A method and terminal for calculating the output of multiple offshore wind farms under the influence of typhoons

Abstract

The present invention discloses a method and terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon. The method predicts the range of the typhoon track by training the acquired historical typhoon track data, thereby simulating the uncertainty of the typhoon track. According to the predicted range of the typhoon track, a vortex wind field model is used to obtain the wind speed scene around it, and a wind speed-wind turbine output physical model is established in combination with the wind speed scene, environmental parameters and wind turbine parameters of the wind farm. Then, the output scenes corresponding to multiple wind farms are calculated according to the wind speed-wind turbine output physical model. In this way, when a typhoon disaster occurs in the working environment of multiple offshore wind farms, the changes brought about by the uncertainty of the typhoon track to the wind power output prediction can be taken into account, which is conducive to the coordinated operation of various types of power sources under typhoon disasters, taking into account both the utilization of wind resources and the safety of power grid operation.

Description

Technical Field

[0001] The present invention relates to the technical field of offshore wind farm group output calculation, and in particular to an output calculation method and a terminal for multiple offshore wind farm groups under the influence of a typhoon. Background Art

[0002] With the continuous improvement of offshore wind turbine manufacturing and grid-connected technology, offshore wind power has received more and more attention. Compared with onshore wind power, offshore wind power has the advantages of high output level, small output fluctuation, stable dominant wind direction, small wind shear, no occupation of land resources and less impact on the environment.

[0003] However, compared with conventional power sources, the access of a high proportion of wind power will bring many uncertainties to the operation and control of the power system, especially under the influence of extreme weather such as typhoons, which may cause large and rapid changes in wind power output, affecting the safe and stable operation of the power system. It is even more noteworthy that the typhoon track is often difficult to accurately predict, and there is also a corresponding uncertainty in the impact on the output characteristics of offshore wind power around the power grid in a specific area, which makes it more difficult for the power grid to respond to the risk of wind power fluctuations in a refined manner. At present, there is a lack of an analysis method for offshore wind power output under the influence of typhoons that comprehensively considers the uncertainty of typhoon tracks and multiple wind power clusters. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon, which can take into account the changes in wind power output prediction caused by the uncertainty of the typhoon, and thus help to coordinate multiple types of power sources under typhoon disasters and ensure the safe operation of the power grid.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for calculating the output of multiple offshore wind farms under the influence of a typhoon, comprising the steps of:

[0007] Acquire historical typhoon track data, train the historical typhoon track data and predict the typhoon track range;

[0008] Different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model;

[0009] A wind speed-wind turbine output physical model is established in combination with wind speed scenarios of offshore wind farms as well as environmental parameters and wind turbine parameters of offshore wind farms. Wind power output scenarios corresponding to multiple wind farms are calculated based on the wind speed-wind turbine output physical model.

[0010] In order to solve the above technical problems, another technical solution adopted by the present invention is:

[0011] A terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the various steps of the above-mentioned method for calculating the output of multiple offshore wind farms under the influence of a typhoon when executing the computer program; the terminal also comprises a human-machine interface module for inputting environmental parameters and wind turbine parameters of the offshore wind farm, and for outputting the calculated wind power output scenarios corresponding to the multiple wind farms.

[0012] The beneficial effects of the present invention are as follows: by training the acquired historical typhoon track data, the typhoon track range is predicted, thereby simulating the uncertainty of the typhoon track; according to the predicted typhoon track range, the vortex wind field model is used to obtain the wind speed scene around it, and the wind speed scene, environmental parameters and wind turbine parameters of the wind farm are combined to establish a wind speed-wind turbine output physical model, and then the wind power output scenes corresponding to multiple wind farms are calculated according to the wind speed-wind turbine output physical model. In this way, when a typhoon disaster occurs in the working environment of multiple offshore wind farms, the changes brought about by the uncertainty of the typhoon track to the wind power output prediction can be taken into account, which is conducive to the coordinated operation of various types of power sources under typhoon disasters, taking into account the utilization of wind resources and the safety of power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a flow chart of a method for calculating the output of multiple offshore wind farms under the influence of a typhoon according to an embodiment of the present invention;

[0014] Figure 2 This is a schematic diagram of an output calculation terminal for multiple offshore wind farms under the influence of a typhoon according to an embodiment of the present invention;

[0015] Figure 3 A flowchart of specific steps of a method for calculating the output of multiple offshore wind farms under the influence of a typhoon according to an embodiment of the present invention;

[0016] Figure 4 A prediction curve diagram of the longitude of a typhoon track by a two-dimensional Gaussian process regression model according to an embodiment of the present invention;

[0017] Figure 5 A prediction curve diagram of the typhoon track latitude by a two-dimensional Gaussian process regression model according to an embodiment of the present invention;

[0018] Figure 6 A display diagram of the actual typhoon track, the average value of the predicted track, and the predicted track range according to an embodiment of the present invention;

[0019] Figure 7 This is a wind speed scenario of wind farm a considering the uncertainty of typhoon track in an embodiment of the present invention;

[0020] Figure 8 This is a wind speed scenario of wind farm b considering the uncertainty of typhoon track in an embodiment of the present invention;

[0021] Fig. 9 The wind power output scenario of wind farm a derived from the wind speed-wind power output physical model according to the embodiment of the present invention;

[0022] Fig.10 This is a wind power output scenario of wind farm b derived from a wind speed-wind power output physical model according to an embodiment of the present invention.

[0023] Description of labels:

[0024] 1. A terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon; 2. A memory; 3. A processor. DETAILED DESCRIPTION

[0025] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0026] Please refer to Figure 1 The embodiment of the present invention provides a method for calculating the output of multiple offshore wind farms under the influence of a typhoon, comprising the steps of:

[0027] Acquire historical typhoon track data, train the historical typhoon track data and predict the typhoon track range;

[0028] Different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model;

[0029] A wind speed-wind turbine output physical model is established in combination with wind speed scenarios of offshore wind farms as well as environmental parameters and wind turbine parameters of offshore wind farms. Wind power output scenarios corresponding to multiple wind farms are calculated based on the wind speed-wind turbine output physical model.

[0030] From the above description, it can be seen that the beneficial effects of the present invention are: by training the acquired historical typhoon track data, the typhoon track range is predicted, thereby simulating the uncertainty of the typhoon track; according to the predicted typhoon track range, the vortex wind field model is used to obtain the wind speed scene around it, and the wind speed scene, environmental parameters and wind turbine parameters of the wind farm are combined to establish a wind speed-wind turbine output physical model, and the wind power output scenes corresponding to multiple wind farms are calculated according to the wind speed-wind turbine output physical model. In this way, when a typhoon disaster occurs in the working environment of multiple offshore wind farms, the changes in the wind power output prediction caused by the uncertainty of the typhoon track can be taken into account, which is conducive to the coordinated operation of various types of power sources under typhoon disasters.

[0031] Furthermore, the historical typhoon track data is trained and the typhoon track range is predicted, including:

[0032] Gaussian kernel training is performed on the historical typhoon track data, and the typhoon track range is predicted through two-dimensional Gaussian process regression, wherein the Gaussian kernel includes an RBF kernel function and an Index kernel function.

[0033] Furthermore, the historical typhoon track data is trained with a Gaussian kernel, and the typhoon track range is predicted by two-dimensional Gaussian process regression, including:

[0034]

[0035]

[0036]

[0037]

[0038] K(X,X')=K inputs (X,X'*K tasks (i,i');

[0039]

[0040] K tasks (i,i')=(BB T +diag(v)) i,i' ;

[0041] In the formula, X represents the two-dimensional input variable of the training sample, which is two groups of time points with equal step length, with a dimension of [2×t], and t represents the number of time points; y represents the two-dimensional output variable of the training sample, which is the longitude and latitude of the typhoon track range;

[0042] K(X,X') represents the covariance matrix of the input variable X and the test point X'. The covariance matrix consists of two parts: the covariance matrix K that considers the correlation between them. inputs (X,X') and the covariance matrix K considering the correlation of the output variables tasks (i,i'), i and i' are the labels corresponding to their output variables;

[0043] represents the variance of Gaussian noise that is independent and identically distributed with the training samples, I represents the identity matrix, f(X') represents the predicted output of the test point X', represents the predicted mean, σ 2 (X') represents the prediction variance, K inputs represents the RBF kernel function, K tasksrepresents the Index kernel function, θ, B and v represent the hyperparameters of the kernel function.

[0044] From the above description, we can see that the traditional Gaussian process regression model only considers the correlation between input variables, while the selected two-dimensional Gaussian process regression model makes up for this deficiency compared to the traditional Gaussian process regression model. The two-dimensional Gaussian process regression model is used to consider the correlation between each output to deal with multi-output problems.

[0045] Furthermore, the different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model include:

[0046]

[0047] Where V i represents the near-ground wind speed at point i during a typhoon, r i represents the distance from point i to the typhoon center, V max Indicates the maximum wind speed inside the horizontal structure of the typhoon, R max Indicates the maximum wind speed radius of a typhoon;

[0048] R max =80-k(950-P c );

[0049] Where P c represents the central air pressure of the typhoon, and k represents the model coefficient of the preset maximum wind speed radius.

[0050] Furthermore, a wind speed-wind turbine output physical model is established by combining the wind speed scenario of the offshore wind farm and the environmental parameters and wind turbine parameters of the offshore wind farm, including:

[0051]

[0052] In the formula, p w Indicates the fan output, p N represents the rated power of the fan, ρ represents the air density, R represents the blade radius of the fan, v represents the wind speed, v N Indicates the rated wind speed corresponding to the rated power of the fan, v in and v out are the cut-in and cut-out wind speeds of the wind turbine, respectively, and K represents the wind force coefficient;

[0053] Where, wind speed v is the wind speed v captured by the wind turbine h , which is related to the near-ground wind speed V i The relationship is:

[0054]

[0055] In the formula, v hrepresents the wind speed captured by the wind turbine, h represents the hub height, m c Represents the friction coefficient.

[0056] From the above description, it can be seen that by combining wind speed scenarios and different parameters of different offshore wind farms, a wind speed-wind turbine output physical model is established. Compared with the conventional wind turbine active output model, the refined wind speed measurement during typhoons is taken into account. In this way, the impact of typhoons on wind power output can be more accurately restored or predicted, thereby improving the accuracy of output prediction.

[0057] Please refer to Figure 2 Another embodiment of the present invention provides an output calculation terminal for multiple offshore wind farms under the influence of a typhoon, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the various steps of the above-mentioned method for calculating the output of multiple offshore wind farms under the influence of a typhoon when executing the computer program; the terminal also includes a human-machine interface module for inputting environmental parameters and wind turbine parameters of the offshore wind farm, and for outputting the calculated wind power output scenarios corresponding to the multiple wind farms.

[0058] The above-mentioned method and terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon in the present invention are applicable to the case where a typhoon disaster occurs in the working environment of multiple offshore wind farms. The uncertainty of the typhoon trajectory can be taken into account to change the wind power output prediction, thereby facilitating the coordinated operation of various types of power sources under typhoon disasters. The following is an explanation through specific implementation methods:

[0059] Embodiment 1

[0060] Please refer to Figure 1 and Figure 3 A method for calculating the output of multiple offshore wind farms under the influence of a typhoon comprises the following steps:

[0061] S1. Obtain historical typhoon track data, train the historical typhoon track data and predict the typhoon track range.

[0062] Gaussian kernel training is performed on the historical typhoon track data, and the typhoon track range is predicted through two-dimensional Gaussian process regression, wherein the Gaussian kernel includes an RBF kernel function and an Index kernel function.

[0063] Among them, the advantage of using a two-dimensional Gaussian process regression model is that the traditional Gaussian process regression model only considers the correlation between input variables, while the two-dimensional Gaussian process regression model makes up for this deficiency. It handles multi-output problems by considering the correlation between each output. The model formula is as follows:

[0064]

[0065]

[0066]

[0067]

[0068] K(X,X')=K inputs (X,X')*K tasks (i,i');

[0069]

[0070] K tasks (i,i')=(BB T +diag(v)) i,i' ;

[0071] The above formula includes the joint prior distribution of the output y of the training sample and the target output f(X') corresponding to the test point X', as well as the posterior probability distribution of f(X'). In the formula of the prior distribution, the training of the Gaussian kernel by the historical typhoon track is mainly the training of the hyperparameters of K; in the formula of the posterior probability distribution, there are two main features, namely the mean function and the covariance matrix. The mean function refers to the predicted mean of the longitude and latitude of the typhoon track.

[0072] Wherein, X represents the two-dimensional input variable of the training sample, which refers to time here, which is two groups of time points with equal step length, and the dimension is [2×t], t represents the number of time points. If the input variable of the training sample in this embodiment is 86 time points evenly divided from 0 to 1, the size of the corresponding input variable is 2×86; y represents the two-dimensional output variable of the training sample, which refers to the longitude and latitude of the typhoon track range here; each longitude and latitude corresponds to a time point.

[0073] K(X,X') represents the covariance matrix of the input variable X and the test point X'. The covariance matrix consists of two parts: the covariance matrix K that considers the correlation between them. inputs (X,X') and the covariance matrix K considering the correlation of the output variables tasks (i,i'), i and i' are the labels of the output variables, the label of longitude is 0, and the label of dimension is 1; the same is true for K(X,X) and K(X',X').

[0074] represents the variance of Gaussian noise that is independent and identically distributed with the training samples, I represents the identity matrix, f(X') represents the predicted output of the test point X', represents the predicted mean, σ 2 (X') represents the prediction variance, K inputs represents the RBF kernel function, K tasksrepresents the Index kernel function, θ, B and v represent the hyperparameters of the kernel function.

[0075] Please refer to Figure 4 and Figure 5 , are respectively the prediction curves of the longitude and latitude of the typhoon trajectory through the two-dimensional Gaussian process regression model, that is, the relationship between the longitude and time of the typhoon position and the relationship between the latitude and time. The scattered points are the actual typhoon trajectories, the solid line is the mean of the typhoon prediction trajectory, and the boundary line of the shaded area is the range of the typhoon prediction trajectory. Figure 6 This is a graph showing the actual typhoon track, the average predicted track, and the predicted track range.

[0076] S2. Different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model.

[0077]

[0078] Where V i represents the near-ground wind speed at point i during a typhoon, r i represents the distance from point i to the typhoon center, V max Indicates the maximum wind speed inside the horizontal structure of the typhoon, R max Indicates the maximum wind speed radius of a typhoon.

[0079] In the above model, key information such as the longitude and latitude of the typhoon center, maximum wind speed, and central air pressure can be obtained from the meteorological information released by the meteorological department, but the maximum wind speed radius is usually difficult to observe directly, so the following model is used for identification:

[0080] R max =80-k(950-P c );

[0081] Where P c It represents the central pressure of the typhoon, and k represents the model coefficient of the preset maximum wind speed radius, which is generally taken as 0.769.

[0082] Please refer to Figure 7 , which is the wind speed scenario of wind farm a considering the uncertainty of typhoon trajectory. The three curves correspond to the mean and upper and lower limits of the typhoon prediction trajectory. Taking the mean of typhoon trajectory as an example, it can be seen that it is divided into two stages: the first stage is the wind speed increase, and the second stage is the wind speed decrease. This is because wind farm a is always located in the typhoon maximum wind speed radius R max In areas outside the area, the closer the typhoon is to the wind farm, the higher the wind speed.

[0083] Please refer to Figure 8, which is the wind speed scenario of wind farm b considering the uncertainty of typhoon track. Taking the mean of typhoon track as an example, it can be seen that it is divided into three stages: the first stage is the increase of wind speed, because wind farm b is located in the maximum wind speed radius R of the typhoon at this time max In the area outside the typhoon, the closer the typhoon is, the higher the wind speed is; in the second stage, the wind speed first decreases and then increases. This is because as the distance from the typhoon center decreases, wind farm b is in the typhoon maximum wind speed radius R max In the area within the typhoon radius, the closer the typhoon is, the lower the wind speed is; in the third stage, wind farm b is again located in the typhoon maximum wind speed radius R max In areas outside the typhoon zone, wind speed decreases as the typhoon moves away.

[0084] S3. A wind speed-wind turbine output physical model is established in combination with the wind speed scenario of the offshore wind farm as well as the environmental parameters and wind turbine parameters of the offshore wind farm. Wind power output scenarios corresponding to multiple wind farms are calculated based on the wind speed-wind turbine output physical model.

[0085] Specifically, the difference between the wind speed-wind power output physical model and the wind turbine active output model is that the wind speed, an important factor affecting wind power output, is taken into account. The wind speed-wind power output physical model can more accurately restore or predict the impact of extreme weather (such as typhoons) on wind speed and wind power output. The physical model is as follows:

[0086]

[0087] In the formula, p w Indicates the fan output, p N represents the rated power of the fan, ρ represents the air density, R represents the blade radius of the fan, v represents the wind speed, v N Indicates the rated wind speed corresponding to the rated power of the fan, v in and v out They are the cut-in and cut-out wind speeds of the wind turbine, and K represents the wind force coefficient, which is generally between 0.2 and 0.6. Among them, the parameter information of the wind turbine can be obtained according to the wind turbine model in the wind farm. For example, the wind turbine model of a wind farm is H210-10, with a hub height of 112m, a blade diameter of 210m, a rated power of 10MW, a rated wind speed of 10.7m / s, and a cut-in and cut-out wind speed of 3.5m / s and 28m / s respectively.

[0088] Where, wind speed v is equal to the wind speed v captured by the wind turbine h , which is related to the near-ground wind speed V i The relationship is:

[0089]

[0090] In the formula, v hrepresents the wind speed captured by the wind turbine, h represents the hub height, m c Represents the friction coefficient.

[0091] Please refer to Fig. 9 and Fig.10 , which is the wind power output scenario of wind farms a and b derived based on the wind speed-wind power output physical model. Both wind farms are in a state of full wind power output as the typhoon moves. When the wind speed exceeds the cut-out wind speed of the wind turbine, the wind power output will drop sharply to 0.

[0092] Embodiment 2

[0093] Please refer to Figure 2 A terminal 1 for calculating the output of multiple offshore wind farms under the influence of a typhoon comprises a memory 2, a processor 3 and a computer program stored in the memory 2 and executable on the processor 3. When the processor 3 executes the computer program, each step of a method for calculating the output of multiple offshore wind farms under the influence of a typhoon in the first embodiment is implemented; the terminal further comprises a human-machine interface module 4 for inputting environmental parameters and wind turbine parameters of the offshore wind farm, and for outputting the calculated wind power output scenarios corresponding to the multiple wind farms.

[0094] In summary, the present invention provides a method and terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon. By training the acquired historical typhoon trajectory data, the typhoon trajectory range is predicted, thereby simulating the uncertainty of the typhoon trajectory; according to the predicted typhoon trajectory range, the vortex wind field model is used to obtain the wind speed scene around it, and the wind speed scene, environmental parameters and wind turbine parameters of the wind farm are combined to establish a wind speed-wind turbine output physical model. After that, the wind power output scenes corresponding to multiple wind farms can be calculated according to the wind speed-wind turbine output physical model. In this way, when a typhoon disaster occurs in the working environment of multiple offshore wind farms, the changes in the wind power output prediction caused by the uncertainty of the typhoon trajectory can be taken into account, which is conducive to the coordinated operation of various types of power sources under typhoon disasters.

[0095] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for calculating the output of multiple offshore wind farms under the influence of a typhoon, characterized in that: Includes steps: Acquire historical typhoon track data, train the historical typhoon track data and predict the typhoon track range; Different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model; A wind speed-wind turbine output physical model is established in combination with wind speed scenarios of offshore wind farms and environmental parameters and wind turbine parameters of offshore wind farms, and wind power output scenarios corresponding to multiple wind farms are calculated according to the wind speed-wind turbine output physical model; Training the historical typhoon track data and predicting the typhoon track range includes: Performing Gaussian kernel training on the historical typhoon track data, and predicting the typhoon track range by two-dimensional Gaussian process regression, wherein the Gaussian kernel includes an RBF kernel function and an Index kernel function; Gaussian kernel training is performed on the historical typhoon track data, and the typhoon track range is predicted by two-dimensional Gaussian process regression, including: ； ； ； ； ； ； K tasks (i, i′)=(BB T +diag(v)) i，i′ ; In the formula, X Represents the two-dimensional input variables of the training samples, which are two groups of time points with equal step lengths, with dimensions [2× t ], t Indicates the number of time points; y The two-dimensional output variable representing the training sample is the longitude and latitude of the typhoon track range; Represents input variables and test points The covariance matrix of the covariance matrix consists of two parts, namely the covariance matrix considering the correlation between them and the covariance matrix taking into account the correlation of the output variables , and They are the labels corresponding to their output variables respectively; represents the variance of Gaussian noise that is independent and identically distributed with the training samples, I represents the identity matrix, Indicates test point The predicted output is represents the predicted mean, represents the prediction variance, represents the RBF kernel function, Represents the Index kernel function, , and All represent the hyperparameters of the kernel function; Combined with the wind speed scenario of the offshore wind farm and the environmental parameters and wind turbine parameters of the offshore wind farm, a wind speed-wind turbine output physical model is established, including: ； In the formula, p w Indicates the fan output. p N Indicates the rated power of the fan. ρ represents the air density, R represents the blade radius of the fan, v Indicates wind speed, v N Indicates the rated wind speed corresponding to the rated power of the fan. v in and v out are the cut-in and cut-out wind speeds of the fan, K represents the wind force coefficient; Where wind speed v Wind speed captured for wind turbines v h , which is related to the near-ground wind speed V i The relationship is: ； In the formula, v h represents the wind speed captured by the wind turbine, h Indicates the hub height, m c Represents the friction coefficient.

2. The method for calculating the output of multiple offshore wind farms under the influence of a typhoon according to claim 1, characterized in that: The different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model include: ； In the formula, V i Indicates that during a typhoon i The wind speed near the ground at point r i express i The distance from the point to the center of the typhoon, V max Indicates the maximum wind speed inside the horizontal structure of the typhoon, R max Indicates the maximum wind speed radius of a typhoon; R max =80－ k (950－ P c )； In the formula, P c Indicates the central pressure of the typhoon. k Represents the model coefficient of the preset maximum wind speed radius.

3. A terminal for calculating the output of multiple offshore wind farms under the influence of a typhoon, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the following steps are implemented: Acquire historical typhoon track data, train the historical typhoon track data and predict the typhoon track range; Different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model; A wind speed-wind turbine output physical model is established in combination with wind speed scenarios of offshore wind farms and environmental parameters and wind turbine parameters of offshore wind farms, and wind power output scenarios corresponding to multiple wind farms are calculated according to the wind speed-wind turbine output physical model; The terminal further comprises a human-machine interface module, which is used to input environmental parameters and wind turbine parameters of the offshore wind farm, and to output wind power output scenarios corresponding to the multiple wind farms obtained by calculation; Training the historical typhoon track data and predicting the typhoon track range includes: Performing Gaussian kernel training on the historical typhoon track data, and predicting the typhoon track range by two-dimensional Gaussian process regression, wherein the Gaussian kernel includes an RBF kernel function and an Index kernel function; Gaussian kernel training is performed on the historical typhoon track data, and the typhoon track range is predicted by two-dimensional Gaussian process regression, including: ； ； ； ； ； ； K tasks (i, i′)=(BB T +diag(v)) i，i′ ; In the formula, X Represents the two-dimensional input variables of the training samples, which are two groups of time points with equal step lengths, with dimensions [2× t ], t Indicates the number of time points; y The two-dimensional output variable representing the training sample is the longitude and latitude of the typhoon track range; Represents input variables and test points The covariance matrix of the covariance matrix consists of two parts, namely the covariance matrix considering the correlation between them and the covariance matrix taking into account the correlation of the output variables , and They are the labels corresponding to their output variables respectively; represents the variance of Gaussian noise that is independent and identically distributed with the training samples, I represents the identity matrix, Indicates test point The predicted output is represents the predicted mean, represents the prediction variance, represents the RBF kernel function, Represents the Index kernel function, , and All represent the hyperparameters of the kernel function; Combined with the wind speed scenario of the offshore wind farm and the environmental parameters and wind turbine parameters of the offshore wind farm, a wind speed-wind turbine output physical model is established, including: ； In the formula, p w Indicates the fan output. p N Indicates the rated power of the fan. ρ represents the air density, R represents the blade radius of the fan, v Indicates wind speed, v N Indicates the rated wind speed corresponding to the rated power of the fan. v in and v out are the cut-in and cut-out wind speeds of the fan, K represents the wind force coefficient; Where wind speed v Wind speed captured for wind turbines v h , which is related to the near-ground wind speed V i The relationship is: ； In the formula, v h represents the wind speed captured by the wind turbine, h Indicates the hub height, m c Represents the friction coefficient.

4. The output calculation terminal for multiple offshore wind farms under the influence of a typhoon according to claim 3 is characterized in that: The different wind speed scenarios corresponding to the typhoon track range calculated and predicted by the vortex wind field model include: ； In the formula, V i Indicates that during a typhoon i The wind speed near the ground at point r i express i The distance from the point to the center of the typhoon, V max Indicates the maximum wind speed inside the horizontal structure of the typhoon, R max Indicates the maximum wind speed radius of a typhoon; R max =80－ k (950－ P c )； In the formula, P c Indicates the central pressure of the typhoon. k Represents the model coefficient of the preset maximum wind speed radius.

Images