Ultrahigh altitude area wind speed vertical extrapolation method based on random forest model

Abstract

The invention relates to the field of wind energy resource assessment, and provides an ultrahigh altitude area wind speed vertical extrapolation method based on a random forest model, which comprises the following steps: collecting and preprocessing anemometer tower data to be subjected to wind speed extrapolation in an ultrahigh altitude area and corresponding horizontal plane total irradiance data; key feature factors are constructed based on the anemometer tower data and the horizontal plane total irradiance data; constructing a training set data set and a prediction set data set of a random forest model based on the preprocessed anemometer tower data, the horizontal plane total irradiance data and the key feature factors; and training a random forest model by using the training set data set, and inputting the prediction set data set into the trained random forest model to realize wind speed extrapolation. According to the method, the strong nonlinear fitting capability and integrated learning advantages of the random forest model are utilized, the complex nonlinear relation in the wind speed data can be automatically learned, the wind speed change rules of different heights are accurately captured, and the wind speed extrapolation precision in the ultrahigh altitude area is higher.

Description

Technical Field

[0001] The present invention relates to the field of wind energy resource assessment, and in particular to a vertical extrapolation method for wind speed in ultra-high altitude areas based on a random forest model. Background Art

[0002] Given today's urgent need for clean energy, the development and utilization of wind energy, a sustainable and widely distributed energy source, has garnered significant attention. With the widespread adoption of large-blade, high-tower wind turbines, an increasing number of wind farm field observations do not reach the turbine hub height, necessitating vertical wind speed extrapolation. Accurately estimating wind speed and wind power density at hub height is crucial for wind energy resource assessment, directly impacting key aspects of wind power project feasibility studies, turbine selection, and power generation estimation.

[0003] At present, in the field of vertical extrapolation of wind speed, traditional methods mostly use the power-law formula to calculate wind speed based on the wind shear index. The power-law formula is derived based on the assumption that the wind speed profile in the atmospheric boundary layer conforms to the power-law distribution law. However, in the actual atmospheric environment, the distribution of wind speed may be affected by a combination of complex factors and does not strictly follow the power-law distribution. This makes the wind speed extrapolation based on this formula have certain uncertainties, especially in ultra-high altitude areas, where this traditional method exposes obvious limitations.

[0004] Ultra-high altitude regions possess unique geographical and climatic characteristics. Their surface is primarily covered by low, barren grasslands and bare ground, significantly reducing surface roughness. Furthermore, atmospheric stratification instability caused by strong solar radiation frequently occurs, weakening the wind speed gradient between the surface and upper atmosphere. Furthermore, the low air density at ultra-high altitudes enhances topographic dynamics, which can easily induce acceleration of airflow due to topographic compression. These combined factors severely compromise the accuracy of wind speed extrapolation using power-law methods at ultra-high altitudes. This low-precision wind speed extrapolation can lead to inappropriate wind turbine selection, failure to fully utilize local wind energy resources, or overestimation of wind energy potential, resulting in wasted project investment. In summary, existing power-law-based vertical wind speed extrapolation methods are unable to meet the high-precision wind speed prediction requirements for wind energy resource assessment in ultra-high altitude regions. New methods are urgently needed to improve the accuracy of vertical wind speed extrapolation. Summary of the Invention

[0005] In response to the above-mentioned problems, the present invention provides a vertical extrapolation method for wind speed in ultra-high altitude areas based on a random forest model to improve the accuracy of vertical extrapolation of wind speed in ultra-high altitude areas.

[0006] The present invention provides a method for vertically extrapolating wind speed in ultra-high altitude areas based on a random forest model, comprising the following steps:

[0007] Collect and pre-process wind tower data and corresponding horizontal total irradiance data for wind speed extrapolation in ultra-high altitude areas;

[0008] Constructing key characteristic factors based on the wind tower data and the horizontal total irradiance data;

[0009] Based on the pre-processed wind tower data, horizontal total irradiance data and key characteristic factors, the training set data group and prediction set data group of the random forest model are constructed;

[0010] The training set data group is used to train a random forest model, and the prediction set data group is input into the trained random forest model to achieve wind speed extrapolation.

[0011] In some embodiments, the wind tower data includes three layers of wind speed, one layer of wind direction, and one layer of air temperature data; wherein the wind direction layer must include the standard deviation of wind direction pulsation.

[0012] In some embodiments, pre-processing the wind tower data includes verifying and interpolating the wind tower data according to relevant regulations and specifications.

[0013] In some embodiments, the horizontal surface total irradiance data refers to the horizontal surface total irradiance data at each time step during the same period of the wind measurement data provided by ERA5 reanalysis data adjacent to the wind tower data.

[0014] In some embodiments, the key characteristic factor is atmospheric stability.

[0015] In some embodiments, constructing key characteristic factors includes:

[0016] Based on the wind tower data and the horizontal total irradiance data, the atmospheric instability classification is performed using the wind speed-radiation-wind direction standard method;

[0017] The classified atmospheric instability is subjected to heat encoding and converted into a numerical form that can be recognized by the random forest model.

[0018] In some embodiments, the atmospheric instability classification using a combined wind speed-radiation-wind direction standard method includes:

[0019] A preliminary classification is made based on the hourly wind direction fluctuation standard deviation data from the wind tower combined with the atmospheric stability classification preliminary determination table;

[0020] According to the total horizontal irradiance data, the corresponding final category determination table is selected;

[0021] The final atmospheric stability is determined based on the preliminary atmospheric stability and multi-layer average wind speed, combined with the corresponding final category determination table.

[0022] In some embodiments, the training set data set and prediction set data set for constructing the random forest model include:

[0023] The wind speed, wind speed ratio, atmospheric stability, total horizontal irradiance, temperature and time at different heights were selected from the collected data to construct the training set data set and prediction set data set of the random forest model;

[0024] The training set data group includes input variables and output variables. The input variables should include the wind speed of the second floor below the height to be pushed, the ratio of the wind speed of the second floor below the height to be pushed to the wind speed of the third floor below the height to be pushed, atmospheric stability, total horizontal irradiance, temperature and time. The output variable is the corrected wind speed of the first floor below the height to be pushed;

[0025] The prediction set data set only includes input variables, including the wind speed of the first layer below the height to be predicted, the ratio of the wind speed of the first layer below the height to be predicted to the wind speed of the second layer below the height to be predicted, atmospheric stability, total horizontal irradiance, temperature and time;

[0026] The selected input variables are normalized to convert data of different dimensions and orders of magnitude into the same scale range.

[0027] In some embodiments, the calculation formula for the corrected wind speed at the next floor below the altitude to be pushed is as follows:

[0028]

[0029] Where V' hub-1 is the wind speed of the first layer under the height to be pushed after correction, α1 is the wind speed of the second layer under the height to be pushed V hub-2 , Wind speed V at the third floor below the height to be pushed hub-3 The resulting overall wind shear index, Z hub-1 、Z hub-2 Refers to the height of 1 layer and 2 layers below the hub height respectively.

[0030] In some embodiments, when constructing a random forest model, hyperparameter tuning is performed based on GridSearchCV grid search, and the parameters to be tuned include the number of decision numbers, the maximum depth of decision numbers, the minimum number of samples required for node partitioning, and the minimum number of samples required for leaf nodes.

[0031] In some embodiments, the prediction set data set is input into a trained random forest model to implement wind speed extrapolation, including:

[0032] The trained random forest model predicts the corrected wind speed V' at the required height based on the learned rules hub;

[0033] Based on the corrected wind speed V' hub Calculate the wind speed V at the height to be pushed hub , the calculation formula is as follows:

[0034]

[0035] Where: α2 is the wind speed V of the next layer based on the height to be pushed hub-1 , Wind speed V at the second floor below the height to be pushed hub-2 The resulting overall wind shear index, Z hub 、Z hub-1 They refer to the hub height and the height one level below the hub height respectively.

[0036] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0037] 1. Compared with the traditional power-law extrapolation model that assumes that the vertical distribution of wind speed conforms to a specific function form, the present invention comprehensively considers the impact of complex meteorological factors in ultra-high altitude areas on the vertical distribution of wind speed, and utilizes the powerful nonlinear fitting ability and integrated learning advantages of the random forest model. It can automatically learn the complex nonlinear relationship in wind speed data, accurately capture the changing laws of wind speed at different altitudes, and achieve higher wind speed extrapolation accuracy in ultra-high altitude areas, thereby realizing accurate vertical extrapolation of wind speed in ultra-high altitude areas, providing reliable data support for related fields such as wind energy resource development and weather forecasting, and has important application value and broad application prospects.

[0038] 2. Compared with traditional methods that rely solely on wind speed for extrapolation, this invention effectively integrates multiple factors such as wind speed, wind direction, temperature, radiation, and atmospheric instability, so that the input characteristics of the random forest model more comprehensively reflect the actual situation of the wind field, breaking the limitation of traditional methods that rely only on a few parameters, and making the wind resource assessment results closer to reality. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flowchart of a method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model is provided in an embodiment of the present invention.

[0040] Figure 2 A schematic structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0042] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but merely represents selected embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0043] like Figure 1 As shown, an embodiment of the present invention proposes a vertical extrapolation method for wind speed in ultra-high altitude areas based on a random forest model. By introducing the nonlinear model of random forest, multiple factors such as wind speed, temperature, radiation, and stability are effectively integrated to improve the accuracy of vertical extrapolation of wind speed in ultra-high altitude areas, and provide reliable technical support for the assessment of wind energy resources in ultra-high altitude areas.

[0044] The method for vertical extrapolation of wind speed in ultra-high altitude areas based on the random forest model includes the following steps:

[0045] S100: Collect and pre-process wind tower data and corresponding horizontal total irradiance data for wind speed extrapolation in ultra-high altitude areas.

[0046] In some embodiments, the wind tower data includes three layers of wind speed, one layer of wind direction, and one layer of temperature data; the wind direction layer must include the variable of wind direction fluctuation standard deviation; in this embodiment, the wind tower data has a duration of at least one year and a temporal resolution of one hour. Preprocessing the wind tower data includes verifying and interpolating the wind tower data according to relevant regulations and specifications to ensure the accuracy and completeness of the wind tower data;

[0047] In some embodiments, the horizontal surface total irradiance data refers to the horizontal surface total irradiance data at each time step (eg, hourly) during the same period of the wind data provided by ERA5 reanalysis data adjacent to the wind tower data.

[0048] S200: Constructing key characteristic factors based on the wind tower data and the horizontal total irradiance data.

[0049] In some embodiments, the key characteristic factor is atmospheric stability. Atmospheric stability is an important factor affecting atmospheric wind shear. It refers to the tendency and degree to which the air mass returns to or moves away from its original equilibrium position due to atmospheric stratification after the air is disturbed in the vertical direction, which directly affects the vertical convection of the near-ground atmosphere. The higher the temperature, the stronger the vertical convection, the more unstable the atmospheric surface, the lower the atmospheric stability in ultra-high altitude areas, and the smaller the wind shear value.

[0050] In some embodiments, constructing key characteristic factors includes:

[0051] Based on the wind tower data and the total horizontal irradiance data, the atmospheric instability is classified using the wind speed-radiation-wind direction standard method, and the atmospheric stability is divided into six categories: A (strong instability), B (unstable), C (weak instability), D (neutral), E (relatively stable), and F (stable);

[0052] These six types of atmospheric instability are heat-encoded and converted into numerical forms that can be recognized by the random forest model, so that the impact of atmospheric instability, a key factor, on the vertical extrapolation of wind speed can be fully considered in the subsequent random forest model.

[0053] The advantage of using the combined wind speed-radiation-wind direction standard method for atmospheric instability classification is that it integrates multiple variables such as wind speed, wind direction, and radiation, and comprehensively considers the unique meteorological characteristics of ultra-high altitude areas. At the same time, the parameters required for its calculation, except for the total horizontal irradiance data, are basically included in the wind tower data, which can effectively improve the applicability of the invention.

[0054] Specifically, the atmospheric instability classification using the wind speed-radiation-wind direction standard method includes:

[0055] First, a preliminary classification is performed based on the hourly wind direction fluctuation standard deviation data of the wind tower in combination with the atmospheric stability classification preliminary determination table; in this embodiment, the atmospheric stability classification preliminary determination table is shown in Table 1;

[0056] Then, the corresponding final category determination table is selected according to the total horizontal irradiance data; in this embodiment, the total horizontal irradiance ≥ 120W / m 2 Table 2 is used for final category determination, and the total irradiance on the horizontal plane is less than 120W / m 2 Table 3 is used for final category determination; the division boundaries and corresponding relationships can be adjusted as needed.

[0057] Finally, the final atmospheric stability is determined based on the preliminary atmospheric stability and multi-layer average wind speed, combined with the corresponding final category determination table.

[0058] Table 1, preliminary determination table of atmospheric stability classification:

[0059] Wind direction fluctuation standard deviation Preliminary assessment of atmospheric stability <![CDATA[22.5≤σ A ]]> A(strongly unstable) <![CDATA[17.5≤σ A <22.5]]> B (unstable) <![CDATA[12.5≤σ A <17.5]]> C (weakly unstable) <![CDATA[7.5≤σ A <12.5]]> D (neutral) <![CDATA[3.8≤σ A <7.5]]> E (more stable) <![CDATA[σ A <3.8]]> F(stable)

[0060] Table 2, total irradiance on horizontal surface ≥ 120W / m 2 Applicable atmospheric stability final determination table:

[0061]

[0062]

[0063] Table 3, total irradiance on horizontal surface <120W / m 2 Applicable atmospheric stability final determination table:

[0064]

[0065] S300: Based on the pre-processed wind tower data, horizontal total irradiance data, and key characteristic factors, a training set data group and a prediction set data group of the random forest model are constructed.

[0066] First, we select key parameters such as wind speed at different heights, wind speed ratio, atmospheric stability, total horizontal irradiance, temperature and time from the collected data to construct the training set data set and prediction set data set of the random forest model. hub Represents the hub height wind speed to be estimated, V hub-1 、V hub-2 、V hub-3 They refer to the wind speeds at the 1st, 2nd and 3rd floors below the hub height respectively.

[0067] The training set data group includes input variables and output variables, and the input variables should include the wind speed V of the second floor below the height to be pushed. hub-2 , the ratio of the wind speed of the second layer below the height to be pushed to the wind speed of the third layer below the height to be pushed V hub-2 / V hub-3 , atmospheric stability, total horizontal irradiance, temperature and time, the output variable is the corrected wind speed V' at the next layer below the height to be estimated hub-1 , which is calculated as follows:

[0068]

[0069] Where α1 is based on V hub-2 、V hub-3 The resulting overall wind shear index, Z hub-1 、Z hub-2 Refers to the height of 1 layer and 2 layers below the hub height respectively.

[0070] The prediction set data set only contains input variables, including the wind speed V of the next layer below the height to be predicted. hub-1, the ratio of the wind speed of the first layer below the height to be pushed to the wind speed of the second layer below the height to be pushed V hub-1 / V hub-2 , atmospheric stability, total horizontal irradiance, temperature and time.

[0071] Secondly, each selected input variable is normalized to convert data of different dimensions and orders of magnitude to the same scale range. In this embodiment, the MinMaxScaler in the Python package sklearn is used to normalize each selected input variable to the interval [0, 1].

[0072] S400: Using the training set data group to train a random forest model, and inputting the prediction set data group into the trained random forest model to achieve wind speed extrapolation. Specifically:

[0073] Model training: The training set data group is input into the random forest model for training. As for the specific training method, the common training method of the random forest model can be selected and will not be described here. Among them, when constructing the random forest model, this embodiment performs hyperparameter tuning based on GridSearchCV grid search. The parameters to be tuned include the number of decision numbers, the maximum depth of the decision number, the minimum number of samples required for node partitioning, and the minimum number of samples required for leaf nodes.

[0074] Wind speed extrapolation: The prediction set data set is input into the trained random forest model, and the trained random forest model predicts the corrected wind speed V' at the required height based on the learned rules. hub , then based on the corrected wind speed V' hub Calculate the wind speed V at the height to be pushed hub In this embodiment, the calculation formula is as follows:

[0075]

[0076] Where: α2 is based on V hub-1 、V hub-2 The resulting overall wind shear index, Z hub 、Z hub-1 They refer to the hub height and the height one level below the hub height respectively.

[0077] From the above, it can be seen that the present invention has significant advantages and innovations in the field of wind energy resource assessment, which are mainly reflected in the following aspects:

[0078] 1. Compared with the traditional power-law extrapolation model that assumes that the vertical distribution of wind speed conforms to a specific function form, the present invention comprehensively considers the impact of complex meteorological factors in ultra-high altitude areas on the vertical distribution of wind speed, and utilizes the powerful nonlinear fitting ability and integrated learning advantages of the random forest model. It can automatically learn the complex nonlinear relationship in wind speed data, accurately capture the changing laws of wind speed at different altitudes, and achieve higher wind speed extrapolation accuracy in ultra-high altitude areas, thereby realizing accurate vertical extrapolation of wind speed in ultra-high altitude areas, providing reliable data support for related fields such as wind energy resource development and weather forecasting, and has important application value and broad application prospects.

[0079] 2. Compared with traditional methods that rely solely on wind speed for extrapolation, this invention effectively integrates multiple factors such as wind speed, wind direction, temperature, radiation, and atmospheric instability, so that the input characteristics of the random forest model more comprehensively reflect the actual situation of the wind field, breaking the limitation of traditional methods that rely only on a few parameters, and making the wind resource assessment results closer to reality.

[0080] Based on the same technical concept, an embodiment of the present invention further provides an electronic device that can implement the process of the vertical extrapolation method of wind speed in ultra-high altitude areas based on the random forest model provided in the above embodiment of the present invention. In one embodiment, the electronic device can be a server, or a terminal device or other electronic device. Figure 2 As shown, the electronic device may include:

[0081] At least one processor, and a memory connected to the at least one processor. The embodiment of the present invention does not limit the specific connection medium between the processor and the memory. Figure 2 The example in this article is that the processor and memory are connected via a bus. Figure 2 The connections between the other components are shown in bold lines, which are only for illustration and not intended to be limiting. The bus can be divided into address bus, data bus, control bus, etc. Figure 2 The processor is represented by a single thick line, but this does not mean that there is only one bus or only one type of bus. Alternatively, the processor can also be called a controller, without any limitation on the name.

[0082] In an embodiment of the present invention, the memory stores instructions that can be executed by at least one processor. The at least one processor can execute the vertical extrapolation method of wind speed in ultra-high altitude areas based on the random forest model discussed above by executing the instructions stored in the memory. The processor can implement Figure 2 The functions of each module in the device shown.

[0083] Among them, the processor is the control center of the device, which can use various interfaces and lines to connect the various parts of the entire control device, and monitor the device as a whole by running or executing instructions stored in the memory and calling data stored in the memory, the various functions of the device and processing data.

[0084] In an optional design, the processor may include one or more processing units, and the processor may integrate an application processor and a modem processor, wherein the application processor primarily processes the operating system, user interface, and application programs, and the modem processor primarily processes wireless communications. It is understood that the modem processor may not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip, or in some embodiments, they may be implemented on separate chips.

[0085] The processor can be a general-purpose processor, such as a CPU, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. A general-purpose processor can be a microprocessor or any conventional processor. In conjunction with the steps of the method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model disclosed in the embodiments of the present invention, the steps can be directly implemented as execution by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.

[0086] As a non-volatile computer-readable storage medium, memory can be used to store non-volatile software programs, non-volatile computer executable programs and modules. Memory can include at least one type of storage medium, for example, can include flash memory, hard disk, multimedia card, card-type memory, random access memory (Random Access Memory, RAM), static random access memory (Static Random Access Memory, SRAM), programmable read-only memory (Programmable Read Only Memory, PROM), read-only memory (Read Only Memory, ROM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), magnetic memory, disk, optical disk, etc. Memory is any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the embodiment of the present invention can also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0087] By designing and programming the processor, the code corresponding to the method for vertically extrapolating wind speed in ultra-high altitude areas based on a random forest model described in the aforementioned embodiment can be embedded in the chip, enabling the chip to execute the steps of the method in the aforementioned embodiment when running. Designing and programming the processor is well known to those skilled in the art and will not be further described here.

[0088] Based on the same inventive concept, an embodiment of the present invention further provides a storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer executes a vertical extrapolation method for wind speed in ultra-high altitude areas based on a random forest model discussed above.

[0089] In some optional embodiments, the present invention also provides various aspects of a method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model, which can also be implemented in the form of a program product, which includes program code. When the program product is run on an apparatus, the program code is used to enable the control device to execute the steps of a method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to various exemplary embodiments of the present invention described above in this specification.

[0090] It should be noted that although several units or subunits of the device are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to an embodiment of the present invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of a unit described above can be further divided into multiple units to be embodied. In addition, although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps.

[0091] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0092] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a server, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0093] Program code for performing the operations of the present invention may be written using any combination of one or more programming languages, including object-oriented programming languages such as Java, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0094] Where a remote computing device is involved, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0095] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0096] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0097] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A vertical extrapolation method for wind speed in ultra-high altitude areas based on a random forest model, characterized in that: include: Collect and pre-process wind tower data and corresponding horizontal total irradiance data for wind speed extrapolation in ultra-high altitude areas; Constructing key characteristic factors based on the wind tower data and the horizontal total irradiance data; Based on the pre-processed wind tower data, horizontal total irradiance data and key characteristic factors, the training set data group and prediction set data group of the random forest model are constructed; The training set data group is used to train a random forest model, and the prediction set data group is input into the trained random forest model to achieve wind speed extrapolation.

2. The vertical extrapolation method of wind speed in ultra-high altitude areas based on the random forest model according to claim 1 is characterized in that: The wind tower data includes three layers of wind speed, one layer of wind direction, and one layer of temperature data; wherein, the wind direction layer must include the standard deviation of wind direction pulsation; the preprocessing of the wind tower data includes checking and interpolating the wind tower data according to relevant regulations and specifications.

3. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on the random forest model according to claim 1 is characterized in that: The horizontal total irradiance data refers to the horizontal total irradiance data for each time step during the same period of the wind measurement data provided by the ERA5 reanalysis data adjacent to the wind tower data.

4. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 1, characterized in that: The key characteristic factor is atmospheric stability.

5. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 4 is characterized in that: The key characteristic factors are constructed, including: Based on the wind tower data and the horizontal total irradiance data, the atmospheric instability classification is performed using the wind speed-radiation-wind direction standard method; The classified atmospheric instability is subjected to heat encoding and converted into a numerical form that can be recognized by the random forest model.

6. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 5, characterized in that: The atmospheric instability classification using the wind speed-radiation-wind direction standard method includes: A preliminary classification is made based on the hourly wind direction fluctuation standard deviation data from the wind tower combined with the atmospheric stability classification preliminary determination table; According to the total horizontal irradiance data, the corresponding final category determination table is selected; The final atmospheric stability is determined based on the preliminary atmospheric stability and multi-layer average wind speed, combined with the corresponding final category determination table.

7. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 1, characterized in that: The training set data group and prediction set data group for constructing the random forest model include: The wind speed, wind speed ratio, atmospheric stability, total horizontal irradiance, temperature and time at different heights were selected from the collected data to construct the training set data set and prediction set data set of the random forest model; The training set data group includes input variables and output variables. The input variables should include the wind speed of the second floor below the height to be pushed, the ratio of the wind speed of the second floor below the height to be pushed to the wind speed of the third floor below the height to be pushed, atmospheric stability, total horizontal irradiance, temperature and time. The output variable is the corrected wind speed of the first floor below the height to be pushed; The prediction set data set only includes input variables, including the wind speed of the first layer below the height to be predicted, the ratio of the wind speed of the first layer below the height to be predicted to the wind speed of the second layer below the height to be predicted, atmospheric stability, total horizontal irradiance, temperature and time; The selected input variables are normalized to convert data of different dimensions and orders of magnitude into the same scale range.

8. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 7, characterized in that: The calculation formula for the corrected wind speed at the first floor below the height to be pushed is as follows: Where V' hub-1 is the wind speed of the first layer under the height to be pushed after correction, α1 is the wind speed of the second layer under the height to be pushed V hub-2 , Wind speed V at the third floor below the height to be pushed hub-3 The resulting overall wind shear index, Z hub-1 , Z hub-2 Refers to the height of 1 layer and 2 layers below the wheel hub height respectively.

9. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 1, characterized in that: When building a random forest model, hyperparameter tuning is performed based on GridSearchCV grid search. The parameters to be tuned include the number of decisions, the maximum depth of decisions, the minimum number of samples required for node partitioning, and the minimum number of samples required for leaf nodes.

10. The method for vertical extrapolation of wind speed in ultra-high altitude areas based on a random forest model according to claim 1, characterized in that: The prediction set data set is input into the trained random forest model to implement wind speed extrapolation, including: The trained random forest model predicts the corrected wind speed V' at the required height based on the learned rules hub ; Based on the corrected wind speed V' hub Calculate the wind speed V at the height to be pushed hub , the calculation formula is as follows: Where: α2 is the wind speed V of the next layer based on the height to be pushed hub-1 , Wind speed V at the second floor below the height to be pushed hub-2 The resulting overall wind shear index, Z hub , Z hub-1 They refer to the hub height and the height one level below the hub height respectively.

Images