transpose of the observation matrix, is the measurement noise covariance matrix; specifically, in the process of wind farm optimization control, the measurement data is often affected by sensor noise, environmental disturbances and wind farm uncertainties. In order to ensure more accurate state estimation, it is necessary to calculate the adaptive state correction coefficient , to dynamically balance the credibility of the predicted state and the measured state, thereby obtaining a more robust state estimate, and calculate the predicted error covariance matrix , estimate the error range of the system state through the predicted state equation, and measure the uncertainty of the system on different state variables. Calculate the observation matrix and the measurement noise covariance matrix , the observation matrix is used to describe the mapping relationship between the measurement variables and the system state variables, and the measurement noise covariance matrix reflects the error level of the sensor measurement. Calculate the adaptive state correction coefficient , use for state correction, balance the influence of the predicted state and the measured state. If the measurement data is more credible (i.e., is small), tends to trust the measurement data. If the measurement data has a large noise (i.e., is large), tends to trust the predicted data;

[0057] Update the data according to the predicted state vector and the adaptive state correction coefficient to obtain the aligned wind field environment data and wind turbine state data within the same time step. The calculation logic for updating the data is: , is the predicted update vector of the data at time, is the measured value of the data at time. Specifically, in the wind field optimization control, the wind turbine state data is collected by multiple sensors. However, due to problems such as different sampling frequencies, data delays, and measurement noises, different data may not be directly aligned, affecting the accuracy of the optimization calculation. To align the wind field environment data and the wind turbine state data within the same time step, data update is required; through the prediction model, calculate the predicted value of the wind turbine state, that is, the optimal estimate obtained based on the historical state and control input when there is no measurement data. The measurement data may contain noise and needs to be corrected in combination with the predicted state to improve the accuracy of the data. Calculate the measurement residual to measure the deviation between the predicted value and the measured value. Use the adaptive state correction coefficient to correct the predicted value to make it closer to the more reliable data. Finally, obtain the updated state to ensure that all wind turbine states are aligned within the same time step.

[0058] The present invention is further set that the wind field dynamic optimization objectives include maximizing the wind turbine power, minimizing the fatigue load, and wake interference;

[0059] Converting the maximization of wind turbine power into the minimization of negative wind turbine power, the calculation logic for maximizing wind turbine power is as follows: , is the number of wind turbines, is the wind turbine power mapping function, , is the air density, is the swept area of the wind turbine, is the power coefficient, is the tip speed ratio, is the pitch angle, is the wind speed; specifically, in the optimization control of the wind farm, maximizing the power output of the wind turbine is one of the core objectives. To unify the format, the maximization of wind turbine power is converted into the minimization of negative wind turbine power. The calculation of the wind turbine power mapping function is based on the aerodynamic model, represents the wind energy that the wind turbine can receive, which is related to the air density and the swept area, is the power coefficient, which describes the efficiency of converting wind energy into mechanical power, is the cube of the wind speed, indicating the influence of the wind speed on the power;

[0060] The calculation logic for minimizing the fatigue load is as follows: , is the fatigue load mapping function, , and are empirical parameters, is the wind turbine acceleration, is the torque fluctuation of the drive train; specifically, during the operation of the wind turbine, components such as the blades, main shaft, and drive train are subjected to alternating loads, which will cause fatigue damage over a long period of time, affecting the lifespan of the wind turbine. The optimization goal of minimizing the fatigue load is to reduce fatigue damage by controlling the operating parameters of the wind turbine, improving the structural stability and service life of the wind turbine;

[0061] The calculation logic for minimizing the wake interference is as follows: , is the wake effect mapping function, , is the induction factor, is the wake diffusion coefficient, is the wind turbine spacing, is the wind turbine diameter. Specifically, during the operation of the wind turbine, it extracts energy from the wind, which causes the wind speed behind it to decrease, forming a wake effect. The wake will reduce the incoming wind speed of the downstream wind turbine, affecting the power generation efficiency, and increasing the uneven load between the wind turbines, resulting in fatigue damage. To reduce the wake influence, it is necessary to minimize the wake wind speed and the free wind speed The difference is that the wake wind speed is the wind speed after the wind turbine is affected by the upstream wind turbine. Since the upstream wind turbine extracts energy from the wind, the downstream wind speed is reduced, affecting its power generation efficiency. First, calculate the free wind speed , that is, the incoming wind speed under undisturbed conditions. Calculate the induction factor of the upstream wind turbine , to quantify the degree of wind energy extraction by the wind turbine. Calculate the degree of wake diffusion, that is, the process of the wake gradually recovering as the distance between wind turbines increases, and calculate the wake wind speed , which is used to estimate the degree of influence on the downstream wind turbine.

[0062] The present invention is further configured such that the calculation logic of the uncertainty degree is: , is the uncertainty degree of the th objective function, is the number of data groups, is the probability density, , is the value of the th objective function in the th group of data; specifically, in the optimization control of the wind farm, the weights of each optimization objective, including maximizing the wind turbine power, minimizing the fatigue load, and minimizing the wake interference, will vary due to data fluctuations. To quantify the instability of these objective functions, it is necessary to calculate the uncertainty degree, which is used to describe the distribution of the objective function values in different data groups and dynamically adjust the objective weights during the optimization process;

[0063] The calculation logic of the dynamic importance degree is: , is the dynamic importance degree of the th objective function; specifically, in the optimization control of the wind farm, the dynamic importance degree is used to measure the relative weight of each objective during the optimization process to ensure that the optimization strategy can adapt to the operating state of the wind farm;

[0064] The calculation logic of the target collaborative contribution rate is: , is the target collaborative contribution rate of the th and the th objective functions, and are the values of the th and the th objective functions, and are the preset thresholds. Specifically, obtain the numerical values and of the objective functions in the current optimization state, evaluate the importance and of each objective in the current optimization task. If , indicating that there is little interaction between the nd and th objective functions, the collaborative contribution rate is set to 0. If , it indicates that there is a certain synergistic effect between the two objectives, and its strength is calculated . If , it indicates that the synergistic effect between the two objectives is very obvious, and the collaborative contribution rate is set to 1. A high collaborative contribution rate indicates that the two objectives can promote each other, and their weights should be increased simultaneously during optimization.

[0065] The present invention is further configured to construct a fuzzy analytic hierarchy process matrix according to the objective collaborative contribution rate , , and according to the fuzzy analytic hierarchy process matrix , solve the eigenvector corresponding to its maximum eigenvalue : , is the maximum eigenvalue of the fuzzy analytic hierarchy process matrix, is the final objective weight vector, , , and are the weights of the objective functions, which are normalized to construct a dynamic optimization objective function: , is the dynamic optimization objective function. Specifically, in wind farm optimization, it is necessary to simultaneously optimize three objectives: maximizing the wind turbine power, minimizing the fatigue load, and minimizing the wake interference. However, due to the interaction between different objectives, directly assigning fixed weights results in poor optimization effects. Therefore, it is necessary to dynamically calculate the objective weights according to the objective collaborative contribution rate through a fuzzy analytic hierarchy process matrix and construct an optimization objective function;

[0066] The present invention is further configured such that the wind farm constraint includes a control input constraint and a wind turbine state variable constraint. The control input constraint includes a pitch angle control constraint and a yaw angle control constraint. The wind turbine state variable constraint includes a wind turbine speed constraint, a fatigue load constraint, and a wake interference constraint. By controlling the pitch angle and the yaw angle within the constraints, the dynamic optimization objective function is solved to obtain an optimal control parameter set. Specifically, the purpose of the wind farm optimal control is to maximize the power output of the wind turbine while minimizing the fatigue load and the wake interference, ensuring the stable operation of the wind turbine group under complex wind conditions. However, during the optimization process, the physical constraints of the wind turbine and the control input limitations must be satisfied. Therefore, the present invention introduces the wind farm constraint, including: control input constraint (limiting the adjustable range of the control variable): pitch angle control constraint, yaw angle control constraint; wind turbine state variable constraint (ensuring the wind turbine operates within a safe range): wind turbine speed constraint, fatigue load constraint, wake interference constraint; finally, by optimizing the pitch angle and the yaw angle within the constraints, the dynamic optimization objective function is solved to obtain an optimal control parameter set, ensuring the wind turbine operates under efficient and safe conditions.

[0067] The present invention is further configured such that the construction logic of the safety barrier function is: , is the safety barrier function, is the maximum eigenvalue of the wind turbine, is the influence of the control input on the wind turbine state. When , through verification, the optimal control parameter set is executed for wind farm dynamic optimization; specifically, in wind farm optimal control, the safety of wind turbine operation is of utmost importance. The optimized control strategy (such as pitch angle and yaw angle adjustment) must ensure that the wind turbine operates within a safe range, avoiding mechanical damage or efficiency reduction caused by overload or unstable operation. Therefore, a safety barrier function is needed to verify the feasibility of the optimization result and ensure that the wind turbine control strategy meets the safety constraints. The safety barrier function is used to measure whether the current state of the wind turbine and the control input meet the safety requirements, control input the influence gradient on the wind turbine state , reflecting whether the current control strategy conforms to the safety constraints; when , the control strategy passes the safety verification, and the optimal control parameter set is executed to achieve wind farm dynamic optimization;

[0068] When , the fault-tolerant control mechanism is entered for adjustment, , is the adjusted control parameter set, is the current control parameter set, is the optimal control parameter set, Taking the safety barrier function as the constraint condition, within the feasible region, the adjusted control parameter set is solved by the projected gradient descent method. The adjusted control parameter set is input into the safety barrier function for verification, and if the verification passes, it is executed. Specifically, in wind farm optimization control, if the safety barrier function is less than 0, the current control parameters may cause the wind turbine to operate unstably or exceed the safety limit. Therefore, a fault-tolerant control mechanism is needed to adjust the control input to meet the safety constraints. This adjustment process uses the projected gradient descent method to optimize the control parameters to make them as close as possible to the optimal control parameter set and satisfy the safety barrier function constraint. Finally, the adjusted control parameters pass the verification of the safety barrier function, ensuring that the wind turbine operates within the safe range, and the optimized control strategy is executed. If it does not pass, the above steps are repeated until the verification of the safety barrier function passes. This fault-tolerant control mechanism automatically detects the safety of the control parameters and makes minimal adjustments in the risk state to restore the safe operation of the wind turbine, ultimately ensuring that the wind turbine can achieve optimal control under different wind conditions, improving the long-term economic benefits and operation stability of the wind farm.

[0069] Embodiment 2

[0070] Please refer to Figure 2 , an exemplary wind farm dynamic optimization system based on real-time data adaptive regulation, which is used to implement the above-mentioned wind farm dynamic optimization method based on real-time data adaptive regulation. The system includes:

[0071] Data acquisition module: Acquire wind farm environment data and wind turbine status data, map the data to the state space model, and align them within the same time step;

[0072] Objective construction module: Construct the wind farm dynamic optimization objective, calculate the uncertainty degree, dynamic importance degree, and objective collaborative contribution rate based on the wind farm dynamic optimization objective, construct a fuzzy analytic hierarchy process matrix according to the objective collaborative contribution rate, solve to obtain the weight of the wind farm dynamic optimization objective, and construct a dynamic optimization objective function based on the wind farm dynamic optimization objective and the weight;

[0073] Optimal generation module: Construct wind farm constraints, solve the dynamic optimization objective function under the condition of meeting the wind farm constraint conditions, and generate an optimal control parameter set;

[0074] Safety verification module: Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set to perform wind farm dynamic optimization.

[0075] It should be noted that a wind farm dynamic optimization system based on real-time data adaptive regulation provided by the above embodiments and a wind farm dynamic optimization method based on real-time data adaptive regulation provided by the above embodiments belong to the same concept. The specific ways in which each module and unit perform operations have been described in detail in the method embodiments, and will not be elaborated here. In practical applications, a wind farm dynamic optimization system based on real-time data adaptive regulation provided by the above embodiments can, as needed, allocate the above functions to different functional modules, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above. No limitation is imposed here either.

[0076] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0077] It should be understood that the term “and / or” in this article is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character “ / ” in this article generally represents an “or” relationship between the associated objects before and after, but it may also represent an “and / or” relationship. The specific meaning can be understood by referring to the context before and after.

[0078] In this application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or a similar expression means any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0079] It should be understood that in various embodiments of this application, the magnitudes of the sequence numbers of the above processes do not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0080] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0081] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0082] In several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in an electrical, mechanical, or other form.

[0083] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, in each embodiment of the present application, each functional unit may be integrated into one processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit.

[0085] If the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0086] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A dynamic optimization method for a wind farm based on real-time data adaptive regulation, characterized in that Including: Obtain wind farm environment data and wind turbine status data, map the data to a state space model, and align them within the same time step; The wind farm environment data includes wind speed, wind direction, wind shear index, turbulence intensity, wake wind speed, and wake diffusion coefficient; the wind turbine status data includes wind turbine power output, rotational speed, tip speed ratio, pitch angle, yaw angle, bearing load, and drivetrain torque; Construct the dynamic optimization objective of the wind farm, calculate the uncertainty degree, dynamic importance degree and target collaborative contribution rate based on the dynamic optimization objective of the wind farm, construct a fuzzy analytic hierarchy process matrix according to the target collaborative contribution rate, solve to obtain the weight of the dynamic optimization objective of the wind farm, and construct a dynamic optimization objective function according to the dynamic optimization objective of the wind farm and the weight; the dynamic optimization objective of the wind farm includes maximizing the wind turbine power, minimizing the fatigue load and wake interference; convert maximizing the wind turbine power into minimizing the negative wind turbine power, and the calculation logic of maximizing the wind turbine power is: , where is the wind turbine power mapping function, , is the air density, is the swept area of the wind turbine, is the power coefficient, is the tip speed ratio, is the pitch angle, is the wind speed; the calculation logic of minimizing the fatigue load is: , is the fatigue load mapping function, , and are empirical parameters, is the wind turbine acceleration, is the torque fluctuation of the drive train; the calculation logic of minimizing the wake interference is: , is the wake effect mapping function, , is the induction factor, is the wake diffusion coefficient, is the wind turbine spacing, is the wind turbine diameter; The calculation logic of the uncertainty degree is as follows: , is the uncertainty degree of the th objective function, is the number of data groups, is the probability density, , is the value of the th objective function in the th group of data; The calculation logic of the dynamic importance degree is as follows: , is the dynamic importance degree of the th objective function; The calculation logic of the target collaborative contribution rate is as follows: , is the target collaborative contribution rate of the th and the th objective functions, and are the values of the th and the th objective functions, and are the preset thresholds; Construct a fuzzy analytic hierarchy process matrix based on the target collaborative contribution rate , , and based on the fuzzy analytic hierarchy process matrix , solve the eigenvector corresponding to its maximum eigenvalue : , is the maximum eigenvalue of the fuzzy analytic hierarchy process matrix, is the final target weight vector, , , and are the weights of the objective function. After normalization, construct a dynamic optimization objective function: , is the dynamic optimization objective function; Construct wind farm constraints, solve the dynamic optimization objective function under the condition of meeting the wind farm constraint conditions, and generate an optimal control parameter set; Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set for wind farm dynamic optimization.

2. The dynamic optimization method for a wind farm based on real-time data adaptive regulation according to claim 1, wherein Mapping the data to the state space model and aligning them within the same time step includes: Construct an initial state vector and covariance matrix according to the wind farm environment data and wind turbine status data; Perform state prediction based on the initial state vector and output the predicted state vector; Calculate the adaptive state correction coefficient according to the covariance matrix; Update the data according to the predicted state vector and the adaptive state correction coefficient to obtain the wind farm environment data and wind turbine status data aligned within the same time step.

3. The dynamic optimization method for a wind farm based on real-time data adaptive regulation according to claim 2, wherein The calculation logic of the predicted state vector is as follows: , is the data at the predicted state vector at time is the state transition matrix, is the data at the predicted update vector at time is the control input matrix, is the control input vector; The calculation logic of the adaptive state correction coefficient is as follows: , is the adaptive state correction coefficient, is the prediction error covariance matrix, is the observation matrix, is the transpose of the observation matrix, is the measurement noise covariance matrix; The calculation logic for updating the data is as follows: , For the data at time, predict the update vector, For the data at time, the measured value.

4. A dynamic optimization method for a wind farm based on real-time data adaptive regulation according to claim 1, characterized in that The wind farm constraints include control input constraints and wind turbine state variable constraints. The control input constraints include pitch angle control constraints and yaw angle control constraints. The wind turbine state variable constraints include wind turbine rotational speed constraints, fatigue load constraints, and wake interference constraints. Solve the dynamic optimization objective function by controlling the pitch angle and yaw angle within the constraint range to obtain the optimal control parameter set.

5. A dynamic optimization method for a wind farm based on real-time data adaptive regulation according to claim 1, characterized in that, The construction logic of the safety barrier function is as follows: , is the safety barrier function, is the maximum eigenvalue of the fan, is the influence of the control input on the fan state. When , through verification, execute the optimal control parameter set to perform dynamic optimization of the wind farm; When it enters the fault-tolerant control mechanism for adjustment, , is the adjusted control parameter set, is the current control parameter set, is the optimal control parameter set, is the constraint condition. Within the feasible region, the adjusted control parameter set is solved by the projected gradient descent method, and the adjusted control parameter set is input into the safety barrier function for verification. If the verification passes, it is executed.

6. A wind farm dynamic optimization system based on real-time data adaptive regulation, which is used to implement a wind farm dynamic optimization method based on real-time data adaptive regulation according to any one of claims 1-5, and is characterized in that, Including: Data acquisition module: Obtain wind farm environment data and wind turbine status data, map the data to a state space model, and align them within the same time step; Objective construction module: Construct the wind farm dynamic optimization objective, calculate the uncertainty degree, dynamic importance, and objective collaborative contribution rate based on the wind farm dynamic optimization objective, construct a fuzzy analytic hierarchy process matrix according to the objective collaborative contribution rate, solve to obtain the weight of the wind farm dynamic optimization objective, and construct a dynamic optimization objective function according to the wind farm dynamic optimization objective and the weight; Optimal generation module: Construct wind farm constraints, solve the dynamic optimization objective function under the condition of meeting the wind farm constraint conditions, and generate an optimal control parameter set; Safety verification module: Input the optimal control parameter set into the safety barrier function for verification. After passing the verification, execute the optimal control parameter set for wind farm dynamic optimization.

Images