Wind power prediction method and device, computer device and storage medium

Abstract

The present application relates to the technical field of clean energy, and discloses a wind power prediction method and device, computer equipment and a storage medium; the wind power prediction method comprises: obtaining geographical features of a target wind power station; constructing a two-way coupled weather forecast model based on the geographical features; taking a first preset time length as a period, performing two-way coupling of a WRF model and a CFD model based on the two-way coupled weather forecast model to obtain first weather forecast data; performing downscaling processing on the first weather forecast data to obtain wind speed features of the target wind power station; inputting the wind speed features into a pre-constructed wind power prediction model to obtain a wind power prediction result; the above method can take the geographical features of the target wind power station as a benchmark, perform two-way coupling correction based on weather forecast data of different spatial scales to obtain more accurate wind speed features, and finally perform wind power prediction using the wind speed features, thereby having the advantage of more accurate wind power prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of clean energy, in particular to a wind power prediction method and device, computer equipment and a storage medium. BACKGROUND

[0002] With the increasing proportion of new energy in the power system, accurate prediction of wind power is of great significance for power grid dispatching, power trading and power supply stability.

[0003] The traditional wind power prediction method uses an independent atmospheric model such as a Weather Research and Forecasting Model (WRF) model and a wake model such as a Jensen model to simulate meteorological conditions to obtain predicted meteorological data, and further inputs the predicted meteorological data into a power prediction model to predict wind power and obtain wind power prediction results. The above wind power prediction method, the WRF model performs excellently in the prediction of wind speed at ultra-short and short time scales, but has problems such as poor grid resolution and difficulty in capturing local terrain features, resulting in inaccurate wind power prediction. For wind power prediction with high grid resolution, a Computational Fluid Dynamics (CFD) model is usually used for micro-simulation to achieve it. However, the CFD model is difficult to respond to large-scale meteorological condition changes in real time, and the wind speed prediction is inaccurate when the boundary layer suddenly changes due to the passage of a front, resulting in inaccurate wind power prediction.

[0004] Therefore, there is an urgent need to propose a wind power prediction method to solve the problem of inaccurate wind power prediction in the related art. SUMMARY

[0005] The present application provides a wind power prediction method, device, computer equipment and storage medium to solve the problem of inaccurate wind power prediction in the related art.

[0006] In a first aspect, the present application provides a wind power prediction method, comprising: obtaining geographical features of a target wind power station; constructing a bidirectional coupling weather forecast model based on the geographical features; wherein the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; performing bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model with a first preset time length as a period to obtain first weather forecast data; in the process of bidirectional coupling, driving the WRF model to perform weather forecasting based on a first constraint condition to obtain second weather forecast data; driving the CFD model to perform weather forecasting based on the second weather forecast data and the geographical features to obtain the first weather forecast data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; performing downscaling processing on the first weather forecast data to obtain wind speed characteristics of the target wind power station; inputting the wind speed characteristics into a pre-constructed wind power prediction model to obtain a wind power prediction result.

[0007] As an exemplary embodiment, the bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model comprises: extracting a boundary layer height and a vertical wind shear profile from the second weather forecast data; inputting the boundary layer height and the vertical wind shear profile as a second constraint condition into the CFD model to perform transient wake weather simulation to obtain the first weather forecast data; extracting an equivalent roughness field from the first weather forecast data; wherein the equivalent roughness field is obtained by the CFD model; inputting the equivalent roughness field as the first constraint condition into the WRF model for coupling.

[0008] As an exemplary embodiment, the inputting the boundary layer height and the vertical wind shear profile as a second constraint condition into the CFD model to perform transient wake weather simulation to obtain the first weather forecast data comprises: calculating a domain height of the CFD model based on the boundary layer height and the geographical features; extracting a Richardson number as boundary layer state information from the second weather forecast data; determining a vertical layered grid layer number based on the boundary layer state information; inputting the domain height, the vertical layered grid layer number, the boundary layer height and the vertical wind shear profile as the second constraint condition into the CFD model to perform transient wake weather simulation to obtain the first weather forecast data.

[0009] As an exemplary embodiment, the wind power prediction method further comprises: obtaining a preset Richardson number threshold; determining atmospheric stratification stability information based on the Richardson number and the preset Richardson number threshold; updating a turbulence model parameter of the CFD model based on the atmospheric stratification stability information to obtain a first corrected CFD model; inputting the domain height, the vertical stratification grid layer number, the boundary layer height, and the vertical wind shear profile into the first corrected CFD model as the second constraint condition for transient wake meteorological simulation to obtain the first meteorological prediction data.

[0010] As an exemplary embodiment, the wind power prediction method further comprises: if the geographical feature satisfies a preset marine geographical feature, obtaining marine meteorological data of the target wind power station based on the geographical feature in hybrid coordinate ocean model data; wherein the marine meteorological data comprises sea surface temperature and ocean mixed layer depth; determining a sea-air flux parameter of the WRF model based on the ocean mixed layer depth; taking a second preset time length as an update period, interpolating the sea surface temperature into an outer grid of the WRF model based on the spatiotemporal characteristics of the sea surface temperature, and adding the sea-air flux parameter to the first constraint condition to obtain a third constraint condition; driving the WRF model based on the third constraint condition to obtain third meteorological prediction data.

[0011] As an exemplary embodiment, after interpolating the sea surface temperature into the outer grid of the WRF model based on the spatial characteristics of the sea surface temperature and adding the sea-air flux parameter to the first constraint condition, the wind power prediction method further comprises: extracting marine temperature data, air temperature data, and ocean mixed layer heat flux data from the third meteorological prediction data; calculating a temperature difference value based on the marine temperature data and the air temperature data; updating a turbulence model parameter of the CFD model based on the temperature difference value to obtain a second corrected CFD model; driving the second corrected CFD model based on the third meteorological prediction data to obtain the first meteorological prediction data.

[0012] As an exemplary embodiment, the wind power prediction method further comprises: extracting local wind speed variation data from the first meteorological prediction data; updating the hybrid coordinate ocean model data based on the local wind speed variation data to obtain updated ocean mixed layer heat flux data; adding the updated ocean mixed layer heat flux data to the first constraint condition.

[0013] In a second aspect, the present application provides a wind power prediction device, comprising: an acquisition module configured to acquire geographical features of a target wind power station; a model construction module configured to construct a bidirectional coupling weather forecast model based on the geographical features; wherein the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; a bidirectional coupling module configured to periodically perform bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model at a first preset time length to obtain first weather forecast data; in the process of bidirectional coupling, the WRF model is driven to perform weather forecast based on a first constraint condition to obtain second weather forecast data; the CFD model is driven to perform weather forecast based on the second weather forecast data and the geographical features to obtain the first weather forecast data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; a downscaling module configured to perform downscaling processing on the first weather forecast data to obtain wind speed features of the target wind power station; and a wind power prediction module configured to input the wind speed features into a pre-constructed wind power prediction model to obtain a wind power prediction result.

[0014] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, which are communicatively connected with each other, and the memory stores computer instructions; the processor executes the computer instructions to perform the wind power prediction method of the first aspect or any of the corresponding embodiments thereof.

[0015] In a fourth aspect, the present application provides a computer readable storage medium, which stores computer instructions for making a computer execute the wind power prediction method of the first aspect or any of the corresponding embodiments thereof.

[0016] The application provides a wind power prediction method, device, computer equipment and storage medium. The wind power prediction method comprises: obtaining geographical features of a target wind power station; constructing a bidirectional coupling weather forecast model based on the geographical features; wherein the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; taking a first preset time length as a period, bidirectionally coupling the WRF model and the CFD model based on the bidirectional coupling weather forecast model to obtain first weather forecast data; in the process of bidirectional coupling, driving the WRF model to perform weather forecasting based on a first constraint condition to obtain second weather forecast data; driving the CFD model to perform weather forecasting based on the second weather forecast data and the geographical features to obtain the first weather forecast data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; performing downscaling processing on the first weather forecast data to obtain wind speed features of the target wind power station; inputting the wind speed features into a pre-constructed wind power prediction model to obtain a wind power prediction result. The WRF model can provide accurate second weather forecast data to the CFD model based on the geographical features, the CFD model can determine the first weather forecast data in response to large-scale weather condition changes caused by, for example, frontal passage, of the second weather forecast data of a larger spatial scale provided by the WRF model, further drive the WRF model as a new first constraint condition, and the WRF model can input a more accurate first constraint condition in response to the first weather forecast data of a smaller spatial scale provided by the CFD model, thereby obtaining updated second weather forecast data based on the more accurate first constraint condition. In the bidirectional coupling process, the finally obtained first weather forecast data can be corrected based on weather forecast data of large and small spatial scales by taking the geographical features of the target wind power station as a benchmark, more accurate first weather forecast data is obtained, further more accurate wind speed features are obtained, and finally wind power prediction is performed by using the wind speed features, thereby having the advantage of more accurate wind power prediction. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the drawings needed in the specific embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings without creative labor based on these drawings.

[0018] Figure 1 is a flowchart of a wind power prediction method according to an embodiment of the present application;

[0019] Figure 2 is a structural block diagram of a wind power prediction device according to an embodiment of the present application;

[0020] Figure 3 is a hardware structure schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

[0022] According to the embodiments of the present application, a wind power prediction method embodiment is provided. It should be noted that the steps shown in the flowchart of the drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that here.

[0023] In the present embodiment, a wind power prediction method is provided, Figure 1 is a flowchart of a wind power prediction method according to an embodiment of the present application, as shown in Figure 1 the flowchart includes the following steps:

[0024] Step S101, obtaining the geographical features of a target wind power station.

[0025] For example, the geographical features can include geographical data information, wherein the geographical data information can include longitude and latitude coordinates, altitude, wind turbine hub height, etc.

[0026] For example, the target wind power station can be a wind power station cluster.

[0027] For example, the target wind power station can be a cluster of offshore wind power stations near the ocean.

[0028] Step S102, constructing a two-way coupled weather forecast model based on the geographical features; wherein the two-way coupled weather forecast model includes a WRF model and a CFD model.

[0029] In the embodiment, the bidirectional coupling weather forecasting model comprises a WRF model and a CFD model; the WRF model can focus on simulating large-scale, mesoscale and microscale atmospheric states (wind, temperature, air pressure, humidity, turbulence, etc.), and does not consider the influence of the fan itself on the airflow; the CFD model forecasting model can simulate the actual fluid flow situation, and the basic principle is to numerically solve the differential equations of the control fluid flow to obtain the discrete distribution of the flow field of the fluid flow in the continuous region, thereby approximating the fluid flow situation; in order to enable the WRF model and the CFD model to consider the geographical features of the target wind power station and adaptively perform weather forecasting, in the embodiment, a bidirectional coupling weather forecasting model is constructed based on geographical features; specifically, when constructing the bidirectional coupling weather forecasting model, the WRF model and the CFD model are separately constructed based on geographical features, and a data interface is configured for the WRF model to input data from the CFD model as a constraint condition, and a data interface is configured for the CFD model to input data from the WRF model as a constraint condition.

[0030] In step S103, the WRF model and the CFD model are bidirectionally coupled based on the bidirectional coupling weather forecasting model with a first preset time length as a period to obtain first weather forecasting data; in the process of bidirectional coupling, the WRF model is driven based on a first constraint condition to perform weather forecasting to obtain second weather forecasting data; the CFD model is driven based on the second weather forecasting data and the geographical features to perform weather forecasting to obtain the first weather forecasting data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecasting data.

[0031] The traditional wind power prediction method uses an independent atmospheric model such as a weather research and forecasting model (WRF model) and a wake model such as a Jensen model to simulate weather conditions to obtain predicted weather data, and further inputs the predicted weather data into a power prediction model to perform power prediction to obtain a wind power prediction result; the above wind power prediction method, the WRF model performs excellently in wind speed prediction at an ultra-short-term and short-term time scale, but has problems such as poor grid resolution and difficulty in capturing local terrain features, thereby leading to inaccurate wind power prediction; for wind power prediction with better grid resolution, a computational fluid dynamics (CFD model) is usually used for micro-simulation to achieve; however, the CFD model is difficult to respond to large-scale weather condition changes in real time, and the wind speed prediction is inaccurate when the weather appears boundary layer mutation caused by frontal passage, thereby leading to inaccurate wind power prediction.

[0032] To solve the above problems, in the embodiment, a two-way coupling mode of the WRF model and the CFD model is adopted to fully consider the advantages of the input data of the WRF model and the CFD model, and meteorological prediction data considering the advantages of the two prediction modules are obtained by coupling.

[0033] Exemplarily, after the two-way coupling model is constructed, the WRF model and the CFD model are two-way coupled with a first preset time length as a period.

[0034] Exemplarily, the first preset time length can be 30 min.

[0035] Specifically, for the first coupling period, that is, the initial coupling process of the two-way coupling model, the WRF model is driven by the geographical features as the first constraint condition to obtain second meteorological prediction data; the CFD model is driven based on the second meteorological prediction data and the geographical features to obtain first meteorological prediction data; further, the WRF model is driven by the geographical features and the first meteorological prediction data as the first constraint condition to obtain second meteorological prediction data, and the CFD model is driven based on the second meteorological prediction data and the geographical features to obtain the first meteorological prediction data, so as to complete the two-way coupling of the WRF model and the CFD model.

[0036] For the non-first coupling period, the WRF model is driven by the geographical features and the first meteorological prediction data as the first constraint condition to obtain second meteorological prediction data, and the CFD model is driven based on the second meteorological prediction data and the geographical features to obtain the first meteorological prediction data, so as to complete the two-way coupling of the WRF model and the CFD model.

[0037] Exemplarily, the spatial resolution of the second meteorological prediction data is 3 km, including boundary layer parameter data such as boundary layer height and vertical wind shear profile.

[0038] Exemplarily, the spatial resolution of the first meteorological prediction data is 300 m, including wind farm overall drag coefficient data, wind speed data, wind direction data and the like.

[0039] The WRF model can provide accurate second meteorological prediction data to the CFD model based on geographical features. The CFD model can determine the first meteorological prediction data in response to large-scale meteorological condition changes caused by, for example, the passage of a front, which causes a sudden change in the boundary layer, of the second meteorological prediction data provided by the WRF model of a larger spatial scale. Further, the first meteorological prediction data is used as a new first constraint condition to re-drive the WRF model. The WRF model can input the first meteorological prediction data provided by the CFD model of a smaller spatial scale to a more accurate first constraint condition, so as to obtain updated second meteorological prediction data based on the more accurate first constraint condition. Through the above two-way coupling process, the first meteorological prediction data obtained finally can be corrected based on meteorological prediction data of large and small spatial scales in two ways with the geographical features of the target wind power station as the benchmark, so as to obtain more accurate first meteorological prediction data.

[0040] In step S104, the first meteorological prediction data is down-scaled to obtain the wind speed characteristics of the target wind power station.

[0041] After obtaining the first meteorological prediction data, the first meteorological prediction data is down-scaled to obtain the wind speed characteristics of the target wind power station.

[0042] Illustratively, after obtaining the first meteorological prediction data, the first meteorological prediction data is interpolated to obtain third meteorological prediction data.

[0043] Illustratively, the resolution of the third meteorological prediction data is 10 m.

[0044] After obtaining the third meteorological prediction data, the wind speed characteristics of the target wind power station are determined based on the geographical location of the target wind power station and the third meteorological prediction data.

[0045] In step S105, the wind speed characteristics are input into a pre-constructed wind power prediction model to obtain a wind power prediction result.

[0046] The application provides a wind power prediction method, which comprises the following steps: acquiring geographical features of a target wind power station; constructing a bidirectional coupling weather forecast model based on the geographical features; wherein the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; periodically performing bidirectional coupling on the WRF model and the CFD model based on the bidirectional coupling weather forecast model with a first preset time length as a period to obtain first weather forecast data; in the process of bidirectional coupling, driving the WRF model to perform weather forecasting based on a first constraint condition to obtain second weather forecast data; driving the CFD model to perform weather forecasting based on the second weather forecast data and the geographical features to obtain first weather forecast data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; performing downscaling processing on the first weather forecast data to obtain wind speed characteristics of the target wind power station; inputting the wind speed characteristics into a pre-constructed wind power prediction model to obtain a wind power prediction result; the wind power prediction method, the WRF model can provide accurate second weather forecast data to the CFD model based on the geographical features, the CFD model can determine the first weather forecast data in response to the large-scale weather condition changes caused by the boundary layer mutation of the second weather forecast data provided by the WRF model, for example, the frontal passage, further drive the WRF model as a new first constraint condition, and the WRF model can input more accurate first constraint conditions in response to the smaller spatial scale first weather forecast data provided by the CFD model, so as to obtain updated second weather forecast data based on more accurate first constraint conditions; in the above bidirectional coupling process, the finally obtained first weather forecast data can be based on the geographical features of the target wind power station, and the large-scale and small-scale weather forecast data can be bidirectionally coupled and corrected, so that more accurate first weather forecast data is obtained, further more accurate wind speed characteristics are obtained, and finally the wind power prediction is performed based on the wind speed characteristics, thereby achieving the advantage of more accurate wind power prediction.

[0047] As an exemplary embodiment, the multiple rounds of bidirectional coupling based on the WRF model and the CFD model comprise the following steps: extracting a boundary layer height and a vertical wind shear profile from the second weather forecast data; inputting the boundary layer height and the vertical wind shear profile into the CFD model as a second constraint condition to perform transient wake weather simulation to obtain the first weather forecast data; extracting an equivalent roughness field from the first weather forecast data; wherein the equivalent roughness field is obtained by the CFD model based on wake induction; and inputting the equivalent roughness field into the WRF model as the first constraint condition for coupling.

[0048] As described above, the CFD model is difficult to respond to large-scale meteorological condition changes in real time, and the wind speed prediction is inaccurate when the boundary layer suddenly changes due to, for example, the passage of a front. For the second meteorological prediction data with a large spatial scale, the boundary layer height and the vertical wind shear profile can accurately reflect the large-scale meteorological condition changes of the target wind farm. Therefore, in order to solve the problem that the CFD model is difficult to respond to large-scale meteorological condition changes in real time, the boundary layer height and the vertical wind shear profile are extracted from the second meteorological prediction data. The boundary layer height and the vertical wind shear profile are input as the second constraint condition into the CFD model for transient wake meteorological simulation, and the first meteorological prediction data is obtained.

[0049] The WRF model has problems such as poor grid resolution and difficulty in capturing local terrain features. For the first meteorological prediction data with a small spatial scale, the equivalent roughness field can accurately reflect the local terrain features of the target wind farm. Therefore, in order to solve the problems such as poor grid resolution and difficulty in capturing local terrain features of the WRF model, the equivalent roughness field is input as the first constraint condition into the WRF model for coupling, so as to obtain relatively accurate second meteorological prediction data.

[0050] As an exemplary embodiment, the inputting of the boundary layer height and the vertical wind shear profile into the CFD model for transient wake meteorological simulation to obtain the first meteorological prediction data includes: calculating a domain height of the CFD model based on the boundary layer height and the geographical features; extracting a Richardson number as boundary layer state information in the second meteorological prediction data; determining a vertical layered grid layer number based on the boundary layer state information; inputting the domain height, the vertical layered grid layer number, the boundary layer height, and the vertical wind shear profile as the second constraint condition into the CFD model for transient wake meteorological simulation to obtain the first meteorological prediction data.

[0051] In this embodiment, the domain height of the CFD model is calculated according to the boundary layer height to optimize the CFD model. At the same time, the Richardson number representing the boundary layer state information is used to further optimize the boundary parameters of the CFD model, so that the CFD model can determine the first meteorological prediction data in response to the large-scale meteorological condition changes such as the boundary layer mutation caused by the passage of a front in the second meteorological prediction data with a large spatial scale provided by the WRF model.

[0052] In one embodiment, the WRF model can obtain the boundary layer height by calculating a Richardson number (Bulk Richardson Number, Ri), calculating a turbulent kinetic energy vertical distribution, calculating a potential temperature gradient threshold, and / or calculating an entrainment layer thickness, and then pass the boundary layer height to the CFD model.

[0053] Exemplarily, the height at which the turbulent kinetic energy decays to 10% of the ambient value is taken as the boundary layer height.

[0054] Exemplarily, the height at which the temperature vertical gradient first exceeds 0.005 K / m is taken as the boundary layer height.

[0055] Exemplarily, the height at which the entrainment layer occupies 15%-20% is taken as the boundary layer height.

[0056] As an exemplary embodiment, the domain height of the CFD model is calculated by formula (1):

[0057] HCFD=α*PBLH+β*Hterrain (1)

[0058] In formula (1), HCFD represents the calculated domain height of the CFD model, PBLH represents the boundary layer height, Hterrain represents the terrain height determined by the geographical feature; α takes 1.2, and β takes 0.3; β takes 0.3 as a terrain height compensation coefficient, which can prevent steep terrain from being truncated.

[0059] In the embodiment, the vertical stratification network in the CFD model is further optimized based on the second meteorological forecast data; exemplarily, the Richardson number is extracted as boundary layer state information in the second meteorological forecast data; the number of vertical stratification grids is determined based on the boundary layer state information; specifically, the number of vertical stratification grids is inversely related to the Richardson number.

[0060] Exemplarily, after the Richardson number is extracted, when Ri<0, it is determined that the boundary layer state is an unstable stratification; when 0≤Ri<0.25, it is determined that the boundary layer state is a neutral stratification; when Ri>0.25, it is determined that the boundary layer state is a stable stratification; further, for the number of vertical stratification networks in the CFD model, the number of vertical stratification grids corresponding to the unstable layer is 50 layers; the number of vertical stratification grids corresponding to the neutral layer is 40 layers, and the number of vertical stratification grids corresponding to the stable layer is 30 layers.

[0061] Exemplarily, for the grid in the CFD model, the spring approximation method is adopted to smooth the transition when the grid is transitioned, so as to avoid mutation.

[0062] As an exemplary embodiment, the wind power prediction method further comprises: determining the MPI domain decomposition of the CFD model grid corresponding to each boundary layer height based on the boundary layer height.

[0063] In the embodiment, the CFD model is allocated with computing resources based on the boundary layer height; for example, when the MPI domain decomposition of the CFD model grid corresponding to each boundary layer height is determined based on the boundary layer height, more computing nodes are allocated to the region with the boundary layer height greater than a preset value, and grid coarsening is enabled for the region with the boundary layer height less than or equal to the preset value, so as to reduce resource allocation.

[0064] As an exemplary embodiment, after the domain height of the CFD model is calculated based on the boundary layer height and the geographical feature, the wind power prediction method further comprises: obtaining an initial domain height in a previous preset time length; calculating a difference value between the domain height and the initial domain height; when the difference value is greater than a preset value, adjusting the difference value to the preset value to obtain a corrected difference value; and correcting the corresponding domain height based on the corrected difference value.

[0065] In the embodiment, the preset value can be 0.2 times the initial domain height, so that the adjustment range of the domain height in each round of bidirectional coupling process does not exceed 20% of the previous coupling process.

[0066] As an exemplary embodiment, after the boundary layer height is extracted from the second meteorological forecast data, the method further comprises: calculating the difference between the boundary layer heights of adjacent grid points as a boundary layer height difference value; marking the boundary layer height with a boundary layer height difference value greater than the boundary layer height difference value as an abnormal value; and performing median filtering processing on the abnormal value to obtain a corrected boundary layer height; in the embodiment, the abnormal point is determined based on the difference between the boundary layer heights of adjacent two points, and the median filtering processing is used for the abnormal point.

[0067] As an exemplary embodiment, the wind power prediction method further comprises: obtaining a preset Richardson number threshold; determining atmospheric stratification stability information based on the Richardson number and the preset Richardson number threshold; updating a turbulence model parameter of the CFD model based on the atmospheric stratification stability information to obtain a first corrected CFD model; and inputting the domain height, the vertical stratification grid layer number, the boundary layer height, and the vertical wind shear profile into the first corrected CFD model as the second constraint condition to perform transient wake meteorological simulation, thereby obtaining the first meteorological forecast data.

[0068] In the embodiment, the atmospheric stratification stability information is determined based on the Richardson number and the preset Richardson number threshold, and further, different turbulence models are selected for the CFD model according to different atmospheric stratification stability information.

[0069] For example, the preset Richardson number threshold is 0.25; after the Richardson number is extracted, when 0≤Ri<0.25, it is determined that the boundary layer state is neutral stratification; and when Ri>0.25, it is determined that the boundary layer state is stable stratification.

[0070] wherein, exemplarily, the turbulent flow model can include a k-ε model and a k-ω SST model.

[0071] Exemplarily, for the neutral layer, the k-ε model is selected to obtain a first modified CFD model.

[0072] Exemplarily, for the stable layer, the k-ω SST model is selected to obtain a first modified CFD model.

[0073] Exemplarily, in the process of the bidirectional coupling, the preset Richardson number threshold can be modified based on the geographical features and the local details output by the CFD model; specifically, the preset Richardson number threshold is modified by using formula (2):

[0074] Ricadjusted=0.25*(1+0.1*z0 / H) (2)

[0075] In formula (2), Ricadjusted represents the modified preset Richardson number threshold, z0 represents the terrain roughness output by the CFD model, and H represents the terrain height, which is determined by the geographical features.

[0076] In an embodiment, for the Richardson number in a preset interval, a model weighted fusion method is used for transition; wherein, the preset interval can be [0.2, 0.3]; specifically, the turbulent flow result after fusion can be obtained by formula (3):

[0077] (3)

[0078] In formula (3), represents the turbulent flow result after fusion, represents the turbulent flow prediction result selected by the k-ε model, represents the turbulent flow prediction result obtained by selecting the k-ω SST model, and w is the fusion weight, w=(0.3-Ri) / 0.1.

[0079] Further, the method of training a random forest classifier can be used to update the turbulent flow model parameters of the CFD model based on the atmospheric stratification stability information; in the training of the random forest classifier, the Richardson number in the first predicted meteorological data and the corresponding meteorological data are input, and the corresponding relationship between the Richardson number or the Richardson number and the corresponding meteorological data and the corresponding turbulent flow model and / or the corresponding relationship between the Richardson number and the corresponding meteorological data and the prediction result is learned.

[0080] The above-mentioned embodiments of the present application can realize more accurate wind power prediction by bidirectional coupling of the WRF model and the CFD model, relative to the related art in which a single atmospheric model such as the WRF model and a wake model such as the Jensen model are used to simulate meteorological conditions to obtain predicted meteorological data. However, the inventors have found that, for offshore wind farm stations on the sea, the atmospheric stability output by the CFD model does not take into account the interaction of the ocean boundary layer, resulting in inaccurate wind speed characteristics output by the bidirectional coupling model, and ultimately inaccurate power prediction for the target wind farm station. When the CFD model ignores the relevant parameters of the ocean boundary layer, the output results of the CFD model are seriously distorted due to multiple influencing factors such as missing physical processes, invalid turbulence simulation, and energy transfer imbalance, thereby affecting the output state of the meteorological data. Specifically, first, the ocean boundary layer (OBL) is a thin layer about 50-200 meters below the sea surface. When the OBL is not considered in the CFD model, the error in the simulation of the wind speed profile is large, the error in the calculation of the surface shear stress is greater than 30%, and the prediction deviation of the ocean current velocity is 20-50%. Second, when the ocean stability parameter (such as the Richardson number) is not considered in the CFD model, the turbulent viscosity coefficient calculation is wrong, and the vortex diffusion ability is distorted. Moreover, for a simulation scheme that lacks the modification of the sea surface roughness to the sensible heat exchange coefficient and the latent heat exchange coefficient, the prediction deviation of the tropical cyclone path is greater than 100 kilometers, and the simulation error of the sea surface temperature (SST) is 1-3℃. Therefore, ignoring the ocean boundary layer parameters and not considering the interaction of the ocean boundary layer essentially breaks the energy and material circulation of the sea-air coupling system, and the error is amplified step by step in the CFD model through paths such as turbulence closure schemes, boundary conditions, and buoyancy effects. Therefore, a high-precision ocean CFD must include OBL parameters, otherwise unacceptable deviations will occur.

[0081] To solve the above problems, as an exemplary embodiment, the wind power prediction method further includes: if the geographical feature meets a preset ocean geographical feature, obtaining ocean meteorological data of the target wind farm station in the hybrid coordinate ocean model data based on the geographical feature; wherein the ocean meteorological data includes sea surface temperature and ocean mixed layer depth; determining the sea-air flux parameter of the WRF model based on the ocean mixed layer depth; taking a second preset time length as an update period, interpolating the sea surface temperature into the outer grid of the WRF model based on the spatio-temporal characteristics of the sea surface temperature, and adding the sea-air flux parameter to the first constraint condition to obtain a third constraint condition; driving the WRF model based on the third constraint condition to obtain third meteorological prediction data.

[0082] Exemplarily, the preset ocean geographical feature is a geographical feature corresponding to the offshore wind farm set on the sea.

[0083] Exemplarily, after obtaining the geographical feature, a topographic feature and a geomorphic feature are extracted in the geographical feature; when the topographic feature and the geomorphic feature satisfy preset marine topographic features and preset marine geomorphic features corresponding to the offshore wind farm arranged on the sea, it is confirmed that the geographical feature satisfies the preset marine geographical feature.

[0084] Exemplarily, after obtaining the geographical feature, a geographical position is extracted in the geographical feature; when the geographical position satisfies a preset geographical position corresponding to the offshore wind farm arranged on the sea, it is confirmed that the geographical feature satisfies the preset marine geographical feature.

[0085] In the embodiment, the marine meteorological data is taken as a constraint condition of the WRF model to realize the wind power prediction considering the marine condition by considering the interaction with the marine boundary layer.

[0086] Specifically, the marine meteorological data of the target wind power station is obtained in the Hybrid Coordinate Ocean Model (HYCOM) data based on the geographical feature, wherein the marine meteorological data includes sea surface temperature and marine mixed layer depth; further, the sea-air flux parameter of the WRF model is determined based on the marine mixed layer depth, and the spatiotemporal feature of the sea surface temperature is used to interpolate the sea surface temperature into the outer grid of the WRF model, so as to finally drive the WRF model to obtain the third meteorological prediction data. Exemplarily, the second preset time length is greater than or equal to 30 min.

[0087] Exemplarily, the second preset time length is 30 min.

[0088] As an exemplary embodiment, after the sea surface temperature is interpolated into the outer grid of the WRF model based on the spatial feature of the sea surface temperature, and the sea-air flux parameter is added to the first constraint condition, the wind power prediction method further comprises: extracting marine temperature data, air temperature data and marine mixed layer heat flux data in the third meteorological prediction data; calculating a temperature difference value based on the marine temperature data and the air temperature data; updating the turbulence model parameter of the CFD model based on the temperature difference value to obtain a second corrected CFD model; driving the second corrected CFD model based on the third meteorological prediction data to obtain the first meteorological prediction data.

[0089] In the embodiment, the third meteorological prediction data based on the WRF model output is used to correct the turbulence model of the CFD model; specifically, in the third meteorological prediction data, marine temperature data, air temperature data and marine mixed layer heat flux data are extracted; the temperature difference value is calculated based on the marine temperature data and the air temperature data; when the temperature difference value is greater than 2℃, it is determined that the boundary layer state is a strong unstable stratification, at this time, based on the strong unstable stratification, LES large eddy simulation is selected as the turbulence model of the CFD model; when the temperature difference value is greater than 0.5℃ and less than or equal to 2℃, it is determined that the boundary layer state is a weak unstable stratification, at this time, based on the weak unstable stratification, the k-ε model is selected as the turbulence prediction model; when the temperature difference value is less than-0.5℃, it is determined that the boundary layer state is a neutral layer, at this time, based on the neutral layer, the k-ω SST model is selected as the turbulence model of the CFD model.

[0090] As an exemplary embodiment, the method further comprises: extracting local wind speed change data in the first meteorological prediction data; updating the HYCOM data based on the local wind speed change data to obtain updated marine mixed layer heat flux data; and adding the updated marine mixed layer heat flux data to the first constraint condition.

[0091] In the embodiment, the local wind speed change caused by the wind turbine wake is fed back to the HYCOM data to more accurately update the marine mixed layer heat flux data by considering the influence of the wind turbine wake on the energy balance of the marine mixed layer.

[0092] The embodiment provides a wind power prediction device, as shown in the figure, comprising: Figure 2

[0093] The acquisition module 501 is configured to acquire geographical features of a target wind power station.

[0094] The model construction module 502 is configured to construct a bidirectional coupling meteorological prediction model based on the geographical features; wherein the bidirectional coupling meteorological prediction model comprises a WRF model and a CFD model.

[0095] The bidirectional coupling module 503 is configured to periodically perform bidirectional coupling on the WRF model and the CFD model based on the bidirectional coupling meteorological prediction model at a first preset time length to obtain first meteorological prediction data; in the process of bidirectional coupling, the WRF model is driven to perform meteorological prediction based on a first constraint condition to obtain second meteorological prediction data; the CFD model is driven to perform meteorological prediction based on the second meteorological prediction data and the geographical features to obtain first meteorological prediction data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first meteorological prediction data.

[0096] ​The downsizing module 504 is configured to downsize the first meteorological forecast data to obtain wind speed characteristics of the target wind power station.

[0097] The wind power prediction module 505 is configured to input the wind speed characteristics into a pre-constructed wind power prediction model to obtain a wind power prediction result.

[0098] It should be noted that the above modules and the examples and application scenarios realized by the corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments.

[0099] It should be noted that the above modules as part of the device can be realized by software or hardware, and the hardware environment includes a grid environment.

[0100] The embodiment of the application further provides a computer device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus, the memory is used for storing a computer program, and the processor is used for executing the method in any one of the above embodiments by running the computer program stored in the memory.

[0101] Figure 3 is a structural block diagram of an optional computer device according to the embodiment of the application, as shown in Figure 3 The computer device includes a processor 10, a communication interface 20, a memory 30 and a communication bus 40, wherein the processor 10, the communication interface 20 and the memory 30 complete mutual communication through the communication bus 40, and the memory 30 is used for storing a computer program.

[0102] The memory 30 is used for storing a computer program.

[0103] The processor 10 is used for executing the computer program stored in the memory 30 to realize the wind power prediction method in any one of the above embodiments.

[0104] Optionally, in the embodiment, the above communication bus can be a PCI (Peripheral Component Interconnect, peripheral component interconnect) bus, an EISA (Extended Industry Standard Architecture, extended industry standard architecture) bus or the like. The communication bus can be divided into an address bus, a data bus, a control bus and the like. For the convenience of representation, Figure 3 In the figure, only one thick line is used to represent the communication bus, but it does not mean that there is only one bus or one type of bus.

[0105] The communication interface is used for communication between the above computer device and other devices.

[0106] The memory can include a RAM and can also include a non-volatile memory, for example, at least one disk memory. Optionally, the memory can also be at least one storage device located remotely from the aforementioned processor.

[0107] The processor can be a general-purpose processor, which can include, but is not limited to, a CPU (Central Processing Unit), a NP (Network Processor), and the like; and can also be a DSP (Digital Signal Processing), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component.

[0108] Optionally, the specific examples in the embodiments can refer to the examples described in the above embodiments, and the embodiments will not be described here.

[0109] Those skilled in the art can understand that, Figure 3 The structure shown is only schematic, and the device for implementing the method of any one of the above embodiments can be a terminal device, which can be a smart phone (such as an Android phone, an IOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, and the like. Figure 3 This does not limit the structure of the above electronic device. For example, the terminal device can further include more or less components (such as a mesh interface, a display device, etc.) than Figure 3 shown, or have a different configuration from Figure 3 shown.

[0110] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device by a program, which can be stored in a computer readable storage medium, and the storage medium can include a flash disk, a ROM, a RAM, a magnetic disk or an optical disk, and the like.

[0111] As an exemplary embodiment, the present application also provides a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the method steps of any one of the embodiments when running.

[0112] Optionally, in the embodiment, the storage medium can be used to execute the program codes of the method steps of the embodiment.

[0113] Optionally, in the embodiment, the storage medium can be located on at least one of the plurality of grid devices in the grid shown in the above embodiment.

[0114] Optionally, in the embodiment, the storage medium is configured to store the method in the above embodiment.

[0115] Optionally, the specific examples in the embodiment can refer to the examples described in the above embodiment, which will not be repeated here.

[0116] Optionally, in the embodiment, the storage medium can include but is not limited to a U disk, a ROM, a RAM, a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0117] The serial numbers of the embodiments of the application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0118] The integrated units in the above embodiments, if realized in the form of software function units and sold or used as independent products, can be stored in the above computer-readable storage medium. Based on such understanding, the technical solutions of the application essentially or the parts that make contributions to the prior art or the whole or part of the technical solutions can be embodied in the form of software products. The computer software product is stored in the storage medium and includes a plurality of instructions for causing one or more computer devices (which can be personal computers, servers or grid devices, etc.) to execute all or part of the steps of the method in the above embodiment.

[0119] In several embodiments provided by the present application, it should be understood that the disclosed client can be implemented by other manners. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, indirect coupling or communication connection between units or modules, which can be electrical or other forms.

[0120] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on a plurality of grid units. According to actual needs, part or all of the units can be selected to achieve the purpose of the scheme provided in the embodiment.

[0121] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0122] In the above-described embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0123] The above is only the preferred embodiment of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principle of the present application, a number of improvements and refinements can be made, which should also be considered as the protection scope of the present application.

Claims

1. A wind power prediction method, characterized by, The wind power prediction method comprises: obtaining geographical features of a target wind power station; constructing a bidirectional coupling weather forecast model based on the geographical features; wherein the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; performing bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model with a first preset time length as a period to obtain first weather forecast data; in the process of bidirectional coupling, driving the WRF model to perform weather forecasting based on a first constraint condition to obtain second weather forecast data; driving the CFD model to perform weather forecasting based on the second weather forecast data and the geographical features to obtain the first weather forecast data; wherein the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; in constructing the bidirectional coupling weather forecast model, the WRF model and the CFD model are separately constructed based on geographical features, and a data interface for input data from the CFD model as a constraint condition is configured for the WRF model, and a data interface for input data from the WRF model as a constraint condition is configured for the CFD model; in the initial coupling process of the bidirectional coupling model, the geographical features are used as the first constraint condition to drive the WRF model to obtain the second weather forecast data; the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data; further, the geographical features and the first weather forecast data are used as the first constraint condition to drive the WRF model to obtain the second weather forecast data, and the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data; in the non-first coupling period, the geographical features and the first weather forecast data are used as the first constraint condition to drive the WRF model to obtain the second weather forecast data, and the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data, thereby completing bidirectional coupling of the WRF model and the CFD model; performing downscaling processing on the first weather forecast data to obtain wind speed characteristics of the target wind power station; inputting the wind speed characteristics into a pre-constructed wind power prediction model to obtain a wind power prediction result.

2. The wind power prediction method of claim 1, wherein, The bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model comprises: extracting a boundary layer height and a vertical wind shear profile from the second weather forecast data; inputting the boundary layer height and the vertical wind shear profile into the CFD model as a second constraint condition to perform transient wake meteorological simulation to obtain the first weather forecast data; extracting an equivalent roughness field from the first weather forecast data; wherein the equivalent roughness field is obtained by the CFD model through wake induction; inputting the equivalent roughness field into the WRF model as the first constraint condition for coupling.

3. The wind power prediction method according to claim 2, characterized in that, The boundary layer height and the vertical wind shear profile are input as second constraint conditions into the CFD model for transient wake meteorological simulation to obtain the first meteorological prediction data, including: The domain height of the CFD model is calculated based on the boundary layer height and the geographical feature; in the second meteorological prediction data, the Richardson number is extracted as boundary layer state information; The number of layers of the vertical stratification grid is determined based on the boundary layer state information; The domain height, the number of layers of the vertical stratification grid, the boundary layer height and the vertical wind shear profile are input as the second constraint conditions into the CFD model for transient wake meteorological simulation to obtain the first meteorological prediction data.

4. The wind power prediction method according to claim 3, characterized in that, The wind power prediction method further includes: A preset Richardson number threshold is obtained; Atmospheric stratification stability information is determined based on the Richardson number and the preset Richardson number threshold; The turbulence model parameters of the CFD model are updated based on the atmospheric stratification stability information to obtain a first corrected CFD model; The domain height, the number of layers of the vertical stratification grid, the boundary layer height and the vertical wind shear profile are input as the second constraint conditions into the first corrected CFD model for transient wake meteorological simulation to obtain the first meteorological prediction data.

5. The wind power prediction method of claim 1, wherein, The wind power prediction method further includes: If the geographical feature meets a preset marine geographical feature, marine meteorological data of the target wind power station is obtained based on the geographical feature in the hybrid coordinate ocean model data; wherein the marine meteorological data includes sea surface temperature and marine mixed layer depth; The sea-air flux parameters of the WRF model are determined based on the marine mixed layer depth; With a second preset time length as an update period, the sea surface temperature is interpolated into the outer grid of the WRF model based on the spatiotemporal characteristics of the sea surface temperature, and the sea-air flux parameters are added to the first constraint condition to obtain a third constraint condition; The WRF model is driven based on the third constraint condition to obtain third meteorological prediction data.

6. The wind power prediction method according to claim 5, characterized in that After the sea surface temperature is interpolated into the outer grid of the WRF model based on the spatial characteristics of the sea surface temperature, and the sea-air flux parameters are added to the first constraint condition, the wind power prediction method further includes: In the third meteorological prediction data, marine temperature data, air temperature data and marine mixed layer heat flux data are extracted; A temperature difference value is calculated based on the marine temperature data and the air temperature data; The turbulence model parameters of the CFD model are updated based on the temperature difference value to obtain a second corrected CFD model; The second corrected CFD model is driven based on the third meteorological prediction data to obtain the first meteorological prediction data.

7. The wind power prediction method according to claim 6, characterized in that, The wind power prediction method further includes: In the first meteorological prediction data, local wind speed change data is extracted; The hybrid coordinate ocean model data is updated based on the local wind speed change data to obtain updated marine mixed layer heat flux data; The updated marine mixed layer heat flux data is added to the first constraint condition.

8. A wind power prediction device, characterized by, The wind power prediction device includes: An acquisition module is configured to acquire geographical features of a target wind farm station; A model construction module is configured to construct a bidirectional coupling weather forecast model based on the geographical features; the bidirectional coupling weather forecast model comprises a WRF model and a CFD model; A bidirectional coupling module is configured to periodically perform bidirectional coupling of the WRF model and the CFD model based on the bidirectional coupling weather forecast model at a first preset time length to obtain first weather forecast data; in the process of bidirectional coupling, the WRF model is driven to perform weather forecast based on a first constraint condition to obtain second weather forecast data; the CFD model is driven to perform weather forecast based on the second weather forecast data and the geographical features to obtain the first weather forecast data; the first constraint condition is determined by the geographical features or the geographical features and the first weather forecast data; in the construction of the bidirectional coupling weather forecast model, the WRF model and the CFD model are separately constructed based on geographical features, and a data interface is configured for the WRF model to obtain input data from the CFD model as data in a constraint condition, and a data interface is configured for the CFD model to obtain input data from the WRF model as data in a constraint condition; in the initial coupling process of the bidirectional coupling model, the WRF model is driven by the geographical features as the first constraint condition to obtain the second weather forecast data; the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data; further, the WRF model is driven by the geographical features and the first weather forecast data as the first constraint condition to obtain the second weather forecast data, and the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data; in a non-first coupling cycle, the WRF model is driven by the geographical features and the first weather forecast data as the first constraint condition to obtain the second weather forecast data, and the CFD model is driven based on the second weather forecast data and the geographical features to obtain the first weather forecast data, thereby completing the bidirectional coupling of the WRF model and the CFD model; A downscaling module is configured to perform downscaling processing on the first weather forecast data to obtain wind speed features of the target wind farm station; and a wind power prediction module is configured to input the wind speed features into a pre-constructed wind power prediction model to obtain a wind power prediction result.

9. A computer device, comprising: An apparatus comprises: A memory and a processor in communication connection with each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the wind power prediction method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer readable storage medium stores computer instructions for causing a computer to perform the wind power prediction method according to any one of claims 1 to 7.

Images