A feedforward unified pitch control method and device based on wind evolution modeling

Abstract

The present invention relates to the technical field of wind turbine pitch control, and discloses a feedforward unified pitch control method and device based on wind evolution modeling. The method comprises: measuring the cross-sectional wind speed at a specified distance from the front end of the wind rotor using a laser radar located at a predetermined position within the wind farm, and determining the wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor; obtaining a wind turbine power curve and cabin feedback data, and determining the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and cabin feedback data; fitting the cross-sectional wind speed at the specified distance from the front end of the wind rotor and the effective wind speed using a preset neural network to obtain a wind evolution model; using the wind evolution model to obtain the wind turbine pitch angle and lead pitch time, and performing feedforward unified pitch control on the wind turbine based on the wind turbine pitch angle and lead pitch time. The present invention achieves accurate pitch control of the wind turbine and optimizes the control effect of the wind turbine pitch.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine pitch control, and in particular to a feedforward unified pitch control method and device based on wind evolution modeling. Background Art

[0002] The capacity of wind power generation systems has developed into the megawatt level, and their operating environment has become increasingly complex. In practical applications, variable-speed and variable-pitch wind turbines often use speed feedback control to control the balance between aerodynamic torque and electromagnetic torque to stabilize the speed. Specifically, the electromagnetic torque or pitch angle is adjusted in different wind speed zones to stabilize the wind rotor speed.

[0003] However, the above-mentioned variable pitch control method of wind turbines ignores the process of wind evolution in actual space. Since the actual wind speed will be affected by factors such as geography, environment and wind turbine wake, it directly leads to a decrease in the accuracy of the control signal, which in turn causes the wind turbine to be prone to overspeed shutdown under wind conditions with high wind speed and large fluctuation range, increasing the component load, which is not conducive to extending the life of the wind turbine. Summary of the Invention

[0004] In view of this, the present invention provides a feedforward unified pitch control method and device based on wind evolution modeling to solve the problem that the pitch control method of wind turbines ignores the process of wind evolution in actual space, resulting in a decrease in the accuracy of the control signal.

[0005] In a first aspect, the present invention provides a feedforward unified pitch control method based on wind evolution modeling, the method comprising:

[0006] The wind speed at a specified distance from the front end of the wind rotor is measured by a laser radar installed at a predetermined position in the wind farm, and the wind speed time series at the cross section of the wind rotor is determined based on the position of the laser radar and the wind speed at the specified distance from the front end of the wind rotor;

[0007] Obtain the wind turbine power curve and cabin feedback data, and determine the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and cabin feedback data;

[0008] The wind evolution model is obtained by fitting the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network;

[0009] The wind evolution model is used to obtain the wind turbine pitch angle and advance pitch time, and feedforward unified pitch control of the wind turbine is performed based on the wind turbine pitch angle and advance pitch time.

[0010] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. The cross-sectional wind speed at a specified distance from the front end of the wind rotor is measured by a laser radar, and the effective wind speed corresponding to the wind speed time series is determined based on the wind turbine power curve and the cabin feedback data, thereby avoiding the problem of inaccurate wind measurement by the cabin anemometer due to the influence of the wake effect. The cross-sectional wind speed and the effective wind speed at a specified distance from the front end of the wind rotor are fitted by a preset neural network, so that the wind evolution model can more accurately fit the wind evolution process in the actual space. In addition, the wind turbine pitch angle and advance pitch time are obtained by using the wind evolution model, and the wind turbine is fed forward and uniformly pitched based on the wind turbine pitch angle and advance pitch time, thereby optimizing the control effect of the wind turbine pitch, having universal applicability to different geographical environments, making the command value of the feedforward pitch controller more accurate, avoiding the stall and shutdown of the wind turbine under extreme wind conditions, reducing the load on the wind turbine components, and extending the life of the wind turbine.

[0011] In an optional embodiment, determining the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and nacelle feedback data includes:

[0012] Determine the wind turbine speed, wind turbine power and wind turbine pitch angle corresponding to the wind speed time series based on the nacelle feedback data;

[0013] Obtain the rotor radius and, based on the turbine speed, turbine power, turbine pitch angle, and rotor radius, obtain the tip speed ratio by comparing it with the turbine power curve.

[0014] The effective wind speed corresponding to the wind speed time series is calculated based on the tip speed ratio, wind turbine speed and rotor radius.

[0015] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. It uses the wind turbine power curve and combines it with the cabin feedback data to reversely infer the effective wind speed at the wind rotor section, avoiding the problem of inaccurate wind measurement caused by the influence of the wake effect, and laying the foundation for the subsequent accurate control of the feedforward unified pitch.

[0016] In an optional embodiment, a wind evolution model is obtained by fitting the cross-sectional wind speed and the effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network, including:

[0017] Normalize the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor;

[0018] The normalized cross-sectional wind speed at a specified distance from the front end of the wind rotor is input into a preset neural network to generate a predicted wind speed at the wind rotor;

[0019] The loss function is calculated based on the predicted wind speed and effective wind speed at the wind rotor, and the preset neural network is iteratively trained based on the loss function to obtain the wind evolution model.

[0020] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling, which normalizes the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor, avoids the adverse effects of abnormal data, and improves the algorithm convergence speed and the prediction accuracy of the preset neural network. By fitting the cross-sectional wind speed at a specified distance from the front end of the wind rotor measured by the lidar and the effective wind speed obtained by inverse calculation through the preset neural network, the evolution process of the horizontal axis wind speed can be fitted more accurately, thereby improving the prediction accuracy of the wind evolution model.

[0021] In an optional embodiment, a wind turbine pitch angle and an advance pitch time are obtained using a wind evolution model, and feedforward unified pitch control of the wind turbine is performed based on the wind turbine pitch angle and the advance pitch time, including:

[0022] The real-time wind speed of the target wind turbine is measured by a lidar installed at a predetermined location within the wind farm. The real-time wind speed is input into the wind evolution model to obtain the current wind speed at the wind rotor.

[0023] Collect target wind turbine parameters, perform pitch control calculation based on the current wind speed at the wind rotor and target wind turbine parameters, and obtain the wind turbine pitch angle;

[0024] Obtaining the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar, and determining the advance pitch time based on the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar;

[0025] The wind turbine pitch angle and advance pitch time are used to perform feedforward unified pitch control on the target wind turbine.

[0026] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. By utilizing a trained wind evolution model, combined with a lidar wind measurement design and a feedforward pitch controller, an accurately superimposed wind turbine pitch angle and a more precise advance pitch time can be obtained, thereby optimizing the control effect of the wind turbine pitch, reducing the fluctuation of the tower load and output power, and extending the life of the wind turbine.

[0027] In an optional embodiment, performing pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameters to obtain the wind turbine pitch angle includes:

[0028] Use the table lookup method to determine the pitch angle corresponding to the current wind speed at the wind rotor;

[0029] Obtain the current pitch angle of the target wind turbine, and calculate the feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle;

[0030] Obtaining a rotor speed and a target rated speed of a target wind turbine, and calculating a pitch angle based on the rotor speed and the rated speed of the target wind turbine;

[0031] The pitch angle and the feedforward superimposed pitch angle are superimposed and calculated to obtain the wind turbine pitch angle.

[0032] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling, which realizes rapid and accurate acquisition of the pitch angle corresponding to the current wind speed at the wind rotor through a table lookup method, and calculates the feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle, calculates the pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and superimposes the pitch angle and the feedforward superimposed pitch angle on each other. On the basis of the wind evolution model's prediction of the current wind speed at the wind rotor, combined with the pitch controller's processing of relevant parameters of the target wind turbine, accurate superimposed calculation of the pitch angle is realized, thereby optimizing the feedforward unified pitch control of the wind turbine.

[0033] In an optional embodiment, the pitch angle is calculated based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, including:

[0034] A speed error signal is determined based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and pitch control is performed based on the speed error signal to obtain a pitch angle.

[0035] In a second aspect, the present invention provides a feedforward unified pitch control device based on wind evolution modeling, the device comprising:

[0036] a measurement module, configured to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor using a laser radar located at a predetermined position within the wind farm, and determine a wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor;

[0037] A determination module is used to obtain the wind turbine power curve and the cabin feedback data, and determine the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and the cabin feedback data;

[0038] A fitting module is used to fit the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model;

[0039] The control module is used to obtain the wind turbine pitch angle and advance pitch time by using the wind evolution model, and perform feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and advance pitch time.

[0040] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, computer instructions stored in the memory, and the processor executing the computer instructions to execute the feedforward unified pitch control method based on wind evolution modeling of the above-mentioned first aspect or any corresponding embodiment thereof.

[0041] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the feedforward unified pitch control method based on wind evolution modeling of the above-mentioned first aspect or any corresponding embodiment thereof.

[0042] In a fifth aspect, the present invention provides a computer program product comprising computer instructions for causing a computer to execute the feedforward unified pitch control method based on wind evolution modeling of the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0044] Figure 1 1 is a flow chart of a feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention;

[0045] Figure 2 is a schematic diagram of laser radar wind measurement according to an embodiment of the present invention;

[0046] Figure 3 is a flow chart of another feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention;

[0047] Figure 4 is a flow chart of another feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention;

[0048] Figure 5 is a flow chart of a wind evolution modeling process according to an embodiment of the present invention;

[0049] Figure 6 is a flow chart of another feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention;

[0050] Figure 7 1 is a schematic diagram of a feedforward pitch control process according to wind evolution according to an embodiment of the present invention;

[0051] Figure 8 is a structural block diagram of a feedforward unified pitch control device based on wind evolution modeling according to an embodiment of the present invention;

[0052] Figure 9Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0054] To address the lag problem in wind turbine pitch control, lidar wind measurement, as a relatively mature technology, has been widely used in the control of large wind turbines. Technicians have also conducted many studies on lidar wind measurement, such as: 1) designing lidar adaptive feedforward control to reduce structural loads caused by turbulence; 2) designing model predictive control algorithms based on lidar data; and 3) improving the accuracy of lidar wind measurement through optimization algorithms to design more accurate feedforward controllers.

[0055] The following methods are used to design feedforward control for wind evolution:

[0056] A method for calculating the evolving wind speed and a feedforward control design based on this method directly adopts a known evolutionary model to model wind evolution. This method can accurately describe the wind speed variation process in the upstream sensing area of ​​the wind rotor, but does not consider the influence of factors such as the geographical environment and is not general.

[0057] A method for predicting the free-inflow wind speed of a wind turbine establishes a nonlinear mapping model by processing the wind speed measured by the nacelle anemometer and the wind speed measured by lidar. While this method is universal, its accuracy is reduced because the wind speed measured by the nacelle anemometer is affected by the turbine wake, resulting in it not being the actual wind speed at the rotor surface.

[0058] A wind turbine yaw control method comprises the following steps: measuring line-of-sight wind speed and direction data at corresponding positions by a laser radar installed on a nacelle or a rotating shaft of the wind turbine; obtaining radar inversion wind speed and direction data based on the line-of-sight wind speed and direction data measured by the laser radar; substituting the inversion wind speed and direction data into a wind evolution model to calculate the predicted current wind information of the wind turbine; and performing wind turbine yaw control based on the calculated evolved wind speed and direction information and the wind direction deviation information of the current nacelle returned by a wind vane. This method also directly adopts a known evolution model, assuming that the evolution law of wind at the same height is the same and that the wind only decays rather than increases during the evolution process, i.e., ignoring the effects of factors such as gusts and terrain.

[0059] The above methods are all based on the Taylor turbulence freezing hypothesis, ignoring the process of wind evolution in actual space, and directly equating the wind speed on the lidar measurement surface with the actual wind speed at the wind rotor surface; the actual wind speed will be affected by factors such as geography, environment and wind turbine wake, which directly leads to a decrease in the accuracy of the control signal.

[0060] An embodiment of the present invention provides a feedforward unified pitch control method based on wind evolution modeling, which is universal to geographical environments and has high accuracy, making the command value of the feedforward pitch controller more accurate, avoiding wind turbine stall and shutdown under various extreme wind conditions, reducing component loads, and extending wind turbine life.

[0061] According to an embodiment of the present invention, an embodiment of a feedforward unified pitch control method based on wind evolution modeling is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0062] In this embodiment, a feedforward unified pitch control method based on wind evolution modeling is provided, which can be used for server-type devices. Figure 1 is a flow chart of a feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0063] In step S101, a laser radar is placed at a predetermined position in the wind farm to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor, and a wind speed time series at the cross section of the wind rotor is determined based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor.

[0064] Specifically, if Figure 2 As shown, the lidar measures the cross-sectional wind speed at a specified distance d of 100 to 150 m from the front end of the wind rotor and a radius r of 0.7 to 0.8 times the radius of the impeller; the coordinates of the lidar position in free space are [-150, 47.25, 47.25], [-150, -47.25, 47.25], [-150, 47.25, -47.25] and [-150, -47.25, -47.25].

[0065] Furthermore, the wind speed time series t (unit: s) at the rotor section can be expressed as follows:

[0066]

[0067] Where x is the position of the laser radar, that is, the position of the measurement point, in meters. is the average wind speed of the cross section, in m / s.

[0068] Step S102 : obtaining a wind turbine power curve and nacelle feedback data, and determining an effective wind speed corresponding to a wind speed time series based on the wind turbine power curve and the nacelle feedback data.

[0069] Specifically, the effective wind speed of the same time series at the rotor section is inferred through the wind turbine power curve combined with the nacelle feedback data.

[0070] Step S103 , fitting the cross-sectional wind speed and the effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model.

[0071] Step S104 , using the wind evolution model to obtain the wind turbine pitch angle and advance pitch time, and performing feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and advance pitch time.

[0072] Specifically, based on the arbitrary wind speed measured by the lidar, combined with the wind speed and time series at the wind rotor obtained by the wind evolution model, the wind turbine pitch angle and the advance pitch time are obtained by using the design of the feedforward pitch controller.

[0073] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. The cross-sectional wind speed at a specified distance from the front end of the wind rotor is measured by a laser radar, and the effective wind speed corresponding to the wind speed time series is determined based on the wind turbine power curve and the cabin feedback data, thereby avoiding the problem of inaccurate wind measurement by the cabin anemometer due to the influence of the wake effect. The cross-sectional wind speed and the effective wind speed at a specified distance from the front end of the wind rotor are fitted by a preset neural network, so that the wind evolution model can more accurately fit the wind evolution process in the actual space. In addition, the wind turbine pitch angle and advance pitch time are obtained by using the wind evolution model, and the wind turbine is fed forward and uniformly pitched based on the wind turbine pitch angle and advance pitch time, thereby optimizing the control effect of the wind turbine pitch, having universal applicability to different geographical environments, making the command value of the feedforward pitch controller more accurate, avoiding the stall and shutdown of the wind turbine under extreme wind conditions, reducing the load on the wind turbine components, and extending the life of the wind turbine.

[0074] In this embodiment, a feedforward unified pitch control method based on wind evolution modeling is provided, which can be used for the above-mentioned server-type devices. Figure 3 is a flow chart of a feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0075] Step S301: Use a laser radar located at a predetermined position in the wind farm to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor, and determine the wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor. Figure 1Step S101 of the illustrated embodiment will not be described in detail here.

[0076] Step S302 : obtaining a wind turbine power curve and nacelle feedback data, and determining an effective wind speed corresponding to a wind speed time series based on the wind turbine power curve and the nacelle feedback data.

[0077] Specifically, the above step S302 includes:

[0078] Step S3021 : determining the wind turbine speed, wind turbine power, and wind turbine pitch angle corresponding to the wind speed time series based on the nacelle feedback data.

[0079] Step S3022: Obtain the rotor radius, and obtain the tip speed ratio based on the wind turbine speed, wind turbine power, wind turbine pitch angle, and wind rotor radius by comparing with the wind turbine power curve.

[0080] Specifically, the fan power P M The calculation formula is as follows:

[0081] P M =0.5ρ(πR2)v 3 C p (2)

[0082] Where ρ represents the air density, R represents the radius of the wind wheel, v represents the effective wind speed, and C p The wind turbine power curve is a three-dimensional curve consisting of the Z axis C p value, X-axis β value and Y-axis lambda value.

[0083] Furthermore, the following formula can be derived from the above formula (2):

[0084] C p / λ 3 =2P M / (ρπω 3 R5) (3)

[0085] Where λ represents the tip speed ratio and ω represents the fan speed.

[0086] Furthermore, the C obtained by the wind turbine pitch angle β and the above formula (3) is P / λ 3 The relationship between p The curve (i.e., the wind turbine power curve) is used to obtain a point in the three-dimensional curve that meets the conditions, and then the lambda value corresponding to the point is used as the tip speed ratio.

[0087] Step S3023: Calculate the effective wind speed corresponding to the wind speed time series based on the tip speed ratio, the wind turbine speed, and the rotor radius.

[0088] Specifically, the calculation formula of effective wind speed v is as follows:

[0089] v=ωR / λ (4)

[0090] Step S303: Fit the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model. Figure 1 Step S103 of the illustrated embodiment will not be described in detail here.

[0091] Step S304: Use the wind evolution model to obtain the wind turbine pitch angle and advance pitch time, and perform feedforward unified pitch control on the wind turbine based on the wind turbine pitch angle and advance pitch time. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0092] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. It uses the wind turbine power curve and combines it with the cabin feedback data to reversely infer the effective wind speed at the wind rotor section, avoiding the problem of inaccurate wind measurement caused by the influence of the wake effect, and laying the foundation for the subsequent accurate control of the feedforward unified pitch.

[0093] In this embodiment, a feedforward unified pitch control method based on wind evolution modeling is provided, which can be used for the above-mentioned server-type devices. Figure 4 is a flow chart of a feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0094] Step S401: Use a laser radar at a predetermined location within the wind farm to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor, and determine the wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor. Figure 3 Step S301 of the illustrated embodiment will not be described in detail here.

[0095] Step S402: Obtain the wind turbine power curve and cabin feedback data, and determine the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and cabin feedback data. Figure 3 Step S302 of the illustrated embodiment will not be described in detail here.

[0096] Step S403: fitting the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model.

[0097] Specifically, the above step S403 includes:

[0098] Step S4031: normalize the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor.

[0099] Specifically, the expression of normalization processing is as follows:

[0100] y=ymin+[(ymax-ymin)*(vi-vmin) / (vmax-vmin)] (5)

[0101] In the above formula, y represents the normalized data (i.e. the normalized cross-sectional wind speed at a specified distance from the front end of the wind rotor, or the normalized effective wind speed), min represents the lower limit of normalization, y min =1,y max represents the normalized upper limit, y max =1,v max and v min Indicates the maximum and minimum wind speed in the original data (i.e., the cross-sectional wind speed or effective wind speed at a specified distance from the front end of the wind rotor), v i Indicates the cross-sectional wind speed or effective wind speed at a specified distance from the front end of the wind rotor.

[0102] Step S4032: Input the normalized cross-sectional wind speed at a specified distance from the front end of the wind rotor into a preset neural network to generate a predicted wind speed at the wind rotor.

[0103] Specifically, if Figure 5 As shown, an artificial neural network (i.e., a preset neural network) is used to fit the cross-sectional wind speed and the effective wind speed at a specified distance from the front end of the wind rotor to obtain the wind evolution process.

[0104] Furthermore, the preset neural network adopts a 2-4 layer BP neural network (a multi-layer feedforward neural network trained according to the error back propagation algorithm), selects 8-12 hidden layer neurons, selects the number of training samples as 600-1000, and uses the normalized cross-sectional wind speed at a specified distance from the front end of the wind wheel as input data.

[0105] Step S4033: A loss function is calculated based on the predicted wind speed and the effective wind speed at the wind rotor, and a preset neural network is iteratively trained based on the loss function to obtain a wind evolution model.

[0106] Specifically, the mean square error is selected as the loss function, and the training data (i.e., the cross-sectional wind speed at a specified distance from the front end of the wind rotor after normalization) is input into the preset neural network for training. The loss function is minimized by adjusting the model parameters. Finally, the wind speed-time series after fitting the preset neural network (i.e., the predicted wind speed at the wind rotor) is compared with the effective wind speed to analyze the accuracy and reliability of the wind evolution model.

[0107] Step S404: Use the wind evolution model to obtain the wind turbine pitch angle and advance pitch time, and perform feedforward unified pitch control on the wind turbine based on the wind turbine pitch angle and advance pitch time. Figure 3 Step S304 of the illustrated embodiment will not be described in detail here.

[0108] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling, which normalizes the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor, avoids the adverse effects of abnormal data, and improves the algorithm convergence speed and the prediction accuracy of the preset neural network. By fitting the cross-sectional wind speed at a specified distance from the front end of the wind rotor measured by the lidar and the effective wind speed obtained by inverse calculation through the preset neural network, the evolution process of the horizontal axis wind speed can be fitted more accurately, thereby improving the prediction accuracy of the wind evolution model.

[0109] In this embodiment, a feedforward unified pitch control method based on wind evolution modeling is provided, which can be used for the above-mentioned server-type devices. Figure 6 is a flow chart of a feedforward unified pitch control method based on wind evolution modeling according to an embodiment of the present invention. Figure 6 As shown, the process includes the following steps:

[0110] Step S601: Use a laser radar at a predetermined location within the wind farm to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor, and determine the wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor. Figure 4 Step S401 of the illustrated embodiment will not be described in detail here.

[0111] Step S602: Obtain the wind turbine power curve and cabin feedback data, and determine the effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and cabin feedback data. Figure 4 Step S402 of the illustrated embodiment will not be described in detail here.

[0112] Step S603: Fit the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model. Figure 4 Step S404 of the illustrated embodiment will not be described in detail here.

[0113] Step S604 , using the wind evolution model to obtain the wind turbine pitch angle and advance pitch time, and performing feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and advance pitch time.

[0114] Specifically, the above step S604 includes:

[0115] Step S6041: measure the real-time wind speed of the target wind turbine by using a laser radar installed at a predetermined position in the wind farm, and input the real-time wind speed into the wind evolution model to obtain the current wind speed at the wind rotor.

[0116] Step S6042: Collect target wind turbine parameters, perform pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameters, and obtain the wind turbine pitch angle.

[0117] In some optional embodiments, such as Figure 7 As shown, the above step S6042 includes:

[0118] Step a1: Use a table lookup method to determine the pitch angle corresponding to the current wind speed at the wind rotor.

[0119] Specifically, the current wind rotor wind speed is fitted with the wind speed-pitch angle curve of the wind turbine in a steady state to obtain the pitch angle corresponding to the current wind rotor wind speed.

[0120] Step a2: obtaining the current pitch angle of the target wind turbine, and calculating the feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle.

[0121] Specifically, the calculation formula of the feedforward superimposed pitch angle is as follows:

[0122] β FF =β L -β msr (6)

[0123] Among them, β FF represents the feedforward superimposed pitch angle, β L Indicates the pitch angle corresponding to the current wind speed at the wind rotor, β msr Indicates the current pitch angle.

[0124] Step a3: Obtain the rotor speed of the target wind turbine and the target rated speed of the wind turbine, and calculate the pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine.

[0125] Specifically, a speed error signal is determined based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and pitch control is performed based on the speed error signal to obtain a pitch angle.

[0126] Furthermore, the rotor speed of the target wind turbine ω Lss and rated speed of fan ω ref The speed error signal is input into the pitch controller to output the pitch angle β. FB .

[0127] Step a4: superimpose the pitch angle and the feedforward superimposed pitch angle to obtain the wind turbine pitch angle.

[0128] Specifically, the pitch angle β FB and the feedforward superimposed pitch angle β FF Perform superposition calculation to obtain the wind turbine pitch angle, that is, the wind turbine pitch angle β.

[0129] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling, which realizes rapid and accurate acquisition of the pitch angle corresponding to the current wind speed at the wind rotor through a table lookup method, and calculates the feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle, calculates the pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and superimposes the pitch angle and the feedforward superimposed pitch angle on each other. On the basis of the wind evolution model's prediction of the current wind speed at the wind rotor, combined with the pitch controller's processing of relevant parameters of the target wind turbine, accurate superimposed calculation of the pitch angle is realized, thereby optimizing the feedforward unified pitch control of the wind turbine.

[0130] Step S6043: Acquire the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar, and determine the advance pitch time based on the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar.

[0131] Specifically, the wind turbine pitch actuator is replaced by a first-order inertia link, and its inertia time constant is used as the pitch mechanism execution time τ1. The sampling time of the lidar is T scan .

[0132] Step S6044: Perform feedforward unified pitch control on the target wind turbine using the wind turbine pitch angle and advance pitch time.

[0133] Specifically, the wind turbine pitch angle and advance pitch time are input into the pitch actuator, and the pitch actuator performs feedforward unified pitch control on the wind turbine.

[0134] This embodiment provides a feedforward unified pitch control method based on wind evolution modeling. By utilizing a trained wind evolution model, combined with a lidar wind measurement design and a feedforward pitch controller, an accurately superimposed wind turbine pitch angle and a more precise advance pitch time can be obtained, thereby optimizing the control effect of the wind turbine pitch, reducing the fluctuation of the tower load and output power, and extending the life of the wind turbine.

[0135] In this embodiment, a feedforward unified pitch control device based on wind evolution modeling is also provided, which is used to implement the above-mentioned embodiments and preferred embodiments. The details that have been described will not be repeated here. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

[0136] This embodiment provides a feedforward unified pitch control device based on wind evolution modeling, such as Figure 8 As shown, including:

[0137] The measurement module 801 is configured to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor using a laser radar installed at a predetermined position in the wind farm, and determine a wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor;

[0138] A determination module 802 is configured to obtain a wind turbine power curve and nacelle feedback data, and determine an effective wind speed corresponding to a wind speed time series based on the wind turbine power curve and nacelle feedback data;

[0139] The fitting module 803 is used to fit the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor through a preset neural network to obtain a wind evolution model;

[0140] The control module 804 is used to obtain the wind turbine pitch angle and advance pitch time using the wind evolution model, and perform feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and advance pitch time.

[0141] In some optional implementations, the determining module 802 includes:

[0142] A first determining unit is configured to determine the wind turbine speed, wind turbine power, and wind turbine pitch angle corresponding to the wind speed time series based on the nacelle feedback data;

[0143] The second determining unit is configured to obtain a rotor radius by comparing the turbine power curve with the turbine speed, turbine power, turbine pitch angle, and rotor radius to obtain a tip speed ratio;

[0144] The first calculation unit is used to calculate the effective wind speed corresponding to the wind speed time series based on the tip speed ratio, the wind turbine speed and the wind rotor radius.

[0145] In some optional implementations, the fitting module 803 includes:

[0146] A normalization processing unit, used to normalize the cross-sectional wind speed and effective wind speed at a specified distance from the front end of the wind rotor;

[0147] A generating unit is used to input the normalized cross-sectional wind speed at a specified distance from the front end of the wind rotor into a preset neural network to generate a predicted wind speed at the wind rotor;

[0148] The training unit is used to calculate the loss function based on the predicted wind speed and the effective wind speed at the wind rotor, and iteratively train the preset neural network based on the loss function to obtain a wind evolution model.

[0149] In some optional implementations, the control module 804 includes:

[0150] A measurement unit is used to measure the real-time wind speed of the target wind turbine using a laser radar located at a predetermined position in the wind farm, and input the real-time wind speed into the wind evolution model to obtain the current wind speed at the wind rotor;

[0151] The second calculation unit is used to collect target wind turbine parameters, perform pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameters, and obtain the wind turbine pitch angle;

[0152] a third determining unit, configured to obtain a dynamic time of the wind turbine pitch actuator and a sampling time of the laser radar, and determine an advance pitch time based on the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar;

[0153] The control unit is used to perform feedforward unified pitch control on the target wind turbine using the wind turbine pitch angle and advance pitch time.

[0154] In some optional implementations, the second computing unit includes:

[0155] A determination subunit, configured to determine a pitch angle corresponding to the current wind speed at the wind rotor by using a table lookup method;

[0156] A first calculation subunit is configured to obtain a current pitch angle of the target wind turbine, and calculate a feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle;

[0157] a second calculation subunit, configured to obtain a rotor speed of a target wind turbine and a target rated speed of the wind turbine, and calculate a pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine;

[0158] The third calculation subunit is used to perform superposition calculation on the pitch angle and the feedforward superimposed pitch angle to obtain the wind turbine pitch angle.

[0159] In some optional embodiments, the second calculation subunit is specifically configured to determine a speed error signal based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and perform pitch control based on the speed error signal to obtain a pitch angle.

[0160] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0161] In this embodiment, a feedforward unified pitch control device based on wind evolution modeling is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0162] The embodiment of the present invention also provides a computer device having the above Figure 8 A feedforward unified pitch control device based on wind evolution modeling is shown.

[0163] See also Figure 9 , Figure 9 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 9 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 9 A processor 10 is taken as an example.

[0164] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0165] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0166] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0167] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0168] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0169] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0170] A portion of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0171] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A feedforward unified pitch control method based on wind evolution modeling, characterized in that: The method comprises: Measuring the cross-sectional wind speed at a specified distance from the front end of the wind rotor by a laser radar installed at a predetermined position in the wind farm, and determining the wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor; Obtaining a wind turbine power curve and nacelle feedback data, and determining an effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and nacelle feedback data; Fitting the cross-sectional wind speed at a specified distance from the front end of the wind rotor and the effective wind speed through a preset neural network to obtain a wind evolution model; Using the wind evolution model to obtain a wind turbine pitch angle and an advance pitch time, and performing feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and the advance pitch time; The method of using the wind evolution model to obtain a wind turbine pitch angle and an advance pitch time, and performing feedforward unified pitch control on the wind turbine generator set based on the wind turbine pitch angle and the advance pitch time, includes: Measuring the real-time wind speed of the target wind turbine by using a laser radar located at a predetermined position in the wind farm, and inputting the real-time wind speed into the wind evolution model to obtain the current wind speed at the wind rotor; Collect target wind turbine parameters, perform pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameters, and obtain the wind turbine pitch angle; Acquire the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar, and determine the advance pitch time based on the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar; Performing feedforward unified pitch control on a target wind turbine using the wind turbine pitch angle and the advance pitch time; The performing pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameter to obtain the wind turbine pitch angle includes: Determine the pitch angle corresponding to the current wind speed at the wind rotor by using a table lookup method; Obtaining a current pitch angle of the target wind turbine, and calculating a feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle; Obtaining a rotor speed of a target wind turbine and a target rated speed of the wind turbine, and calculating a pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine; The pitch angle and the feedforward superimposed pitch angle are superimposed and calculated to obtain the wind turbine pitch angle.

2. The method according to claim 1, characterized in that The determining, based on the wind turbine power curve and nacelle feedback data, the effective wind speed corresponding to the wind speed time series includes: Determining the wind turbine speed, wind turbine power, and wind turbine pitch angle corresponding to the wind speed time series based on the nacelle feedback data; Obtaining a rotor radius, and obtaining a tip speed ratio based on the wind turbine speed, the wind turbine power, the wind turbine pitch angle, and the wind rotor radius, and comparing the obtained tip speed ratio with the wind turbine power curve; The effective wind speed corresponding to the wind speed time series is calculated based on the tip speed ratio, the wind turbine speed and the wind rotor radius.

3. The method according to claim 1, characterized in that The wind evolution model is obtained by fitting the cross-sectional wind speed at a specified distance from the front end of the wind rotor and the effective wind speed through a preset neural network, including: Normalizing the cross-sectional wind speed at a specified distance from the front end of the wind rotor and the effective wind speed; Inputting the normalized cross-sectional wind speed at a specified distance from the front end of the wind rotor into the preset neural network to generate a predicted wind speed at the wind rotor; A loss function is calculated based on the predicted wind speed at the wind rotor and the effective wind speed, and the preset neural network is iteratively trained based on the loss function to obtain the wind evolution model.

4. The method according to claim 1, wherein The calculating of the pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine includes: A speed error signal is determined based on the rotor speed of the target wind turbine and the rated speed of the wind turbine, and pitch control is performed based on the speed error signal to obtain the pitch angle.

5. A feedforward unified pitch control device based on wind evolution modeling, characterized in that: The device comprises: a measurement module, configured to measure the cross-sectional wind speed at a specified distance from the front end of the wind rotor using a laser radar located at a predetermined position within the wind farm, and determine a wind speed time series at the cross section of the wind rotor based on the position of the laser radar and the cross-sectional wind speed at the specified distance from the front end of the wind rotor; a determination module, configured to obtain a wind turbine power curve and nacelle feedback data, and determine an effective wind speed corresponding to the wind speed time series based on the wind turbine power curve and nacelle feedback data; A fitting module, configured to fit the cross-sectional wind speed at a specified distance from the front end of the wind rotor and the effective wind speed through a preset neural network to obtain a wind evolution model; a control module, configured to obtain a wind turbine pitch angle and an advance pitch time by using the wind evolution model, and perform feedforward unified pitch control on the wind turbine set based on the wind turbine pitch angle and the advance pitch time; The control module includes: A measurement unit is used to measure the real-time wind speed of the target wind turbine using a laser radar located at a predetermined position in the wind farm, and input the real-time wind speed into the wind evolution model to obtain the current wind speed at the wind rotor; The second calculation unit is used to collect target wind turbine parameters, perform pitch control calculation on the current wind speed at the wind rotor and the target wind turbine parameters, and obtain the wind turbine pitch angle; a third determining unit, configured to obtain a dynamic time of the wind turbine pitch actuator and a sampling time of the laser radar, and determine an advance pitch time based on the dynamic time of the wind turbine pitch actuator and the sampling time of the laser radar; A control unit for performing feedforward unified pitch control on a target wind turbine using a wind turbine pitch angle and an advance pitch time; The second computing unit includes: A determination subunit, configured to determine a pitch angle corresponding to the current wind speed at the wind rotor by using a table lookup method; A first calculation subunit is configured to obtain a current pitch angle of the target wind turbine, and calculate a feedforward superimposed pitch angle based on the pitch angle corresponding to the current wind speed at the wind rotor and the current pitch angle; a second calculation subunit, configured to obtain a rotor speed of a target wind turbine and a target rated speed of the wind turbine, and calculate a pitch angle based on the rotor speed of the target wind turbine and the rated speed of the wind turbine; The third calculation subunit is used to perform superposition calculation on the pitch angle and the feedforward superimposed pitch angle to obtain the wind turbine pitch angle.

6. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the feedforward unified pitch control method based on wind evolution modeling according to any one of claims 1 to 4 by executing the computer instructions.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the feedforward unified pitch control method based on wind evolution modeling according to any one of claims 1 to 4.

8. A computer program product, characterized in that The method comprises computer instructions, wherein the computer instructions are used to enable a computer to execute the feedforward unified pitch control method based on wind evolution modeling according to any one of claims 1 to 4.

Images