Wind power prediction method and system considering meteorological conditions and hydromechanics

Abstract

The invention discloses a wind power prediction method and system considering meteorological conditions and hydromechanics, and the method comprises the steps: building a wind power curve database according to the data of a numerical simulation sample, and obtaining a wind power curve according to the principle of wind energy conversion; generating output power sample data of the wind turbine generator based on the output power curve of the wind turbine generator in the wind power curve database; constructing a wind power plant meteorological prediction model based on Bayesian deep learning, and predicting wind speed data and turbulence intensity data according to historical operation data, roughness data, topographic condition data and NWP data; and constructing a fan power prediction model based on Bayesian deep learning, and predicting the output power of the fan according to the historical operation data and the wind speed data and turbulence intensity data predicted by the wind power plant meteorological prediction model. Redundant physical calculation is avoided, and the real-time requirement of wind power prediction and the safe and stable operation requirement of a power system can be met.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power prediction, and in particular relates to a wind power prediction method and system taking meteorological conditions and fluid mechanics into consideration. Background Art

[0002] As a non-polluting, widely available renewable energy source, wind power has become an important part of the power grid and plays an important role in power supply. However, due to the randomness and limited predictability of wind energy, wind power has become an intermittent power source, posing a major challenge to the safe and stable operation of the power system.

[0003] Accurate wind power prediction (WPP) is crucial for the safe and stable operation of power systems. However, traditional physical wind power prediction methods still face too high computational complexity and cannot meet the real-time requirements of WPP, while data-driven methods require a large amount of wind farm operation data, which limits their applicability to newly built wind farms. Summary of the invention

[0004] The present invention provides a wind power prediction method and system taking into account meteorological conditions and fluid mechanics, which are used to solve the technical problem that existing wind power prediction schemes cannot meet the real-time demand of wind power prediction and the applicability requirements of newly built wind farms.

[0005] In a first aspect, the present invention provides a method for predicting wind power taking into account meteorological conditions and fluid mechanics, comprising: According to the Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates sample data for numerical simulation; Constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion; Build a wind farm meteorological prediction model based on Bayesian deep learning to predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data and NWP data; A wind turbine power prediction model based on Bayesian deep learning is constructed to predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0006] In a second aspect, the present invention provides a wind power prediction system taking into account meteorological conditions and fluid mechanics, comprising: Generate a module configured to build a numerical simulation database based on computational fluid dynamics, according to the Reynolds-averaged Navier-Stokes equations and standard The turbulence model generates sample data for numerical simulation; A construction module configured to construct a wind power curve database according to the numerical simulation sample data, and generate output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set according to the principle of wind energy conversion; A first prediction module is configured to construct a wind farm meteorological prediction model based on Bayesian deep learning, and predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data, and NWP data; The second prediction module is configured to build a wind turbine power prediction model based on Bayesian deep learning, and predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0007] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the wind power prediction method considering meteorological conditions and fluid mechanics of any embodiment of the present invention.

[0008] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the steps of a wind power prediction method considering meteorological conditions and fluid mechanics according to any embodiment of the present invention.

[0009] The wind power prediction method and system of this application, which takes into account meteorological conditions and fluid mechanics, builds a database based on computational fluid dynamics and wind power curves, provides more realistic and informative prior knowledge for prediction and makes it applicable to newly built wind farms. It uses Bayesian deep learning to perform wind farm meteorological prediction and wind turbine power prediction, and has significant advantages such as probabilistic interpretability, uncertainty modeling of complex terrain distribution of wind farms, and excellent data utilization efficiency. Random variational reasoning is used to update the parameters of the model based on the actual operating data of the wind farm and save them regularly, so as to realize the adaptive update of the model and apply it to the entire life cycle of the wind farm, avoiding cumbersome physical calculations, which is conducive to meeting the real-time needs of wind power prediction and the safe and stable operation requirements of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0011] Figure 1 A flow chart of a method for predicting wind power taking into account meteorological conditions and fluid mechanics provided in one embodiment of the present invention; Figure 2 A structural block diagram of a wind power prediction system considering meteorological conditions and fluid mechanics provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] See also Figure 1 , which shows a flow chart of a wind power prediction method considering meteorological conditions and fluid mechanics of the present application.

[0014] like Figure 1 As shown, the wind power prediction method considering meteorological conditions and fluid mechanics specifically includes the following steps: Step S101, according to the Reynolds-averaged Navier-Stokes equation and the standard The turbulence model generates sample data for numerical simulations.

[0015] In this step, the Reynolds Averaged Navier-Stokes equations (RANS) are a time-averaged Navier-Stokes equation used to describe turbulence in fluid mechanics. The idea is to regard turbulent motion as the superposition of time-averaged and instantaneous pulsation flows, that is, any physical quantity satisfy: , In the formula, is the time mean, is the pulsation value, and the time-average value can be defined as: , In the formula, is the time differential, is the actual value of fluid flow rate; The turbulence characteristics represented by the standard k-ε turbulence model are: , In the formula, is the turbulent kinetic energy constant, is the turbulent viscosity, is the fluid density, is the dissipation rate, is the turbulent kinetic energy.

[0016] The numerical simulation of wind farms based on computational fluid dynamics (CFD) mainly considers inflow conditions, terrain conditions and wind turbine wake effects. The main factors affecting CFD results of wind farms include roughness, terrain, incoming wind speed, incoming wind direction, air pressure, thermal stability, etc.

[0017] It should be noted that obtaining geographic location , terrain conditions and roughness ; Obtain atmospheric boundary conditions, including incoming wind speed , wind direction , Thermal stability and air pressure ; Numerical simulation based on fluid mechanics to obtain the wind speed distribution and turbulence intensity distribution of the wind farm; Extract wind speed from calculated wind speed distribution , extract the turbulence intensity from the calculated turbulence intensity distribution ; Future wind speed Evenly divided into 50 discrete values ​​from 1m / s to 25m / s, incoming wind direction From 0° to 360°, it is evenly divided into 36 discrete values, thermal stability According to the Monin-Obukhov length, the air pressure is divided into 6 discrete values. Evenly divided into 20 discrete values ​​ranging from 0.8 to 1 atmosphere, the expression of the numerical simulation sample data is: , In the formula, For the A sample of numerical simulations in the computational fluid dynamics database.

[0018] Step S102 : constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion.

[0019] In this step, the wind speed is obtained and turbulence intensity ; Get the typical fan power curve provided by the fan manufacturer, the expression is: , In the formula, is the actual wind power, is the air density, is the windward swept area of ​​the fan blades, is the wind energy utilization coefficient affected by turbulence intensity, is the inflow wind speed, is the fan cut-in wind speed, Cut out wind speed for the fan, is the rated wind speed, is the rated power of the fan; Calculate the fan power based on the typical fan power curve and polynomial interpolation method Specifically, the data measured by the typical wind turbine power curve can be interpolated using the Lagrange interpolation method, and then the specific wind speed value can be input to obtain the corresponding wind turbine power. .

[0020] Step S103, constructing a wind farm meteorological prediction model based on Bayesian deep learning, and predicting wind speed data and turbulence intensity data according to historical operation data, roughness data, terrain condition data and NWP data.

[0021] In this step, a Bayesian dense layer is constructed. The basic unit of the Bayesian dense layer is an artificial neuron, and the expression is: , In the formula, To output information, is the activation function, is the weight of the i-th input feature, is the number of neurons, is the i-th input feature, for deviation; By assigning random parameters To construct variational Bayesian neurons; Obtain historical operation data, roughness data, terrain condition data and numerical weather forecast data. The expression is: , In the formula, for Input features at all times, For terrain conditions, is the roughness, for Historical operation data of wind farms at all times, for Historical operation data of wind farms at all times, for Numerical weather forecast data at all times; The gated recurrent unit is used to process the historical operation data, the convolutional neural network is used to process the roughness data and terrain data, the dense layer is used to process the NWP data, and the wind farm meteorological prediction model is constructed based on multiple variational Bayesian dense layers. The wind farm meteorological prediction model outputs the wind speed data and turbulence intensity data of each wind turbine, which is expressed as follows: , In the formula, , The fans are Forecasted wind speed and turbulence intensity at the moment.

[0022] Step S104, constructing a wind turbine power prediction model based on Bayesian deep learning, and predicting the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0023] In this step, the wind turbine power prediction model based on Bayesian deep learning is defined, and the expression is: , In the formula, is the wind power output, is the input feature, is a random parameter set, is the probability density function; According to Bayesian theory, the posterior inference problem is expressed as: , In the formula, For a priori, For sample information, is the sample theoretical probability, is the real distribution; The historical operation data, the wind speed data and the turbulence intensity data predicted by the wind farm meteorological prediction model are obtained, and the expression is: , In the formula, for Wind turbine power prediction data sample at the moment, for Historical operation data of the wind turbine at all times, for Historical operation data of the wind turbine at all times, , The fans are The predicted wind speed and turbulence intensity at the moment; GRU is used to process historical operation data, and dense layer is used to process wind speed data and turbulence intensity data, and the output is the predicted output power of the wind turbine. .

[0024] In a specific embodiment, the parameters of the wind turbine power prediction model and the wind farm meteorological prediction model are updated based on stochastic variational reasoning, and the updated wind turbine power prediction model and the wind farm meteorological prediction model are retrained based on the operating data set of the wind farm.

[0025] It should be noted that as a combination of variational inference, small batch subsampling strategy and variance reduction technology, stochastic variational inference (SVI) has high computational efficiency and efficient memory utilization, so SVI is used for parameter update.

[0026] SVI introduces a series of distributions , using prior knowledge to determine the detailed type and initial parameters of the distribution , by optimizing , find the closest The Kullback-Leibler divergence is used to measure the closeness of different distributions: , In the formula, is the sample posterior, is the Kullback-Leibler divergence between the parameter posterior and the actual distribution, is the actual distribution of the parameters, is the parameter posterior, is the expected value; Therefore, evidence for: , In the formula, To set a limit for evidence; because , the a posteriori reasoning problem is equivalent to maximizing , the parameter update problem is transformed into the ELBO maximization problem, record , the gradient of ELBO is: , In the formula, is a constant and does not participate in the gradient calculation. is the proxy objective function.

[0027] The SVI parameter update process is as follows: First, the BDL-based model is initialized by pre-training it using the prior database If there is a regularly saved BDL model, the most recently saved parameters can be used to initialize Then, repeat the following steps: The entire sample data set Randomly divide into different small batch sample sets ; According to the probability density function conduct Monte Carlo sampling, calculation ; Use each mini-batch sample of calculate ;according to and learning rate renew .

[0028] In summary, the method of this application builds a database based on computational fluid dynamics and wind power curves, provides more realistic and informative prior knowledge for prediction and makes it applicable to newly built wind farms, and uses Bayesian deep learning to perform wind farm meteorological forecasts and wind turbine power forecasts, which has significant advantages such as probabilistic interpretability, uncertainty modeling of complex terrain distribution of wind farms, and excellent data utilization efficiency. Random variational reasoning is used to update the parameters of the model based on the actual operating data of the wind farm and save them regularly, so as to realize the adaptive update of the model and apply it to the entire life cycle of the wind farm, avoiding cumbersome physical calculations, which is conducive to meeting the real-time needs of wind power forecasting and the safe and stable operation requirements of the power system.

[0029] See also Figure 2 , which shows a structural block diagram of a wind power prediction system of the present application that takes into account meteorological conditions and fluid mechanics.

[0030] like Figure 2 As shown, the wind power prediction system 200 includes a generation module 210 , a construction module 220 , a first prediction module 230 and a second prediction module 240 .

[0031] The generation module 210 is configured to construct a numerical simulation database based on computational fluid dynamics, according to the Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates numerical simulation sample data; a construction module 220 is configured to construct a wind power curve database based on the numerical simulation sample data, and generate output power sample data of the wind turbine based on the output power curve of the wind turbine according to the principle of wind energy conversion; a first prediction module 230 is configured to construct a wind farm meteorological prediction model based on Bayesian deep learning, and predict wind speed data and turbulence intensity data according to historical operation data, roughness data, terrain condition data and NWP data; a second prediction module 240 is configured to construct a wind turbine power prediction model based on Bayesian deep learning, and predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0032] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0033] In some other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the wind power prediction method considering meteorological conditions and fluid mechanics in any of the above method embodiments; As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows: According to the Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates sample data for numerical simulation; Constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion; Build a wind farm meteorological prediction model based on Bayesian deep learning to predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data and NWP data; A wind turbine power prediction model based on Bayesian deep learning is constructed to predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0034] The computer-readable storage medium 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 a wind power prediction system that takes into account meteorological conditions and fluid mechanics, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the wind power prediction system that takes into account meteorological conditions and fluid mechanics 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.

[0035] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the connection through the bus is taken as an example. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the wind power prediction method considering meteorological conditions and fluid mechanics in the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the wind power prediction system considering meteorological conditions and fluid mechanics. The output device 340 may include display devices such as display screens.

[0036] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0037] As an implementation mode, the electronic device is applied to a wind power prediction system taking into account meteorological conditions and fluid mechanics, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: According to the Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates sample data for numerical simulation; Constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion; Build a wind farm meteorological prediction model based on Bayesian deep learning to predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data and NWP data; A wind turbine power prediction model based on Bayesian deep learning is constructed to predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

[0038] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0039] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A wind power prediction method considering meteorological conditions and fluid mechanics, characterized in that: include: According to the Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates sample data for numerical simulation; Constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion; Build a wind farm meteorological prediction model based on Bayesian deep learning to predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data and NWP data; A wind turbine power prediction model based on Bayesian deep learning is constructed to predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

2. A method for predicting wind power considering meteorological conditions and fluid mechanics according to claim 1, characterized in that: After predicting the output power of the wind turbine according to the historical operation data, the wind speed data and the turbulence intensity data predicted by the wind farm meteorological prediction model, the method further includes: The parameters of the wind turbine power prediction model and the wind farm meteorological prediction model are updated based on stochastic variational reasoning, and the updated wind turbine power prediction model and the wind farm meteorological prediction model are retrained based on the operating data set of the wind farm.

3. A method for predicting wind power considering meteorological conditions and fluid mechanics according to claim 1, characterized in that: The Reynolds-averaged Navier-Stokes equations and the standard The turbulence model generates numerical simulation sample data including: Get geographic location , terrain conditions and roughness ; Obtain atmospheric boundary conditions, including incoming wind speed , wind direction , Thermal stability and air pressure ; Numerical simulation based on fluid mechanics to obtain the wind speed distribution and turbulence intensity distribution of the wind farm; Extract wind speed from calculated wind speed distribution , extract the turbulence intensity from the calculated turbulence intensity distribution ; Future wind speed Evenly divided into 50 discrete values ​​from 1m / s to 25m / s, incoming wind direction From 0° to 360°, it is evenly divided into 36 discrete values, thermal stability According to the Monin-Obukhov length, the air pressure is divided into 6 discrete values. Evenly divided into 20 discrete values ​​ranging from 0.8 to 1 atmosphere, the expression of the numerical simulation sample data is: ， In the formula, For the A sample of numerical simulations in the computational fluid dynamics database.

4. A method for predicting wind power considering meteorological conditions and fluid mechanics according to claim 1, characterized in that: The step of constructing a wind power curve database according to the numerical simulation sample data, and generating output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set in the wind power curve database according to the principle of wind energy conversion includes: Get wind speed and turbulence intensity ; Get the typical fan power curve provided by the fan manufacturer, the expression is: ， In the formula, is the actual wind power, is the air density, is the windward swept area of ​​the fan blades, is the wind energy utilization coefficient affected by turbulence intensity, is the inflow wind speed, is the fan cut-in wind speed, Cut out wind speed for the fan, is the rated wind speed, is the rated power of the fan; Calculate the fan power based on the typical fan power curve and polynomial interpolation method .

5. A method for predicting wind power considering meteorological conditions and fluid mechanics according to claim 1, characterized in that: The wind farm meteorological prediction model based on Bayesian deep learning is constructed, and the wind speed data and turbulence intensity data are predicted according to the historical operation data, roughness data, terrain condition data and NWP data, including: Construct a Bayesian dense layer. The basic unit of the Bayesian dense layer is an artificial neuron, and the expression is: ， In the formula, To output information, is the activation function, is the weight of the i-th input feature, is the number of neurons, is the i-th input feature, for deviation; By assigning random parameters To construct variational Bayesian neurons; Obtain historical operation data, roughness data, terrain condition data and numerical weather forecast data. The expression is: ， In the formula, for Input features at all times, For terrain conditions, is the roughness, for Historical operation data of wind farms at all times, for Historical operation data of wind farms at all times, for Numerical weather forecast data at all times; The gated recurrent unit is used to process the historical operation data, the convolutional neural network is used to process the roughness data and terrain data, the dense layer is used to process the NWP data, and the wind farm meteorological prediction model is constructed based on multiple variational Bayesian dense layers. The wind farm meteorological prediction model outputs the wind speed data and turbulence intensity data of each wind turbine, which is expressed as follows: ， In the formula, , The fans are Forecasted wind speed and turbulence intensity at the moment.

6. A method for predicting wind power considering meteorological conditions and fluid mechanics according to claim 1, characterized in that: The wind turbine power prediction model based on Bayesian deep learning is constructed to predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model, including: The wind turbine power prediction model based on Bayesian deep learning is defined as follows: ， In the formula, is the wind power output, is the input feature, is a random parameter set, is the probability density function; According to Bayesian theory, the posterior inference problem is expressed as: ， In the formula, For a priori, For sample information, is the sample theoretical probability, is the real distribution; The historical operation data, the wind speed data and the turbulence intensity data predicted by the wind farm meteorological prediction model are obtained, and the expression is: ， In the formula, for Wind turbine power prediction data sample at the moment, for Historical operation data of the wind turbine at all times, for Historical operation data of the wind turbine at all times, , The fans are The predicted wind speed and turbulence intensity at the moment; GRU is used to process historical operation data, and dense layer is used to process wind speed data and turbulence intensity data, and the output is the predicted output power of the wind turbine. .

7. A wind power prediction system considering meteorological conditions and fluid mechanics, characterized in that: include: Generate a module configured to build a numerical simulation database based on computational fluid dynamics, according to the Reynolds-averaged Navier-Stokes equations and standard The turbulence model generates sample data for numerical simulation; A construction module configured to construct a wind power curve database according to the numerical simulation sample data, and generate output power sample data of the wind turbine generator set based on the output power curve of the wind turbine generator set according to the principle of wind energy conversion; A first prediction module is configured to construct a wind farm meteorological prediction model based on Bayesian deep learning, and predict wind speed data and turbulence intensity data based on historical operation data, roughness data, terrain condition data, and NWP data; The second prediction module is configured to build a wind turbine power prediction model based on Bayesian deep learning, and predict the output power of the wind turbine according to historical operation data, wind speed data and turbulence intensity data predicted by the wind farm meteorological prediction model.

8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images