A short-term prediction method, device and medium for wind turbine power generation

Abstract

The present application discloses a short-term prediction method, device and medium for wind turbine power generation, which mainly relates to the field of wind turbine power generation. The method first trains a meteorological model based on historical predicted meteorological data and historical actual meteorological data; then, trains a short-term prediction model based on historical actual meteorological data and historical short-term actual power; then, calls the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; finally, calls the short-term prediction model to obtain the target short-term predicted power corresponding to the second target meteorological data according to the second target meteorological data, so as to report. The method calls the meteorological model to obtain the second target meteorological data according to the first target meteorological data, and uses the second target meteorological data as the input of the short-term prediction model to obtain the target short-term predicted power for reporting, which effectively guarantees the reporting rate of the short-term predicted power.

Description

Technical Field

[0001] The present application relates to the field of wind turbine power generation, and in particular to a short-term prediction method, device and medium for wind turbine power generation. Background Art

[0002] The short-term predicted power is calculated based on meteorological data. It refers to the predicted power generation of the wind turbine a few days from the current time. It is the data that the wind farm must report to the power grid every day. The reporting rate and accuracy of this data are important criteria for the power grid to assess the wind farm. During the reporting process, the meteorological data is transmitted from the remote end where the meteorological data is measured to the wind farm end. When the meteorological data cannot be transmitted to the wind farm end due to emergencies such as network interruption and reverse isolation failure, the wind farm needs to start the fault-tolerant mechanism to obtain the short-term predicted power and report it.

[0003] Currently, when the meteorological data transmission is interrupted, the short-term forecast power cannot be obtained and therefore cannot be successfully reported.

[0004] It can be seen that how to improve the reporting rate of short-term predicted power is an urgent problem to be solved by those skilled in the art. Summary of the invention

[0005] The purpose of the present application is to provide a method, device and medium for short-term prediction of wind turbine power generation, which are used to perform short-term prediction of wind turbine power generation and improve the reporting rate and accuracy of short-term predicted power.

[0006] In order to solve the above technical problems, the present application provides a short-term prediction method for wind turbine power generation, comprising:

[0007] The meteorological model is trained based on historical forecast meteorological data and historical actual meteorological data; wherein the historical forecast meteorological data and historical actual meteorological data correspond to each other in time;

[0008] Training a short-term prediction model based on historical actual meteorological data and historical short-term actual power; wherein the historical short-term actual power corresponds to the historical actual meteorological data according to the time;

[0009] Calling a meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at a target time, and the target time is after the time corresponding to the historical short-term actual power;

[0010] The short-term prediction model is called to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, so as to report it.

[0011] Preferably, before calling the meteorological model to obtain the second target meteorological data corresponding to the first target meteorological data according to the meteorological model, the method further includes:

[0012] Determine whether to generate reporting data;

[0013] If not generated, the fault tolerance mechanism is started, and the step of calling the meteorological model to obtain the second target meteorological data according to the first target meteorological data is entered.

[0014] Preferably, the first target meteorological data is a plurality of groups;

[0015] Correspondingly, calling the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model includes:

[0016] Calling a meteorological model to obtain second target meteorological data corresponding to each first target meteorological data according to the first target meteorological data;

[0017] An average value is taken for all the second target meteorological data, and the average value is used as the second target meteorological data output by the meteorological model.

[0018] Preferably, after obtaining the target short-term predicted power, the method further includes:

[0019] The target short-term predicted power is reported as reporting data.

[0020] Preferably, before training the meteorological model, the method further includes:

[0021] Preprocess the historical forecast meteorological data and historical actual meteorological data.

[0022] Preferably, after calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, the method further includes:

[0023] Get the actual weather data at the target time;

[0024] calibrating the meteorological model according to the first target meteorological data and the actual meteorological data at the target time;

[0025] Obtain the actual power generation of the wind turbine at the target time;

[0026] The short-term prediction model is calibrated based on the actual power generation of the wind turbine at the target time and the actual meteorological data at the target time.

[0027] In order to solve the above technical problems, the present application also provides a short-term prediction device for wind turbine power generation, comprising:

[0028] The first training module is used to train a meteorological model based on historical forecast meteorological data and historical actual meteorological data; wherein the historical forecast meteorological data and the historical actual meteorological data correspond to each other in time;

[0029] The second training module is used to train a short-term prediction model based on historical actual meteorological data and historical short-term actual power; wherein the historical short-term actual power corresponds to the historical actual meteorological data in time;

[0030] A first calling module is used to call the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at a target time, and the target time is after the time corresponding to the historical short-term actual power;

[0031] The second calling module is used to call the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, so as to report it.

[0032] In order to solve the above technical problems, the present application also provides a short-term prediction device for wind turbine power generation, comprising:

[0033] Memory for storing computer programs;

[0034] A processor is used to implement the steps of the above-mentioned short-term prediction method of wind turbine power generation when executing a computer program.

[0035] In order to solve the above technical problems, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned short-term prediction method of wind turbine power generation are implemented.

[0036] The present application provides a short-term prediction method for wind turbine power generation, which first trains a meteorological model based on historical predicted meteorological data and historical actual meteorological data; then, trains a short-term prediction model based on historical actual meteorological data and historical short-term actual power; then, calls the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at the target time; finally, calls the short-term prediction model to obtain the target short-term predicted power corresponding to the second target meteorological data according to the second target meteorological data, so as to report. The method calls the meteorological model to obtain the second target meteorological data according to the first target meteorological data, and uses the second target meteorological data as the input of the short-term prediction model to obtain the target short-term predicted power, so as to report, effectively ensuring the reporting rate of the short-term predicted power. In addition, since the meteorological model is trained based on historical predicted meteorological data and historical actual meteorological data, and the first target meteorological data is the predicted meteorological data at the target time, the second target meteorological data obtained based on the first target meteorological data can be regarded as the latest and most accurate meteorological data predicted at the target time, and the accuracy of the target short-term predicted power obtained based on the second target meteorological data is effectively guaranteed.

[0037] In addition, the short-term prediction device and medium for wind turbine power generation provided in the present application correspond to the short-term prediction method for wind turbine power generation, and the effects are as above. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A flow chart of a short-term prediction method for wind turbine power generation provided in this application;

[0040] Figure 2 A structural diagram of a short-term prediction device for wind turbine power generation provided in this application;

[0041] Figure 3 A structural diagram of another short-term prediction device for wind turbine power generation provided in this application. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0043] The core of this application is to provide a short-term prediction method, device and medium for wind turbine power generation, which are used to perform short-term prediction of wind turbine power generation and improve the reporting rate and accuracy of short-term predicted power.

[0044] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0045] It should be noted that the short-term prediction method for wind turbine power generation provided in the present application is applicable to wind farms, especially the short-term prediction of wind turbine power generation in the event of an emergency, wherein the emergency includes but is not limited to the inability to obtain the predicted meteorological data due to network failure, reverse isolation failure, etc., and the lack of the latest and most accurate predicted meteorological data, resulting in the inability to start the short-term prediction model.

[0046] Figure 1 This is a flow chart of a short-term prediction method for wind turbine power generation provided in this application. Figure 1 As shown, the method includes:

[0047] S1: Train the meteorological model based on historical forecast meteorological data and historical actual meteorological data.

[0048] In this embodiment, the historical forecast meteorological data refers to the meteorological data prepared in advance and predicted based on the geographical location of the wind turbine, and the historical actual meteorological data corresponds to the historical forecast meteorological data in time. It can be understood that meteorological data is a set of data reflecting the weather, mainly including wind speed, wind direction, temperature, air pressure, etc. Since the longer the interval between the time of meteorological forecast and the actual time of the predicted meteorological data, the lower the accuracy of the predicted meteorological data, therefore, in order to ensure the accuracy of the meteorological model, the historical forecast meteorological data should be predicted at a time close to the time corresponding to the historical actual meteorological data, for example, it can be predicted three days before the time corresponding to the historical actual meteorological data, because in these three days, a set of meteorological data corresponding to the time of the historical actual meteorological data is predicted every day, therefore, for each set of historical actual meteorological data, there are three sets of historical forecast meteorological data corresponding to it, that is, when training the meteorological model, for each set of historical actual meteorological data, three sets of historical forecast meteorological data can be used to combine with it for training respectively to optimize the meteorological model. In addition, when training the meteorological model, the historical predicted meteorological data is used as input and the historical actual meteorological data is used as output. After trial and error, the meteorological model obtained using the xgboost algorithm has the highest accuracy.

[0049] S2: Train the short-term prediction model based on historical actual meteorological data and historical short-term actual power.

[0050] In this embodiment, the historical short-term actual power is the measured historical power generation of the wind turbine, and each historical short-term actual power has a set of historical predicted meteorological data corresponding to it. Obviously, in order to ensure the accuracy of the short-term prediction model, the time corresponding to the historical actual meteorological data should be shorter than the time corresponding to the historical short-term actual power. For example, after obtaining a set of historical actual meteorological data, the power generation of the wind turbine the next day can be predicted based on the set of historical actual meteorological data, and the measured power generation of the wind turbine the next day can be obtained as the historical short-term actual power corresponding to the set of historical actual meteorological data. In addition, when training the short-term prediction model, the historical actual meteorological data is used as input and the historical short-term actual power is used as output. After trying, the accuracy of the short-term prediction model obtained using the xgboost algorithm is the highest.

[0051] S3: calling the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model.

[0052] In this embodiment, the first target meteorological data is the predicted meteorological data at the target time. It should be noted that the target time is after the time corresponding to the historical short-term actual power. Since the target short-term predicted power is predicted based on the latest and most accurate meteorological data predicted at the target time, and it is currently impossible to guarantee that the first target meteorological data is the latest and most accurate meteorological data predicted at the target time, it is necessary to process the first target meteorological data, and obtain the second target meteorological data corresponding to the first target meteorological data according to the meteorological model. At this time, the second target meteorological data can be regarded as the latest and most accurate meteorological data predicted at the target time for short-term prediction. In order to improve the accuracy of the second target meteorological data, the first target meteorological data should be the meteorological data predicted at a time closer to the target time. For example, the first target meteorological data can include the meteorological data predicted at the target time three days before the target time. Since the meteorological data of the next week is usually predicted every day, the first target meteorological data at this time includes three groups of meteorological data as input to the meteorological model. Obviously, the first target meteorological data can also have only one group of meteorological data, and this embodiment does not limit this.

[0053] In the specific implementation, the meteorological model is first called to obtain the second target meteorological data corresponding to the first target meteorological data based on the input first target meteorological data. At this time, the second target meteorological data can be regarded as the latest and most accurate meteorological data of the predicted wind turbine at the target time, so that it can be used as the input of the short-term prediction model to obtain the short-term predicted power of the wind turbine at the target time.

[0054] S4: calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, so as to report it.

[0055] Since the short-term prediction model is based on historical actual meteorological data and historical short-term actual power, after inputting the second target meteorological data, the short-term prediction model can be called to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data. At this time, the target short-term prediction power obtained is the short-term prediction power of the wind turbine at the target time. After obtaining the target short-term prediction power, a reporting file is generated according to the target short-term prediction power for reporting to ensure the reporting rate of the short-term prediction power.

[0056] The present embodiment provides a short-term prediction method for wind turbine power generation, which first trains a meteorological model based on historical predicted meteorological data and historical actual meteorological data; then, trains a short-term prediction model based on historical actual meteorological data and historical short-term actual power; then, calls the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at the target time; finally, calls the short-term prediction model to obtain the target short-term predicted power corresponding to the second target meteorological data according to the second target meteorological data, so as to report. The method calls the meteorological model to obtain the second target meteorological data according to the first target meteorological data, and uses the second target meteorological data as the input of the short-term prediction model to obtain the target short-term predicted power, so as to report, effectively ensuring the reporting rate of the short-term predicted power. In addition, since the meteorological model is trained based on historical predicted meteorological data and historical actual meteorological data, and the first target meteorological data is the predicted meteorological data at the target time, the second target meteorological data obtained based on the first target meteorological data can be regarded as the latest and most accurate meteorological data for the predicted target time, and the accuracy of the target short-term predicted power obtained based on the second target meteorological data is effectively guaranteed.

[0057] Before the above embodiment calls the meteorological model to obtain the second target meteorological data corresponding to the first target meteorological data according to the meteorological model, it is necessary to first determine whether a reporting file for short-term predicted power has been generated. It is understandable that the model needs to be called for short-term power prediction only when a reporting file for short-term predicted power has not been generated. This step includes:

[0058] Determine whether to generate reporting data;

[0059] If not generated, the fault tolerance mechanism is started, and the step of calling the meteorological model to obtain the second target meteorological data according to the first target meteorological data is entered.

[0060] In the specific implementation, every day during the preset time period, a check is performed to see whether the short-term predicted power to be reported is generated. If it has been generated, the detection is terminated. If it has not been generated, the fault-tolerant mechanism is started to perform short-term power prediction and generate a reporting file based on the obtained short-term predicted power for reporting.

[0061] Before calling the model for short-term power prediction, this embodiment first determines whether to generate reporting data, and triggers the fault-tolerant mechanism when no reporting data is generated, and enters the step of calling the meteorological model to obtain the second target meteorological data based on the first target meteorological data. When the reporting data has been generated, there is no need to perform short-term power detection, which can effectively save time.

[0062] In the above embodiment, it is mentioned that the first target meteorological data can be one group or multiple groups. When the first target meteorological data is multiple groups, correspondingly, the second target meteorological data corresponding to each first target meteorological data obtained by calling the meteorological model is also multiple groups. At this time, the second target meteorological data needs to be processed and then output. This step includes:

[0063] Calling a meteorological model to obtain second target meteorological data corresponding to each first target meteorological data according to the first target meteorological data;

[0064] An average value is taken for all the second target meteorological data, and the average value is used as the second target meteorological data output by the meteorological model.

[0065] In a specific implementation, if the first target meteorological data is the meteorological data predicted at the target time three days before the target time, the first target meteorological data at this time includes three groups of meteorological data, so the second target meteorological data obtained by calling the meteorological model is also three groups. When the meteorological model outputs the second target meteorological data, in order to facilitate user query and use, all the second target meteorological data can be directly averaged or assigned different weights to obtain a weighted average. Specifically, all the parameters with the same name in the second target meteorological data can be directly averaged or assigned different weights to obtain a weighted average. For example, the weighted average of the temperature parameters in the three groups of second target meteorological data can be obtained according to the different weights assigned, and the obtained average value can be used as the temperature parameter in the second target meteorological data that is finally output for output. The output of the average value of all the second target meteorological data proposed in this embodiment is only a preferred implementation method. Obviously, all the second target meteorological data can also be directly output.

[0066] When there are multiple groups of first target meteorological data, there are also multiple groups of second target meteorological data obtained based on the first target meteorological data. When outputting the second target meteorological data, this embodiment takes the average value of all the second target meteorological data and then outputs them. Compared with directly outputting all the second target meteorological data, when the number of groups of second target meteorological data is large, taking the average value for output is more convenient for users to query and use.

[0067] On the basis of the above embodiment, after obtaining the target short-term predicted power, this embodiment further includes: reporting the target short-term predicted power as reporting data.

[0068] Specifically, after obtaining the target short-term predicted power, a reporting file will be generated based on the target short-term predicted power. At this time, the target short-term predicted power is the reporting data in the reporting file, and reporting the reporting file can meet the reporting rate of the short-term predicted power.

[0069] After obtaining the target short-term predicted power, this embodiment reports the target short-term predicted power as reporting data to meet the reporting rate of the short-term predicted power.

[0070] On the basis of the above embodiment, this embodiment further includes: pre-processing the historical forecast meteorological data and the historical actual meteorological data before training the meteorological model.

[0071] In a specific implementation, since the historical forecast meteorological data and the historical actual meteorological data correspond to each other in time, the historical forecast meteorological data and the historical actual meteorological data can be sorted in time for the convenience of analysis and processing. During preprocessing, since each group of historical forecast meteorological data has the historical actual meteorological data corresponding thereto, the difference between each parameter in the historical forecast meteorological data and the historical actual meteorological data corresponding thereto can be calculated for analysis. After the calculation is completed, one or more parameters can be selected as a reference. When the absolute value of the selected parameter difference is greater than the preset threshold value, it is considered that the accuracy of the historical forecast meteorological data corresponding to the parameter difference is low, and the historical forecast meteorological data and the historical actual meteorological data corresponding to the parameter difference will be eliminated to avoid affecting the accuracy of the trained meteorological model. It can be understood that when comparing the absolute value of the selected parameter difference with the preset threshold value, for different parameters, such as temperature and wind speed, the threshold values ​​used are different because the units of the parameters are different. In addition, the specific value of the threshold value is set according to the actual situation, which is not explained in this embodiment.

[0072] Before training the meteorological model, this embodiment pre-processes the historical predicted meteorological data and the historical actual meteorological data, and eliminates the historical predicted meteorological data with low accuracy and the corresponding historical actual meteorological data, so as to improve the accuracy of the trained meteorological model and effectively ensure the accuracy of the short-term predicted power.

[0073] On the basis of the above embodiment, after calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, this embodiment calibrates the meteorological model and the short-term prediction model respectively. This step includes:

[0074] Get the actual weather data at the target time;

[0075] calibrating the meteorological model according to the first target meteorological data and the actual meteorological data at the target time;

[0076] Obtain the actual power generation of the wind turbine at the target time;

[0077] The short-term prediction model is calibrated based on the actual power generation of the wind turbine at the target time and the actual meteorological data at the target time.

[0078] Specifically, for the meteorological model, first obtain the actual meteorological data at the target time, and then adjust the meteorological model so that the second target meteorological data output by the meteorological model based on the first target meteorological data is consistent with the actual meteorological data at the target time, so as to optimize the meteorological model; for the short-term prediction model, first obtain the actual power generation of the wind turbine at the target time, and then adjust the short-term prediction model so that the target short-term prediction power output by the short-term prediction model based on the actual meteorological data at the target time is consistent with the actual power generation of the wind turbine at the target time, so as to optimize the short-term prediction model.

[0079] After obtaining the target short-term predicted power, this embodiment calibrates the meteorological model according to the first target meteorological data and the actual meteorological data at the target time, and calibrates the short-term prediction model according to the actual power generation of the wind turbine at the target time and the actual meteorological data at the target time, thereby improving the accuracy of the meteorological model and the short-term prediction model and facilitating subsequent use.

[0080] In the above embodiments, the short-term prediction method of wind turbine power generation is described in detail, and the present application also provides an embodiment corresponding to the short-term prediction device of wind turbine power generation. It should be noted that the present application describes the embodiments of the device part from two perspectives, one is based on the functional module perspective, and the other is based on the hardware perspective.

[0081] Figure 2 This is a structural diagram of a short-term prediction device for wind turbine power generation provided in this application. Figure 2 As shown, the device comprises:

[0082] The first training module 10 is used to train a meteorological model based on historical forecast meteorological data and historical actual meteorological data; wherein the historical forecast meteorological data and the historical actual meteorological data correspond to each other in time;

[0083] The second training module 11 is used to train a short-term prediction model based on historical actual meteorological data and historical short-term actual power; wherein the historical short-term actual power corresponds to the historical actual meteorological data in time;

[0084] The first calling module 12 is used to call the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at a target time, and the target time is after the time corresponding to the historical short-term actual power;

[0085] The second calling module 13 is used to call the short-term prediction model to obtain the target short-term predicted power corresponding to the second target meteorological data according to the second target meteorological data, so as to report it.

[0086] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, which will not be repeated here.

[0087] The short-term prediction device for wind turbine power generation provided in this embodiment trains the meteorological model according to the historical predicted meteorological data and the historical actual meteorological data through the first training module; trains the short-term prediction model according to the historical actual meteorological data and the historical short-term actual power through the second training module; calls the meteorological model through the first calling module to obtain the second target meteorological data corresponding to the first target meteorological data according to the meteorological model; and calls the short-term prediction model through the second calling module to obtain the target short-term predicted power corresponding to the second target meteorological data according to the second target meteorological data, so as to report. The device calls the meteorological model to obtain the second target meteorological data according to the first target meteorological data, and uses the second target meteorological data as the input of the short-term prediction model to obtain the target short-term predicted power, so as to report, effectively ensuring the reporting rate of the short-term predicted power. In addition, since the meteorological model is trained according to the historical predicted meteorological data and the historical actual meteorological data, and the first target meteorological data is the predicted meteorological data at the target time, therefore, the second target meteorological data obtained based on the first target meteorological data can be regarded as the latest and most accurate meteorological data predicted at the target time, and the accuracy of the target short-term predicted power obtained based on the second target meteorological data is effectively guaranteed.

[0088] Figure 3 A structural diagram of a short-term prediction device for wind turbine power generation provided in another embodiment of the present application, such as Figure 3 As shown, the device comprises:

[0089] A memory 20, for storing computer programs;

[0090] The processor 21 is used to implement the steps of the short-term prediction method of wind turbine power generation mentioned in the above embodiment when executing the computer program.

[0091] The short-term prediction device for wind turbine power generation provided in this embodiment may include but is not limited to a smart phone, a tablet computer, a laptop computer or a desktop computer.

[0092] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 can be implemented in at least one hardware form of a digital signal processor (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 21 may be integrated with a graphics processing unit (GPU), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 21 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0093] The memory 20 may include one or more computer-readable storage media, which may be non-transitory. The memory 20 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201, wherein, after the computer program is loaded and executed by the processor 21, it can implement the relevant steps of the short-term prediction method of wind turbine power generation disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. Data 203 may include but is not limited to the first target meteorological data, etc.

[0094] In some embodiments, the short-term prediction device for wind turbine power generation may further include a display screen 22 , an input / output interface 23 , a communication interface 24 , a power supply 25 , and a communication bus 26 .

[0095] Those skilled in the art will understand that Figure 3 The structure shown in the figure does not constitute a limitation on the short-term prediction device for wind turbine power generation, and may include more or fewer components than those shown in the figure.

[0096] The short-term prediction device for wind turbine power generation provided in this embodiment includes a memory and a processor. When the processor executes the program stored in the memory, it can implement the above-mentioned short-term prediction method for wind turbine power generation, and the effect is the same as above.

[0097] Finally, the present application also provides an embodiment corresponding to a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the short-term prediction method for wind turbine power generation as described in the above method embodiment are implemented.

[0098] It is understandable that if the method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0099] The computer-readable storage medium provided in this embodiment includes the above-mentioned short-term prediction method for wind turbine power generation, and the effect is the same as above.

[0100] The above is a detailed introduction to the short-term prediction method, device and medium for wind turbine power generation provided by the present application. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the embodiments can be referenced to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of the present application.

[0101] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A short-term prediction method for wind turbine power generation, characterized in that: include: Training a meteorological model based on historical forecast meteorological data and historical actual meteorological data; wherein the historical forecast meteorological data and the historical actual meteorological data correspond to each other in time; Training a short-term prediction model according to the historical actual meteorological data and the historical short-term actual power; wherein the historical short-term actual power corresponds to the historical actual meteorological data according to the time; Calling the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at a target time, and the target time is after the time corresponding to the historical short-term actual power; calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, so as to report it; After calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, the method further includes: Acquire actual meteorological data at the target time; calibrating the meteorological model according to the first target meteorological data and actual meteorological data at the target time; Acquire the actual power generation of the wind turbine at the target time; Calibrate the short-term prediction model according to the actual power generation of the wind turbine at the target time and the actual meteorological data at the target time; Wherein, the first target meteorological data is a plurality of groups; Correspondingly, calling the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model includes: calling the meteorological model to obtain the second target meteorological data corresponding to each of the first target meteorological data according to the first target meteorological data; taking an average value of all the second target meteorological data, and using the average value as the second target meteorological data output by the meteorological model; Before calling the meteorological model to obtain the second target meteorological data corresponding to the first target meteorological data according to the meteorological model, the method further includes: Determine whether to generate reporting data; If not generated, the fault tolerance mechanism is started, and the step of calling the meteorological model to obtain the second target meteorological data according to the first target meteorological data is entered.

2. The short-term prediction method for wind turbine power generation according to claim 1, characterized in that: After obtaining the target short-term predicted power, the method further includes: The target short-term predicted power is reported as the reporting data.

3. The short-term prediction method for wind turbine power generation according to claim 1, characterized in that: Before training the meteorological model, it also includes: The historical forecast meteorological data and the historical actual meteorological data are preprocessed.

4. A short-term prediction device for wind turbine power generation, characterized in that: include: A first training module is used to train a meteorological model based on historical forecast meteorological data and historical actual meteorological data; wherein the historical forecast meteorological data and the historical actual meteorological data correspond to each other in time; A second training module is used to train a short-term prediction model based on the historical actual meteorological data and the historical short-term actual power; wherein the historical short-term actual power corresponds to the historical actual meteorological data according to the time; a first calling module, configured to call the meteorological model to obtain second target meteorological data corresponding to the first target meteorological data according to the meteorological model; wherein the first target meteorological data is the predicted meteorological data at a target time, and the target time is after the time corresponding to the historical short-term actual power; A second calling module, used for calling the short-term prediction model to obtain the target short-term prediction power corresponding to the second target meteorological data according to the second target meteorological data, so as to report; Wherein, the short-term prediction device for wind turbine power generation is also used for: Acquire actual meteorological data at the target time; calibrating the meteorological model according to the first target meteorological data and actual meteorological data at the target time; Acquire the actual power generation of the wind turbine at the target time; Calibrate the short-term prediction model according to the actual power generation of the wind turbine at the target time and the actual meteorological data at the target time; Wherein, the first target meteorological data is a plurality of groups; The first calling module is used for: calling the meteorological model to obtain the second target meteorological data corresponding to each of the first target meteorological data according to the first target meteorological data; taking an average value of all the second target meteorological data, and using the average value as the second target meteorological data output by the meteorological model; Wherein, the short-term prediction device for wind turbine power generation is also used for: Determine whether to generate reporting data; If not generated, the fault tolerance mechanism is started, and the step of calling the meteorological model to obtain the second target meteorological data according to the first target meteorological data is entered.

5. A short-term prediction device for wind turbine power generation, characterized in that: comprising a memory for storing a computer program; A processor is used to implement the steps of the short-term prediction method for wind turbine power generation as described in any one of claims 1 to 3 when executing the computer program.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the short-term prediction method for wind turbine power generation according to any one of claims 1 to 3 are implemented.

Images