Wind power plant wake flow dynamic yaw control method, device, equipment and medium
Patent Information
- Application Number
- CN202411067020.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-05
- Publication Date
- 2024-11-12
AI Technical Summary
The prior art cannot achieve precise control of the dynamic yaw of the wind farm wake, resulting in low power generation efficiency.
By obtaining wind farm environmental information, determining the power information of the wind farm, and using genetic algorithms to optimize the yaw angle sequence, adjusting the yaw angle of the wind turbine in real time to maximize the power generation power of the wind farm and respond to instantaneous wind speed changes.
Accurate control of the dynamic yaw of the wake of the wind farm is achieved, which improves the overall power generation efficiency of the wind farm and reduces the impact of the wake effect.
Smart Images

Figure CN118934447A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of wind power generation, and specifically provides a method, device, equipment and medium for dynamic yaw control of wake flow in a wind farm. Background Art
[0002] Wind farm yaw coordinated optimization control is an effective means to reduce the wake effect of wind farms and increase the total power generation of wind farms. According to whether the yaw angle is dynamically adjusted over time, the wake deflection control can be divided into two categories: fixed yaw control and dynamic yaw control.
[0003] However, neither the fixed yaw control method nor the dynamic yaw control method can achieve precise control of the dynamic yaw of the wind farm wake. Summary of the invention
[0004] In order to overcome the above defects, the present application is proposed to provide a solution or at least partially solve the technical problem that the existing methods cannot achieve accurate control of the dynamic yaw of the wind farm wake. The present application provides a method, device, equipment and medium for controlling the dynamic yaw of the wind farm wake.
[0005] In a first aspect, the present application provides a method for dynamic yaw control of a wind farm wake, characterized in that the method includes: acquiring environmental information in the wind farm; determining power information of the wind farm based on the environmental information; determining an optimal yaw angle sequence based on the power information of the wind farm; and performing yaw control on each wind turbine in the wind farm based on the optimal yaw angle sequence.
[0006] In one embodiment of the present application, the environmental information includes the inflow wind speed of each wind turbine set at the wind rotor, and the inflow wind speed of each wind turbine set at the wind rotor is obtained by the following steps: acquiring the thrust coefficient, yaw angle and instantaneous inflow wind speed of each wind turbine set of the wind farm; generating a wind speed Taylor background field of the wind farm based on the instantaneous inflow wind speed of the wind farm; determining a first steady-state speed loss based on the thrust coefficient and the yaw angle; and determining the inflow wind speed of each wind turbine set at the wind rotor based on the wind speed Taylor background field and the first steady-state speed loss.
[0007] In one embodiment of the present application, determining the first steady-state speed loss based on the thrust coefficient and the yaw angle includes generating a second steady-state speed loss based on the thrust coefficient and the yaw angle; determining the first steady-state speed loss based on the second steady-state speed loss, the first steady-state speed loss is a steady-state speed loss of multiple units, and the second steady-state speed loss is a steady-state speed loss of each wind turbine unit.
[0008] In one embodiment of the present application, determining the inlet wind speed of each wind turbine at the wind rotor based on the wind speed Taylor background field and the first steady-state speed loss includes: determining the instantaneous wake field of multiple units based on the wind speed Taylor background field and the first steady-state speed loss; and obtaining the inlet wind speed of each wind turbine at the wind rotor based on the instantaneous wake field of multiple units.
[0009] In one embodiment of the present application, determining the power information of the wind farm based on the environmental information includes: determining the instantaneous power of each wind turbine based on the environmental information, the instantaneous power being the instantaneous power of the wind turbine at any time in the prediction time domain; determining the power information of the wind farm in the prediction time domain based on the instantaneous power of each wind turbine.
[0010] In one embodiment of the present application, determining the optimal yaw angle sequence based on the power information of the wind farm includes: determining an expression of an optimization target based on the power information of the wind farm; establishing an initial population, wherein the initial population includes multiple individuals, each individual is a yaw action sequence of each wind turbine in the wind farm; starting from the initial population, using a genetic algorithm to perform target optimization iterations to maximize the optimization target; when the optimization target converges, obtaining the optimal yaw action sequence of each wind turbine; and determining the optimal yaw angle sequence of each wind turbine based on the optimal yaw action sequence.
[0011] In one embodiment of the present application, the target optimization iteration using a genetic algorithm includes: obtaining the constraints of the yaw angle sequence of each wind turbine; and performing target optimization iteration using a genetic algorithm starting from the initial population under the constraints of the yaw angle sequence of each wind turbine.
[0012] In a second aspect, a wind farm wake dynamic yaw control device is provided, the device comprising:
[0013] An acquisition module, wherein the acquisition module is configured to acquire environmental information in the wind farm;
[0014] A first determination module, wherein the first determination module is configured to determine power information of the wind farm based on the environmental information;
[0015] a second determination module, the second determination module being configured to determine an optimal yaw angle sequence based on power information of the wind farm;
[0016] A control module is configured to perform yaw control on each wind turbine in the wind farm based on the optimal yaw angle sequence.
[0017] In a third aspect, a computer device is provided, comprising:
[0018] at least one processor;
[0019] and, a memory communicatively coupled to the at least one processor;
[0020] Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the aforementioned wind farm wake dynamic yaw control method is implemented.
[0021] In a fourth aspect, a computer-readable storage medium is provided, wherein a plurality of program codes are stored in the computer-readable storage medium, wherein the program codes are suitable for being loaded and run by a processor to execute any of the wind farm wake dynamic yaw control methods described above.
[0022] The above one or more technical solutions of this application have at least one or more of the following Beneficial effects:
[0023] A method for controlling the dynamic yaw of a wind farm wake in the present application specifically includes: obtaining environmental information in the wind farm; determining the power information of the wind farm based on the environmental information; determining the optimal yaw angle sequence based on the power information of the wind farm; and controlling the yaw of each wind turbine in the wind farm based on the optimal yaw angle sequence. By real-time monitoring and dynamic adjustment of the yaw angle, the wind farm's ability to respond to instantaneous wind speed changes is improved, the impact of the wake effect is reduced, and the overall power generation efficiency of the wind farm is improved, and precise control of the dynamic yaw of the wind farm wake is achieved. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The disclosure of the present application will become more easily understood with reference to the accompanying drawings. It is easy for those skilled in the art to understand that these drawings are only for illustrative purposes and are not intended to limit the scope of protection of the present application. In addition, similar numbers in the drawings are used to represent similar components, among which:
[0025] Figure 1 This is a schematic diagram of the main process of a method for dynamic yaw control of a wind farm wake in one embodiment of the present application;
[0026] Figure 2 It is a schematic diagram of the structure of a convolutional neural network in one embodiment of the present application;
[0027] Figure 3 It is a complete schematic diagram of a process for predicting the instantaneous wake field of multiple units in one embodiment of the present application;
[0028] Figure 4 It is a complete flow chart of a method for dynamic yaw control of a wind farm wake in one embodiment of the present application;
[0029] Figure 5It is a schematic diagram of the main structure of a wind farm wake dynamic yaw control device in one embodiment of the present application;
[0030] Figure 6 It is a schematic diagram of the structure of a computer device in one embodiment of the present application. DETAILED DESCRIPTION
[0031] Some embodiments of the present application are described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.
[0032] In the description of the present application, "module" and "processor" may include hardware, software or a combination of the two. A module may include hardware circuits, various suitable sensors, communication ports, memory, and may also include software parts, such as program code, or a combination of software and hardware. The processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor or any other suitable processor. The processor has data and / or signal processing functions. The processor may be implemented in software, hardware or a combination of the two. Non-temporary computer-readable storage media include any suitable medium that can store program code, such as a disk, a hard disk, an optical disk, a flash memory, a read-only memory, a random access memory, etc. The term "A and / or B" means all possible combinations of A and B, such as only A, only B or A and B. The term "at least one A or B" or "at least one of A and B" has a similar meaning to "A and / or B" and may include only A, only B or A and B. The singular terms "one" and "the" may also include plural forms.
[0033] Wind farm yaw coordinated optimization control is an effective means to reduce the wind farm wake effect and increase the total power generation of the wind farm. According to whether the yaw angle is dynamically adjusted over time, the wake deflection control can be divided into two categories: fixed yaw control and dynamic yaw control.
[0034] Fixed yaw control usually refers to a control method that uses a steady-state wake model to evaluate the time-averaged power generation of a wind farm, and then uses a heuristic algorithm to obtain the optimal yaw angle of each unit in the wind farm, and the optimal yaw angle of each unit remains constant during the control cycle. Although a large number of studies have proven that fixed yaw control can increase the overall power generation of a wind farm. However, since fixed yaw control is optimized based on a steady-state flow field, it is impossible to dynamically adjust the yaw angle of each unit according to the turbulent pulsation of the instantaneous inflow wind speed, and therefore the power enhancement potential of wake deflection control is not fully utilized.
[0035] The dynamic wake model has certain accuracy limitations when evaluating the instantaneous wake velocity distribution within a wind farm.
[0036] Therefore, neither the fixed yaw control method nor the dynamic yaw control method can achieve precise control of the dynamic yaw of the wind farm wake.
[0037] To this end, the present application proposes a method, device, equipment and medium for dynamic yaw control of a wind farm wake.
[0038] See attached Figure 1 , Figure 1 It is a schematic flow chart of the main steps of a method for dynamic yaw control of a wind farm wake according to an embodiment of the present application.
[0039] like Figure 1 As shown, the wind farm wake dynamic yaw control method in the embodiment of the present application mainly includes the following steps S10 to S40.
[0040] Step S10: Acquire environmental information of the wind farm.
[0041] The environmental information includes the state parameters of each wind turbine and the inflow wind speed of each wind turbine at the wind rotor.
[0042] The state parameters at least include the thrust coefficient and yaw angle of each wind turbine generator set.
[0043] The inflow wind speed of a wind turbine at the wind rotor is the speed of the air flow passing through the wind rotor blades.
[0044] Step S20: determining power information of the wind farm based on the environmental information.
[0045] The power information of a wind farm refers to the ability of the wind farm to convert wind energy into electrical energy through its generator sets and the relevant data on its actual output power.
[0046] Step S30: determining an optimal yaw angle sequence based on the power information of the wind farm.
[0047] The yaw angle sequence refers to the yaw angle setting of each wind turbine in a wind farm obtained under preset wind conditions and other operating conditions.
[0048] Step S40: performing yaw control on each wind turbine generator set based on the optimal yaw angle sequence.
[0049] Based on the above steps S10 to S40, firstly, the environmental information in the wind farm is obtained; based on the environmental information, the power information of the wind farm is determined; based on the power information of the wind farm, the optimal yaw angle sequence is determined; and based on the optimal yaw angle sequence, the yaw of each wind turbine is controlled. By real-time monitoring and dynamic adjustment of the yaw angle, the responsiveness of the wind farm to instantaneous wind speed changes is improved, the influence of the wake effect is reduced, thereby improving the overall power generation efficiency of the wind farm, and realizing precise control of the dynamic yaw of the wind farm wake.
[0050] The above steps S10 to S40 are further explained below.
[0051] Specifically, for the above step S10, the environmental information includes the inflow wind speed of each wind turbine set at the wind rotor, and the inflow wind speed of each wind turbine set at the wind rotor can be obtained through the following steps S101 to S104.
[0052] Step S101: acquiring the thrust coefficient, yaw angle and instantaneous inflow wind speed of each wind turbine generator set.
[0053] Specifically, the instantaneous inflow wind speed of the wind farm can be obtained through the wind turbine cabin radar. The yaw angle can be obtained by the difference between the wind direction angle of the wind turbine and the steering angle of the wind turbine. The thrust coefficient can be calculated from the monitoring data of the wind turbine SCADA monitoring system.
[0054] Step S102: Generate a wind speed Taylor background field of the wind farm based on the instantaneous inflow wind speed of the wind farm.
[0055] Specifically, the wind speed Taylor background field includes the streamwise wind speed Taylor background field and the spanwise wind speed Taylor background field, which are generated based on the instantaneous inflow wind speed of the wind farm by adopting the Taylor frozen flow assumption.
[0056] The Taylor background field of streamwise wind speed refers to the spatial distribution of the streamwise component of wind speed in a wind farm. The Taylor background field of streamwise wind speed describes the variation characteristics of the streamwise wind speed in a wind farm and reflects the spatial non-uniformity of the streamwise wind speed.
[0057] The spanwise wind speed Taylor background field refers to the spatial distribution of the spanwise component of wind speed in a wind farm. The spanwise wind speed Taylor background field describes the variation characteristics of the spanwise direction in a wind farm and reflects the spatial non-uniformity of the spanwise wind speed.
[0058] The Taylor frozen flow hypothesis states that the spatial structure of turbulent motion does not change significantly after passing through a fixed point, that is, the spatial structure of turbulence is like being frozen. According to the Taylor frozen flow hypothesis, if the average wind speed in the flow direction of the flow field is assumed to be U, then the velocity vector u(x, t) at the flow direction position x at time t should be equal to the velocity vector u(x+Uτ, t+τ) at the downstream position x+Uτ at time t+τ, where τ is the flow evolution time. The corresponding expression can be expressed as: u(x+Uτ,t+τ)=u(x,t)
[0059] Therefore, according to the Taylor frozen flow hypothesis, we can get an approximate snapshot of the flow field that is continuous in space by arranging the instantaneous inflow wind speeds measured at the entrance of the flow field in sequence according to the flow distance. That is, the Taylor background field of the flow direction wind speed u can be obtained according to the instantaneous inflow wind speed of the wind farm. t and spanwise wind speed Taylor background field v t , and use them as two input features of the convolutional neural network.
[0060] Taylor background field refers to the wind speed background field obtained by Taylor frozen flow assumption. Since wind speed has three components: flow direction, span direction and vertical direction, this application determines the background field of wind speed in the flow direction and span direction, that is, the wind speed Taylor background field u t and spanwise wind speed Taylor background field v t .
[0061] Step S103: determining a first steady-state speed loss based on the thrust coefficient and the yaw angle, where the first steady-state speed loss is a multi-unit steady-state speed loss.
[0062] Steady-state speed loss refers to the reduction in wind speed relative to the inflow wind speed after the wind passes through the wind turbine under steady-state conditions.
[0063] In a specific implementation, step S103 may be implemented through the following steps S1031 to S1032.
[0064] Step S1031: generating a second steady-state speed loss based on the thrust coefficient and the yaw angle, where the second steady-state speed loss is the steady-state speed loss of each wind turbine generator set.
[0065] Specifically, the overall wake velocity loss is first generated according to the thrust coefficient and yaw angle of the wind turbine. According to the Shapiro steady-state wake model, the calculation formula for the overall wake velocity loss is:
[0066] The offset of the wake center is determined according to the thrust coefficient and yaw angle of the wind turbine, where the calculation formula for the offset of the wake center is:
[0067] The calculation formula of the normalized wake width is:
[0068] The calculation formula for determining the steady-state speed loss based on the overall wake speed loss, the offset of the wake center, and the normalized wake width is:
[0069] in, is the steady-state wake loss of the i-th wind turbine under different operating conditions; δu(x) is the overall wake velocity loss; y c (x) is the offset of the wake center; d w (x) is the normalized wake width; u i is the inflow wind speed of the i-th wind turbine; x and y are the flow direction and span direction distances from the center of the wind rotor, respectively; C T is the thrust coefficient of the wind turbine; γ is the yaw angle; R is the rotor radius; D is the rotor diameter; σ0 is the standard deviation of the wake Gaussian distribution, σ0 = 0.235D; k w is the wake expansion coefficient; erf is the error function.
[0070] Step S1032: Determine the first steady-state speed loss based on the second steady-state speed loss.
[0071] Specifically, there are usually multiple wind turbines in a wind farm. By using the square sum superposition method to superimpose the steady-state speed losses of each wind turbine in the wind farm, the speed losses of multiple wind turbines can be obtained. (First steady-state speed loss), the specific calculation formula is:
[0072] Among them, u0 is the average inflow wind speed of the wind farm; N is the number of upstream units; x and y are the flow direction and span direction distances respectively.
[0073] Step S104: determining the inflow wind speed of each wind turbine at the wind rotor based on the wind speed Taylor background field and the first steady-state speed loss.
[0074] The inflow wind speed of a wind turbine at the wind rotor refers to the wind speed at the location of the wind rotor of the wind turbine, that is, the wind speed cut when the wind rotor rotates.
[0075] In a specific implementation, step S104 may be implemented through the following steps S1041 to S1042.
[0076] Step S1041: Determine the instantaneous wake field of multiple units based on the wind speed Taylor background field and the first steady-state speed loss.
[0077] The multi-unit instantaneous wake field refers to the instantaneous velocity field including the wakes of all wind turbines in a wind farm. The multi-unit instantaneous wake field in this application can characterize the instantaneous wake velocity distribution of all wind turbines in a wind farm under different operating conditions.
[0078] Specifically, the Taylor background field u of the flow velocity can be t , spanwise wind speed Taylor background field v t The first steady-state speed loss is input into the convolutional neural network, and the instantaneous wake field of multiple units is output.
[0079] Convolutional Neural Networks (CNN) is a deep learning algorithm. For example, ResNet, VGGNet, etc. can be used as examples of the convolutional neural network.
[0080] In a specific embodiment of the present application, the convolutional neural network may include an autoencoder module and a linear weight layer, wherein the autoencoder module is used to encode and decode the input features (wind speed Taylor background field and the first steady-state velocity loss) to obtain single-channel image features; the linear weight layer is used to jump-connect the network input features and the single-channel image features output by the autoencoder module, and perform linear weighting as the output of the convolutional neural network. The autoencoder module may include an encoder and a decoder connected in sequence, wherein the encoder includes at least five convolutional layers connected in sequence, and the decoder includes at least five deconvolutional layers connected in sequence.
[0081] The calculation formula for linear weighting of network input features and single-channel image features is: u w =A1·u t +A2·u CNN
[0082] Among them, u w is the output of the convolutional neural network; u t is the network input feature; u CNN is a single-channel image feature; A1 and A2 are parameter matrices to be trained with the same size as the output; · is the operation operator, representing the multiplication of corresponding elements of the matrix.
[0083] Figure 2 A schematic diagram of the structure of a convolutional neural network in one embodiment of the present application.
[0084] Specific as Figure 2 As shown, the input features (wind speed Taylor background field u) are first converted through the encoder and decoder. t , spanwise wind speed Taylor background field v t and the first steady-state velocity loss) to obtain a single-channel image feature u with the same resolution as the target flow field snapshot CNN , then the linear weight layer skips the convolution layer of the encoder and the deconvolution layer of the decoder, and converts u in the input feature t (stream wind speed Taylor background field) and the decoder output u CNN Perform skip connections and use linear weighting as the final output of the neural network model.
[0085] The convolutional neural network includes an encoder, a decoder and a linear weight layer connected in sequence, wherein the encoder includes five convolutional layers connected in sequence, and the decoder includes five deconvolutional layers connected in sequence. A rectified linear unit (ReLU) activation function can also be connected after the convolution and deconvolution layers for activation to increase the nonlinear fitting ability of the network.
[0086] Specifically, the first convolutional layer is used to input the network input features (flow direction wind speed Taylor background field u t , spanwise wind speed Taylor background field v t and the first steady-state speed loss) to extract features to obtain the first convolution feature, and then the first feature is input into the second convolution layer for further feature extraction to obtain the second convolution feature, and so on, until the output features of the fifth deconvolution layer are further input into the linear weight layer.
[0087] The linear weight layer is used to deconvolution the single-channel image feature u output by the fifth deconvolution layer. CNN Taylor background field u t Linear weighting is performed to obtain the final prediction results of the instantaneous wake field of multiple units.
[0088] The linear weight layer consists of two trainable parameter matrices of the same size as the output. The linear weight layer transforms the single-channel image feature u output by the decoder CNN Taylor background field u t Linear superposition is performed to obtain the final output of the model (the prediction results of the instantaneous wake field of multiple units).
[0089] Step S1042: obtaining the inflow wind speed of each wind turbine at the wind rotor based on the instantaneous wake field of the multiple turbines.
[0090] Specifically, the instantaneous wake field of multiple wind turbines refers to the instantaneous wake velocity distribution of all wind turbines in a wind farm, which includes the inflow wind speed of each wind turbine at the wind rotor. Therefore, the inflow wind speed at the corresponding wind rotor of each wind turbine can be obtained from the instantaneous wake field of multiple wind turbines.
[0091] According to the key parameters such as the thrust coefficient of the wind turbine, the yaw angle and the instantaneous inflow wind speed of the wind farm, it is helpful to improve the accuracy of the inflow wind speed calculation. The use of the wind speed Taylor background field to simulate the instantaneous wind speed distribution better reflects the complex flow characteristics in the actual wind farm. Accurate inflow wind speed data helps to improve the wind energy conversion effect, thereby enhancing the overall power generation of the wind farm.
[0086] The above is a further description of step S10 , and the following is a further description of step S20 .
[0087] Specifically, step S10 can be implemented through the following steps S201 to S202.
[0088] Step S201: Determine the instantaneous power of each wind turbine generator set based on environmental information, where the instantaneous power is the instantaneous power of the wind turbine generator set at any time in the prediction time domain.
[0089] The prediction time domain is the time period used to predict future system behavior, which can be adaptively adjusted according to the actual application scenario.
[0090] The instantaneous power of a wind turbine is the average instantaneous power P of each wind turbine at any time in the prediction time domain. wf (t).
[0091] Specifically, for any time t in the prediction time domain, the average instantaneous power P of each unit in the wind farm is wf The expression of (t) is:
[0092] Among them, N tur is the total number of units in the wind farm; P i (t) is the instantaneous power of the i-th wind turbine; ρ is the air density, which is 1.225 kg / m 3 ; R is the radius of the wind wheel; C P is the power coefficient of the wind turbine; u d,i is the inflow wind speed of the i-th wind turbine at the wind rotor; γ i is the yaw angle of the i-th wind turbine.
[0093] Step S202: Determine the average power of the wind farm in the forecast time domain based on the instantaneous power of each wind turbine
[0092] Specifically, the prediction time domain T starting from time t0 P Average power of wind farms within It can be expressed as:
[0093] Among them, N P =T P / ΔT is the total number of samples in the prediction period; ΔT is the sampling period.
[0094] The above is a further description of step S20 , and the following is a further description of step S30 .
[0095] Specifically, step S30 can be implemented through the following steps S301 to S305.
[0096] Step S301: determining an expression of an optimization target based on power information of a wind farm.
[0097] Specifically, the expression of the optimization objective (Obj) can be to maximize the power information of the wind farm, which can be specifically expressed as:
[0098] in, The size of the yaw angle sequence in the prediction time domain is Determined.
[0099] In the embodiment of the present application, constraint conditions for the yaw angle sequence of each wind turbine generator set may also be established to constrain the optimization process.
[0100] Illustratively, the following constraint conditions of the yaw angle sequence may be used as examples of the constraint conditions, but are not limited thereto.
[0101] Step S302: establishing an initial population, wherein the initial population includes a plurality of individuals, each of which is a yaw action sequence of each wind turbine in the wind farm.
[0102] The yaw action sequence of a wind turbine generator set refers to the difference between the yaw angle of each wind turbine generator set at the next moment and the current moment.
[0103] Specifically, an initial population may be randomly generated using uniform distribution, the initial population including a plurality of individuals (eg, N, where N is a positive integer), each of which is a yaw action sequence Δγ of each wind turbine in the wind farm.
[0104] Step S303: Starting from the initial population, a genetic algorithm is used to perform target optimization iterations to maximize the optimization target.
[0105] Specifically, each individual is first converted into the yaw angle of each wind turbine, that is, the yaw action sequence of each wind turbine in the wind farm is converted into the yaw angle of each wind turbine, that is:
[0106] Then according to the above average power of wind farm The calculation formula is used to calculate the average power of the wind farm corresponding to each individual in the parent generation.
[0107] Continue to perform crossover and mutation operations on the individuals of the parent generation to generate new offspring, and convert the individual genes of the offspring into yaw angles.
[0108] According to the above average wind farm power The calculation formula for calculating the average power of the wind farm of the next generation is
[0109] Merge the parent and offspring populations to obtain a merged population. Select N wind farm individuals with the highest power from the merged population to form a new population. Repeat the above steps until the optimization target converges.
[0110] Step S304: when the optimization target converges, the optimal yaw action sequence of each wind turbine is obtained.
[0111] Specifically, when the optimization target converges, the optimal yaw action sequence of each wind turbine at the current moment is obtained.
[0112] Step S305: determining the optimal yaw angle sequence of each wind turbine generator system based on the optimal yaw action sequence.
[0113] Specifically, the optimal yaw action sequence is converted into the optimal yaw angle of the wind turbine, and the corresponding yaw control is performed on the wind turbine according to the yaw angle.
[0114] By optimizing the yaw action of wind turbines, the power output of the entire wind farm is increased. By optimizing the yaw action, the mutual interference between wind turbines is reduced, the wake effect is reduced, and the total power generation of the wind farm is increased. By determining the optimal yaw action sequence, the interference between adjacent wind turbines can be effectively reduced, and the overall efficiency of the wind farm can be improved.
[0115] Figure 3 It is a complete flow diagram of predicting the instantaneous wake field of multiple units in one embodiment of the present application. The inflow wind speed of each wind turbine unit at the wind rotor can be obtained based on the predicted instantaneous wake field of multiple units.
[0116] like Figure 3 As shown, the dynamic wake field model framework in this application can be divided into two parts as a whole: a physical model and a neural network. The physical model includes the Taylor frozen flow hypothesis and the Shapiro steady-state wake model, wherein the Taylor frozen flow hypothesis is used to obtain the Taylor background field of the flow wind speed and the Taylor background field of the span wind speed, and the Shapiro steady-state wake model is used to obtain the steady-state speed loss of a single unit, and then the steady-state speed loss of multiple units is obtained by the square sum superposition method. Furthermore, firstly, according to the measured inflow wind speed and the operating state parameters of the unit, the physical model is used to generate the input features of the neural network, and then it is input into the convolutional neural network to characterize the mapping relationship from the physical input features to the instantaneous velocity distribution of the wakes of multiple units, thereby realizing the real-time prediction of the dynamic wake field of multiple units.
[0117] Figure 4 1 is a schematic diagram of a complete process of a method for controlling a dynamic yaw of a wind farm wake in one embodiment of the present application. Specifically, the complete process of the method for controlling a dynamic yaw of a wind farm wake can be obtained through the following steps.
[0118] S1. At the initial moment when the control starts, the instantaneous inflow wind speed of the wind farm at the current moment and the current yaw angle of each unit are obtained.
[0119] S2, prediction time domain T P The steady-state speed loss of the wind farm in the prediction time domain T is predicted, and the wind farm speed loss in the prediction time domain T is calculated accordingly. P The specific prediction method of the steady-state speed loss of the wind farm can be referred to the aforementioned step S10, which will not be described in detail here.
[0120] In the prediction time domain T P At any time t, first calculate the average instantaneous power P of each wind turbine in the wind farm wf (t).
[0121] When performing model predictive control, it is necessary to evaluate the prediction time domain T P The wind farm power within the control period is used to determine the optimal yaw angle within the control period, where the prediction time domain T starting from time t0 P The average power of wind farms within Please refer to the above calculation method, which will not be repeated here.
[0122] S3, using genetic algorithm to predict the time domain T P The yaw angle sequence is optimized by finding the P The yaw angle sequence of each unit in the wind farm that maximizes the power.
[0123] Using genetic algorithm for optimization, the objective function can be written as:
[0124] in, The size of the yaw angle sequence in the prediction time domain is Determined.
[0125] The prediction time domain T P Set to 200 seconds to control the time domain T c To reduce the amount of calculation, the sampling period ΔT is set to 10 seconds. It should be understood that the prediction time domain T P , control time domain T c The sampling period ΔT can be adaptively adjusted according to specific scenarios, and no specific limitation is given to this.
[0126] At the same time, the yaw angle sequence should meet the following constraints:
[0127] The specific optimization steps of the genetic algorithm are as follows:
[0128] Step 1): Initialize the population. Population size N pop is 70, the encoding method is real number encoding, and the initial population is randomly generated using uniform distribution. The gene of the individual represents the yaw action Δγ at each sampling moment.
[0129] Step 2): Convert to the yaw angle of the wind turbine: When executing the nth control command, the yaw angle at the current moment should be equal to the accumulation of all control actions before this moment, that is:
[0130] Step 3): Parental fitness evaluation: Use the power calculation module to calculate the corresponding power generation of each individual in the parent.
[0131] Step 4): Crossover and mutation: Perform crossover and mutation operations on the individual genes of the parent generation to generate new offspring.
[0132] Step 5): Convert to yaw angle of wind turbine: As in step 2), convert the individual genotype of the offspring into yaw angle;
[0133] Step 6): Offspring fitness evaluation: The power calculation module is also used to calculate the wind farm power of the offspring.
[0134] Step 7): Merge the parent and offspring populations to obtain a population size of 2N pop of the combined population.
[0135] Step 8): Select and generate a new population; select N from the merged population in turn pop The individuals with the highest power generation capacity form a new population.
[0136] Step 9): Iterative optimization: Repeat the above steps 2) to 8) until convergence to obtain the control time domain T c The yaw angle action sequence within.
[0137] S4, output control time domain T c The yaw angle command within.
[0138] S5. After the control instruction of the current control time domain is executed, the next new control time domain is entered immediately and the above process is repeated, thereby realizing the real-time rolling optimization control of the dynamic yaw of the wind farm.
[0139] It should be pointed out that although the various steps in the above embodiments are described in a specific order, those skilled in the art can understand that in order to achieve the effect of the present application, different steps do not have to be executed in such an order. They can be executed simultaneously (in parallel) or in other orders. These changes are within the scope of protection of the present application.
[0133] Furthermore, the present application also provides a wind farm wake dynamic yaw control device.
[0134] See attached Figure 5 , Figure 5 It is a main structural block diagram of a wind farm wake dynamic yaw control device according to an embodiment of the present application.
[0135] like Figure 5 As shown, the wind farm wake dynamic yaw control device in the embodiment of the present application mainly includes an acquisition module 11, a first determination module 12, a second determination module 13 and a control module 14. In some embodiments, one or more of the acquisition module 11, the first determination module 12, the second determination module 13 and the control module 14 can be combined into one module.
[0136] In some embodiments, the acquisition module 11 may be configured to acquire environmental information in the wind farm.
[0137] The first determination module 12 may be configured to determine power information of the wind farm based on the environmental information.
[0138] The second determination module 13 may be configured to determine an optimal yaw angle sequence based on power information of the wind farm.
[0139] The control module 14 is configured to perform yaw control on each wind turbine in the wind farm based on the optimal yaw angle sequence.
[0140] In one implementation, the description of the specific implementation functions can refer to step S10 to step S40.
[0141] The above-mentioned wind farm wake dynamic yaw control device is used to perform Figure 1 The embodiment of the wind farm wake dynamic yaw control method shown in the figure has similar technical principles, technical problems solved and technical effects produced. Technicians in this technical field can clearly understand that for the convenience and conciseness of description, the specific working process and related instructions of the wind farm wake dynamic yaw control device can refer to the contents described in the embodiment of the wind farm wake dynamic yaw control method, which will not be repeated here.
[0142] Further, it should be understood that since the setting of each module is only for illustrating the functional units of the device of the present application, the physical devices corresponding to these modules may be the processor itself, or a part of the software in the processor, a part of the hardware, or a part of the combination of software and hardware. Therefore, the number of each module in the figure is only schematic.
[0143] It is understood by those skilled in the art that each module in the device can be adaptively split or merged. Such splitting or merging of specific modules will not cause the technical solution to deviate from the principle of the present application, and therefore, the technical solutions after splitting or merging will fall within the protection scope of the present application.
[0144] It is understood by those skilled in the art that all or part of the processes in the method for implementing the above-mentioned embodiment of the present application can also be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable storage medium may include: any entity or device, medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code.
[0145] Furthermore, the present application also provides a computer device, which may include at least one processor; and a memory connected to the at least one processor; wherein a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method for controlling the dynamic yaw of the wake of a wind farm described in any of the above embodiments is implemented. Figure 6 As shown, Figure 6 exemplarily shows the structure of a computer device, which includes a processor 100 and a memory 200.
[0146] Furthermore, the present application also provides a computer-readable storage medium. In a computer-readable storage medium embodiment according to the present application, the computer-readable storage medium can be configured to store a program for executing the wind farm wake dynamic yaw control method of the above-mentioned method embodiment, and the program can be loaded and run by the processor to implement the above-mentioned wind farm wake dynamic yaw control method. For ease of explanation, only the parts related to the embodiment of the present application are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present application. The computer-readable storage medium can be a memory device formed by various electronic devices. Optionally, the computer-readable storage medium in the embodiment of the present application is a non-temporary computer-readable storage medium.
[0147] So far, the technical solutions of the present application have been described in conjunction with the specific implementations shown in the accompanying drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific implementations. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present application.
Claims
1. A method for dynamic yaw control of wind farm wake, It is characterized in that The method comprises: Acquiring environmental information in the wind farm; determining power information of the wind farm based on the environmental information; determining an optimal yaw angle sequence based on power information of the wind farm; The yaw control is performed on each wind turbine in the wind farm based on the optimal yaw angle sequence.
2. The method for dynamic yaw control of a wind farm wake according to claim 1, It is characterized in that The environmental information includes the inflow wind speed of each wind turbine set at the wind rotor, and the inflow wind speed of each wind turbine set at the wind rotor is obtained by the following steps: Obtain the thrust coefficient, yaw angle and instantaneous inflow wind speed of each wind turbine; Generate a wind speed Taylor background field of the wind farm based on the instantaneous inflow wind speed of the wind farm; determining a first steady-state speed loss based on the thrust coefficient and the yaw angle; The inflow wind speed of each wind turbine set at the wind rotor is determined based on the wind speed Taylor background field and the first steady-state speed loss.
3. The wind farm wake dynamic yaw control method according to claim 2, It is characterized in that The determining a first steady-state speed loss based on the thrust coefficient and the yaw angle comprises: generating a second steady-state speed loss based on the thrust coefficient and the yaw angle; The first steady-state speed loss is determined based on the second steady-state speed loss, the first steady-state speed loss is a steady-state speed loss of multiple units, and the second steady-state speed loss is a steady-state speed loss of each wind turbine unit.
4. The method for dynamic yaw control of a wind farm wake according to claim 2, It is characterized in that The determining the inflow wind speed of each wind turbine set at the wind rotor based on the wind speed Taylor background field and the first steady-state speed loss comprises: Determine the instantaneous wake field of multiple units based on the wind speed Taylor background field and the first steady-state speed loss; The inflow wind speed of each wind turbine set at the wind rotor is obtained based on the instantaneous wake field of the multiple turbine sets.
5. The wind farm wake dynamic yaw control method according to claim 1, It is characterized in that The determining the power information of the wind farm based on the environmental information includes: Determine the instantaneous power of each wind turbine generator set based on the environmental information, where the instantaneous power is the instantaneous power of the wind turbine generator set at any time in the prediction time domain; The power information of the wind farm in the prediction time domain is determined based on the instantaneous power of each wind turbine generator set.
6. The wind farm wake dynamic yaw control method according to claim 1, It is characterized in that The determining of the optimal yaw angle sequence based on the power information of the wind farm comprises: Determining an expression of an optimization target based on power information of the wind farm; Establishing an initial population, wherein the initial population includes a plurality of individuals, each of which is a yaw action sequence of each wind turbine in the wind farm; Starting from the initial population, using a genetic algorithm to perform target optimization iterations to maximize the optimization target; When the optimization target converges, an optimal yaw action sequence of each wind turbine is obtained; An optimal yaw angle sequence of each wind turbine generator set is determined based on the optimal yaw action sequence.
7. The method for dynamic yaw control of a wind farm wake according to claim 6, It is characterized in that The target optimization iteration using genetic algorithm includes: Obtain the constraint conditions of the yaw angle sequence of each wind turbine; Under the constraint condition of the yaw angle sequence of each wind turbine generator set, starting from the initial population, a genetic algorithm is used to perform target optimization iteration.
8. A wind farm wake dynamic yaw control device, It is characterized in that The device comprises: An acquisition module, wherein the acquisition module is configured to acquire environmental information in the wind farm; A first determination module, wherein the first determination module is configured to determine power information of the wind farm based on the environmental information; a second determination module, the second determination module being configured to determine an optimal yaw angle sequence based on power information of the wind farm; A control module is configured to perform yaw control on each wind turbine in the wind farm based on the optimal yaw angle sequence.
9. A computer device, It is characterized in that include: at least one processor; and, a memory communicatively coupled to the at least one processor; Wherein, a computer program is stored in the memory, and when the computer program is executed by the at least one processor, the method for dynamic yaw control of the wake of a wind farm according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium having a plurality of program codes stored therein, It is characterized in that The program code is suitable for being loaded and run by a processor to execute the wind farm wake dynamic yaw control method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Wind power plant wake flow management method and system based on model
CN110397553A
Yaw optimization control method, device equipment and medium by considering wake flow of wind power plant
CN112459965A
Wind power plant yaw cooperative control optimization method, electronic equipment and storage medium
CN113139326A
Wind power plant cooperative yaw control method based on multilayer artificial intelligence system
CN116221021A
Wind power plant layout method and device considering wake flow control, equipment and medium
CN116739313A
Cited By
Wind power plant wake flow scheduling method under wind direction monitoring
CN119593944A
Wind power plant yaw control method, system and device based on layered optimization and medium
CN119755011A
Wind farm yaw control method, system, device and medium based on hierarchical optimization
CN119755011B
Wake flow control method and system for wind turbine generator of wind power plant
CN120140145A
Wind power plant yaw control method and system based on time-of-use electricity price and medium
CN121007088A
