Method and device for predicting future wind speed based on historical wind speed data

Abstract

The invention discloses a method, device and equipment for predicting future wind speed based on historical wind speed data and a storage medium. The method for predicting the future wind speed based on the historical wind speed data comprises the steps of obtaining wind speed data of a wind resource in n years; counting monthly data characteristics of the wind speed data of the n years according to the wind speed data of the n years; and constructing wind resource standard data in the next year according to the monthly data characteristics of the wind speed data in the n years. According to the technical scheme, the wind speed data of the corresponding month in the future is predicted through the distribution characteristics of the historical wind speed data, the particle dimension of the month data and the historical monthly data with the maximum similarity.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation, and in particular to a method, device, equipment and storage medium for predicting future wind speed based on historical wind speed data. Background Art

[0002] In recent years, renewable energy such as wind power has achieved rapid development worldwide. However, the randomness, intermittency, and volatility of wind power mean that large-scale wind power may bring severe challenges to the safe and economic operation of the power system after being connected to the grid. At the same time, the development of power trading, refined operation and maintenance, etc., puts forward higher requirements for optimized production. At present, some of the needs of power grid dispatching and day-ahead and day-midday trading can be met by short-term power forecasting systems. In order to better formulate wind farm production assessment indicators, maintenance plans, and long-term business strategies, it is particularly important to improve the level of future medium- and long-term power generation assessment.

[0003] According to the medium- and long-term power generation forecast and evaluation, the time range with low power generation and unfavorable meteorological conditions can be reliably predicted, thereby providing sufficient time planning window for maintenance. Reliable medium- and long-term power generation forecast and evaluation can reasonably include wind power reliable capacity in the power generation plan, assist power grid dispatching in activating the start-stop peak-shaving capacity of large units, and improve dispatching economy and power grid security. In addition, reliable medium- and long-term power generation forecast and evaluation can also provide a basis for medium- and long-term power trading, further promoting wind power consumption. Reliable medium- and long-term wind power generation assessment technology can effectively grasp the fluctuation law of wind power generation, improve the level of wind power consumption, and provide a basic guarantee for fine scientific operation and safe and economical operation of power systems. Therefore, the technical methods for studying medium- and long-term power generation assessment are of great practical significance for decision-making in production operation and maintenance, power trading, and power grid dispatching.

[0004] At present, in the medium- and long-term wind power generation forecast, which is mainly based on the forecast of power generation in the next 12 months, the use of long-term historical return forecast data is not effective, which affects the accuracy of the forecast results. This can be summarized as the following factors: Although the historical return forecast data of local wind resources is large, the data does not necessarily directly and prominently reflect the situation of the wind farm location; the wind farm is usually established for a short time, and the amount of data is small, which is not enough to be used specifically for the correction of historical return forecast data; some properties of the historical return forecast data itself need to be mined through special means, but many current forecasting algorithms do not independently include this part of the content. The above problems have caused most algorithms to be either limited in accuracy or limited in use by factors such as data quality and dimension. How to properly solve the above problems has become a topic that the industry needs to solve urgently. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for predicting future wind speed based on historical wind speed data, which is used to predict the wind speed data of the corresponding month in the future by using the distribution characteristics of the historical wind speed data, the granularity dimension of the monthly data and the historical monthly data with the maximum similarity.

[0006] According to a first aspect of the present invention, a method for predicting future wind speed based on historical wind speed data is provided, the method for predicting future wind speed based on historical wind speed data comprising:

[0007] Get wind speed data for n years of wind resources;

[0008] According to the wind speed data of the n years, monthly data characteristics of the wind speed data of the n years are counted;

[0009] According to the monthly data characteristics of the wind speed data of the n years, the standard wind resource data for the next year is constructed.

[0010] In one embodiment, it includes:

[0011] Counting monthly data characteristics of wind speed data for the n years;

[0012] According to the characteristics of all monthly data in the same month, a two-parameter Weibull distribution is fitted;

[0013] The data distribution of each month in the wind speed data of the n years is matched with the two-parameter Weibull distribution to obtain the most similar monthly data of each month;

[0014] The most similar monthly data of each month constitutes the standard wind resource data for the next year.

[0015] In one embodiment, it includes:

[0016] Using a preset error calculation method to match the two-parameter Weibull distribution, the preset error calculation method includes a mean absolute error method or a mean absolute percentage error method;

[0017] According to the array composed of scale parameters and deformation parameters, the corresponding mean absolute error is calculated, and the monthly data with the smallest mean absolute error is counted, and the most similar monthly data of each month is formed by traversing twelve months.

[0018] In one embodiment, it includes:

[0019] The Euclidean distance of scale parameter, deformation parameter and mean absolute error value was calculated by cluster analysis, and the compactness of cluster analysis was evaluated based on the silhouette method.

[0020] In one embodiment, it further includes:

[0021] Cluster analysis is performed based on any combination of the scale parameter, the scale parameter and the deformation parameter, and the scale parameter, the deformation parameter and the average wind speed.

[0022] In one embodiment, it further includes:

[0023] The wind power generation capacity in the next year is calculated based on the standard wind resource data in the next year and the a priori power curve formula.

[0024] According to a second aspect of the present invention, there is provided a device for predicting future wind speed based on historical wind speed data, comprising:

[0025] The acquisition module is used to obtain the wind speed data of wind resources for n years;

[0026] A statistical module, used for collecting monthly data characteristics of the wind speed data of the n years according to the wind speed data of the n years;

[0027] The construction module is used to construct the wind resource standard data for the next year according to the monthly data characteristics of the wind speed data of the n years.

[0028] According to a third aspect of the present invention, there is provided an electronic device, the electronic device comprising: a communication interface, a processor, and a memory;

[0029] The memory is used to store program instructions, and when the program instructions are executed by the processor that is communicatively connected to the memory through the communication interface, any of the above-mentioned methods for predicting future wind speed based on historical wind speed data is implemented.

[0030] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a computer (e.g., a processor in a computer), implement any of the above-mentioned methods for predicting future wind speed based on historical wind speed data.

[0031] In summary, the present invention provides a method and device for predicting future wind speed based on historical wind speed data, the method comprising: obtaining wind speed data of wind resources for n years; based on the wind speed data of n years, statistically analyzing the monthly data characteristics of the wind speed data of n years; and constructing standard wind resource data for the next year based on the monthly data characteristics of the wind speed data of n years. The technical solution of the present application predicts the wind speed data of the corresponding month in the future based on the distribution characteristics of the historical wind speed data, the granularity dimension of the monthly data, and the historical monthly data with the maximum similarity.

[0032] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0033] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 A flow chart of a method for predicting future wind speed based on historical wind speed data provided by an embodiment of the present invention;

[0036] Figure 2 A flowchart of another method for predicting future wind speed based on historical wind speed data provided by an embodiment of the present invention;

[0037] Figure 3 A flowchart of another method for predicting future wind speed based on historical wind speed data provided by an embodiment of the present invention;

[0038] Figure 4 A structural diagram of a device for predicting future wind speed based on historical wind speed data provided by an embodiment of the present invention;

[0039] Figure 5 A structural diagram of an electronic device provided by an embodiment of the present invention;

[0040] Figure 6 An effect diagram of an optimized contour method after cluster analysis for predicting future wind speed based on historical wind speed data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0042] It should be noted that, in this article, 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 "include", "comprise" 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, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0043] like Figure 1 As shown, the present invention provides a method for predicting future wind speed based on historical wind speed data, and the method for predicting future wind speed based on historical wind speed data includes:

[0044] In step S11, wind speed data of wind resources for n years is obtained;

[0045] In step S12, monthly data features of the wind speed data of the n years are counted based on the wind speed data of the n years;

[0046] In step S13, the wind resource standard data for the next year is constructed according to the monthly data characteristics of the wind speed data for n years.

[0047] In one embodiment, wind speed data for n years is obtained, which is the basic input data for the whole process. N years is usually a relatively long time (such as 20 years) to ensure data representativeness and reliability of prediction. After obtaining historical data, monthly characteristic statistics are performed on the wind speed data for n years. The monthly data characteristics can extract the statistical variation law of wind speed in each month, including the monthly average wind speed, the distribution pattern of wind speed (such as through Weibull distribution fitting), etc. Based on the aforementioned monthly data characteristics, standard wind resource data for the next year is constructed. The wind speed characteristics of each month are fitted in combination with a probability density function (such as Weibull distribution) to generate a data set that can represent the next year, which can reflect the wind speed fluctuations and changing trends in the next year, and provide a reference for subsequent power generation forecasts.

[0048] Based on the analysis of n years of data, representative statistical seasonality is found, and it is combined with the standard data of wind resources to find the (Weibull distribution) probability density function for each statistical season in the next year to meet the needs of calculation and use.

[0049] Regarding the characteristics, distribution and fitting of wind speed data: Depending on the conditions of different wind fields, there are many possible distribution characteristics of wind speed, such as Weibull, Gamma, Kappa and other distribution methods. The wind speed (long-term forecast) data represented by Weibull distribution is also considered according to the two-parameter Weibull distribution in actual use. The probability density function of the two-parameter Weibull distribution is as follows:

[0050]

[0051] Where v is the independent variable wind speed, λ>0 and k>0 are called scale parameter and deformation parameter, respectively, which are related to the width and shape of the distribution curve.

[0052] The selection and construction of wind resource standard data. In the construction process, several different probability density functions are used, which are represented by the following notations (where y∈{1, 2, …, 20}, m∈{1, 2, …, 12}):

[0053] PDF y，m : Probability density function of the mth month of the forecast data in the yth year

[0054] PDF c，m : The probability density function of the mth month accumulated over all years

[0055] PDF W，m : Probability density function of the mth month in the wind resource standard data

[0056] Specific construction process:

[0057] 1 Collect all 20 years of data monthly, find out the monthly data characteristics, and fit the Weibull distribution PDFy，m .

[0058] 2 Perform Weibull fitting on all the data of the corresponding month to obtain PDF c，m .

[0059] 3 For each month m0∈{1, 2, …, 12}, find a year y0∈{1, 2, …, 20} such that Close enough to and will As Finally, we get a set of years corresponding to 12 months, and then use the data of the months corresponding to these years to form the standard data of wind resources. That is to say, in the standard data of wind resources for the next year constructed using 20 years of data, the data of January is for year y1, which makes PDF c，1 and Very similarly, the data for February is for year y2, which makes and It is close enough, and so on, we get:

[0060]

[0061] The year-month set {(y1, 1), (y2, 2), ..., (y 12 , 12)} is the standard wind resource data representing the wind speed conditions in the next year.

[0062] For the approximate standard, error calculation methods such as mean absolute error (MAE) or mean absolute percentage error (MAPE) can be used, or manual inspection can be used. This patent uses MAE as the judgment standard. and PDF c，i , i = 1, 2, ..., 12, get the corresponding scale parameter λ and deformation parameter k, and then respectively form an array of these two parameters

[0063]

[0064] Find the corresponding MAE value for all possible year groups {y1, y2, ..., y 12} to find a group Make these two MAEs reach the minimum. When there is no MAE that meets the conditions (i.e. one is the lowest but the other is not), select the group with the smallest deformation parameter from the year group with the smallest scale parameter MAE, and then select the group with the smallest scale parameter from the year group with the smallest deformation parameter MAE. Take these two groups and perform all subsequent processes separately and select the one with better results.

[0065] When calculating the standard data of wind resources, y，m The scale parameters, deformation parameters and average wind speed of the corresponding months |y∈{1, 2, …, 20}, m∈{1, 2, …12}} are clustered and analyzed to obtain the statistical seasonality of wind speed for use in the prediction algorithm.

[0066] The process of cluster analysis, the calculation method and process of statistical seasonality, the calculation of statistical seasonality uses the k-means algorithm, that is, a given set of data (x1, x2, ..., x n ) (where each data point is an m-dimensional vector), divided into k≤n clusters C={C1,C2,…,C k}, so that the points in each cluster are close to the center of the cluster μ i , i = 1, 2, ..., k, the sum of squared Euclidean distances is minimal:

[0067]

[0068] Among them, the center

[0069] The evaluation of clustering effect is based on the silhouette method, which is to quantify the similarity between points in the same cluster and points in other clusters, judge the rationality of the clustering of data points, and reflect the compactness and separation of clusters. The definition of the silhouette value (or silhouette coefficient) used in the silhouette method consists of the following parts:

[0070] 1) In cluster C i The average distance between point i in the cluster and all other points in the cluster is

[0071] where d(i, j) is the cluster C l The distance between midpoints i and j.

[0072] 2) For the point i mentioned in 1), the minimum average distance between it and all the points in all other clusters, that is, the average distance between it and the nearest point in other clusters, is

[0073]

[0074] where d(i, j) is the cluster C i Point i in and another cluster C k The distance of point j in .

[0075] 3) Based on a(i) and b(i) above, define the contour value as

[0076]

[0077] Here, the range of s(i) is [-1, 1]. Values ​​greater than 0 and closer to 1 indicate better clustering of data points. Conversely, values ​​less than 0 and closer to -1 indicate a greater likelihood that data points are classified into the wrong cluster. (In the definition, in order to prevent the number of clusters from increasing dramatically, the profile value of a cluster with only one sample is set to 0.)

[0078] At the same time, the silhouette value can also help determine the optimal number of clusters. The silhouette coefficient is calculated using the following formula:

[0079]

[0080] in It represents the average silhouette value of the data after being divided into k clusters. When k is within a limited optional range, the silhouette coefficient represents the best clustering effect when a specific k is selected within this range, so the corresponding k can be considered as the optimal number of clusters.

[0081] The general process for establishing statistical seasonality using the above techniques is as follows:

[0082] 1) Fit the time series data of each month of each year in the 20-year data to the Weibull distribution and obtain the corresponding distribution parameters.

[0083] 2) Based on three different variable combinations of scale parameter, scale parameter & shape parameter, and scale parameter & shape parameter & average wind speed, one-dimensional, two-dimensional, and three-dimensional cluster analysis are performed respectively, and the number of clusters is optimized according to the silhouette method

[0084] 3) Arrange the cluster analysis results by year and month to produce three charts

[0085] 4) According to the distribution law and statistical information of each month in each year, the 20-year wind speed data is divided into 2-4 statistical seasons according to the month. Finally, a set of statistical seasonal month distributions is obtained:

[0086]

[0087] Among them, i+j+…+l=12.

[0088] For example, the historical return forecast wind speed data of a wind farm in the past 20 years is obtained after going through steps 1 and 2 above. Figure 6 .

[0089] From left to right are the results of one-dimensional, two-dimensional, and three-dimensional cluster analysis and optimization using the contour method. Based on the height and width of the Weibull distribution of the months corresponding to the grids of different colors, it can be judged that the months of the white grids belong to the "low statistical season", while the months corresponding to the black grids belong to the "high statistical season". It should be noted that the high and low here do not refer to the high and low wind speeds, but to the height and width of the wind speed distribution curve. According to the situation shown in the above three figures, in the three cluster analyses, the three months of July, August, and September, which are high statistical seasons in all years, are selected as high statistical season months that can represent all years, while the other nine months are selected as low statistical season months. Therefore, the final statistical seasonal month distribution is

[0090] {(7, 8, 9), (1, 2, 3, 4, 5, 6, 10, 11, 12)}

[0091] By combining the wind resource standard data with statistical seasonality, we can obtain the wind resource standard data {(y1, 1), (y2, 2), …, (y 12 , 12)}, and the statistical seasonal month distribution Then, according to the statistical seasonal month distribution, the data of the corresponding months in the wind resource standard data are taken out in turn, and Weibull fitting is performed to obtain the probability density function PDF of each statistical season. i , PDF j , …, PDF l .

[0092] The power generation in the corresponding statistical season can be calculated using the existing power curve:

[0093]

[0094] The distribution and wind speed information of each statistical season can also be put into the modeling software and combined with the measured data to complete the calculation of the predicted power generation; or it can be used to correct and complete the measured data in order to obtain more accurate results in other prediction algorithms.

[0095] The technical solution in this embodiment predicts the future corresponding monthly wind speed data based on the distribution characteristics of historical wind speed data and the granularity dimension of monthly data using the historical monthly data with the greatest similarity.

[0096] In one embodiment, Figure 2 As shown, step 13 includes the following steps S21-S24:

[0097] In step S21, monthly data features of the wind speed data of the n years are counted;

[0098] In step S22, a two-parameter Weibull distribution is fitted based on the monthly data characteristics of all the same months;

[0099] In step S23, the data distribution of each month in the wind speed data of the n years is matched to the two-parameter Weibull distribution to obtain the most similar monthly data of each month;

[0100] In step S24, the most similar monthly data of each month is used to form the wind resource standard data for the next year.

[0101] In one embodiment, the monthly data features are statistically analyzed after the historical wind speed data is obtained, and the wind speed data of n years are statistically analyzed on a monthly basis, and the wind speed characteristics of each month are extracted to form representative monthly data features. The monthly data features may include average wind speed, wind speed variation range and distribution mode. Fitting the two-parameter Weibull distribution, according to the monthly data features of all the same months in the historical data, fitting the two-parameter Weibull distribution, using the two parameters (scale parameter and shape parameter) of the Weibull distribution, establish a probability density function that can accurately reflect the wind speed distribution of the month. The two-parameter Weibull distribution is widely used because it can fit the wind speed data well. Through this fitting, a mathematical model that is consistent with the monthly wind speed features can be obtained. Matching the monthly data with the two-parameter Weibull distribution, after obtaining the two-parameter Weibull distribution, the data distribution of each month in the wind speed data of n years is matched with the two-parameter Weibull distribution, and the historical monthly data most similar to the distribution can be found. The matching process may use some error measurement method (such as mean absolute error or mean absolute percentage error) to determine which month in the historical data is closest to the fitted Weibull distribution. To form the standard wind resource data for the next year, the most similar historical monthly data for each month are combined to form the standard wind resource data for the next year. The wind speed forecast data set for the next year not only covers the characteristics of different months, but also maintains the natural fluctuation law of each representative month in the historical data, which enhances the reliability and representativeness of the data.

[0102] Monthly characteristics are extracted from historical data to provide the statistical basis required for analysis; monthly data characteristics are converted into a probability distribution model through two-parameter Weibull distribution fitting to help quantitatively describe the monthly distribution characteristics of wind speed; typical months in historical data are aligned with the fitted distribution to ensure that the selected monthly data can accurately reflect the changing trend of monthly wind speed in the future; forecast data for the next year is formed by combining the most similar data of each month. This standard data set is both representative and predictive, providing a scientific basis for medium- and long-term forecasts of wind power generation.

[0103] In one embodiment, Figure 3 As shown, step 23 includes the following steps S31-S32:

[0104] In step S31, a preset error calculation method is used to match the two-parameter Weibull distribution, and the preset error calculation method includes a mean absolute error method or a mean absolute percentage error method;

[0105] In step S32, the corresponding mean absolute error is calculated based on the array composed of the scale parameter and the deformation parameter, and the monthly data with the smallest mean absolute error is counted, and the most similar monthly data of each month is formed by traversing the twelve months.

[0106] In one embodiment, each month in the historical data is matched with a two-parameter Weibull distribution to find the monthly data that best fits the future wind speed forecast. By calculating and comparing the errors, the data closest to the predicted distribution can be selected for each month. An error calculation method is selected to match the historical monthly data with the two-parameter Weibull distribution. The preset error calculation method includes the mean absolute error (MAE) or the mean absolute percentage error (MAPE). The purpose of selecting the error calculation method is to evaluate the similarity between the historical data and the predicted distribution. By calculating the error, the fitting quality of the data of a certain month with the Weibull distribution model can be quantitatively measured. In the matching process, the scale parameter and the deformation parameter are used to describe the characteristics of the two-parameter Weibull distribution. After fitting the wind speed data of each month in the historical data to the Weibull distribution, the scale parameter and the deformation parameter of each month are obtained respectively. Through the above parameters, an array is formed, and the mean absolute error between the data of each month and the two-parameter Weibull distribution is obtained according to the array, so as to quantify the deviation of the data of each month from the standard distribution. After calculating the error, the monthly data with the smallest error is selected. The monthly data with the smallest error represents the historical data that is closest to the two-parameter Weibull distribution, that is, the monthly data that best fits the future prediction model. In this process, each historical monthly data is traversed to find the month with the smallest error in turn, and finally the most similar monthly data set is formed, covering the wind speed conditions for the next 12 months.

[0107] By selecting the error calculation method (MAE or MAPE), a quantitative standard is provided for the matching process to measure the degree of fit between the historical data of each month and the forecast model. MAE and MAPE are commonly used error calculation methods, of which MAE is more suitable for processing absolute value errors, while MAPE is suitable for evaluating percentage deviations. Obtaining the error according to the scale parameter and the deformation parameter is the key to fitting the historical monthly data to the Weibull distribution. The scale parameter and the deformation parameter are the core features of the Weibull distribution and determine the distribution shape of the wind speed data. Through these two parameters, the actual data can be accurately compared with the model, and the fitting effect can be measured by the error. The monthly data with the smallest error is selected to ensure that the data of each month conforms to the characteristics of the Weibull distribution model as much as possible, thereby ensuring the representativeness of the forecast for the next year and forming a collection of the most similar monthly data. The technical solution in this embodiment ensures the similarity of the data of each month by parameter matching, thereby improving the reliability of the forecast model.

[0108] In one embodiment, Figure 4 FIG. 1 is a block diagram of a device for predicting future wind speed based on historical wind speed data according to an exemplary embodiment. Figure 4 As shown, the device for predicting future wind speed based on historical wind speed data includes an acquisition module 41, a statistical module 42 and a construction module 43.

[0109] The acquisition module 41 is used to obtain wind speed data of wind resources for n years;

[0110] A statistical module 42 is used to collect monthly data characteristics of the wind speed data of the n years according to the wind speed data of the n years;

[0111] The construction module 43 is used to construct the wind resource standard data for the next year according to the monthly data characteristics of the wind speed data for the n years.

[0112] The acquisition module 41, the statistical module 42 and the construction module 43 included in the device block diagram for predicting future wind speed based on historical wind speed data are controlled to execute the method for predicting future wind speed based on historical wind speed data described in any of the above embodiments.

[0113] like Figure 5 As shown, the present invention provides an electronic device 500, the electronic device comprising: a communication interface, a processor 501, and a memory 502;

[0114] Among them, the memory 502 is used to store program instructions, and when the program instructions are executed by the processor 501 that is communicatively connected to the memory 502 through the communication interface, the wind speed data of wind resources for n years is obtained; based on the wind speed data of n years, the monthly data characteristics of the wind speed data of n years are counted; based on the monthly data characteristics of the wind speed data of n years, standard data of wind resources within the next year are constructed.

[0115] The present invention provides a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, wind speed data of wind resources for n years are obtained; monthly data features of the wind speed data for n years are counted based on the wind speed data for n years; and standard wind resource data for the next year are constructed based on the monthly data features of the wind speed data for n years.

[0116] It should be understood that the specific features, operations and details described hereinabove about the method of the present invention may also be similarly applied to the device and system of the present invention, or, vice versa. In addition, each step of the method of the present invention described above may be performed by the corresponding parts or units of the device or system of the present invention.

[0117] It should be understood that each module / unit of the device of the present invention can be implemented in whole or in part by software, hardware, firmware or a combination thereof. Each module / unit can be embedded in the processor of the computer device in the form of hardware or firmware or independent of the processor, or can be stored in the memory of the computer device in the form of software for the processor to call to perform the operation of each module / unit. Each module / unit can be implemented as an independent component or module, or two or more modules / units can be implemented as a single component or module.

[0118] In one embodiment, a computer device is provided, which includes a memory and a processor, and the memory stores computer instructions executable by the processor, and the computer instructions instruct the processor to execute each step of the method of the embodiment of the present invention when executed by the processor. The computer device can be a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities in a broad sense. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected through a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices through a network. The steps of the method of the present invention are executed by the processor.

[0119] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed, and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0120] It will be understood by those skilled in the art that the method steps of the present invention can be completed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program can be stored in a non-temporary computer-readable storage medium, and the steps of the present invention are executed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (PROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0121] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting future wind speed based on historical wind speed data, characterized in that: include: Get wind speed data for n years of wind resources; According to the wind speed data of the n years, monthly data characteristics of the wind speed data of the n years are counted; According to the monthly data characteristics of the wind speed data of the n years, the standard wind resource data for the next year is constructed.

2. The method for predicting future wind speed based on historical wind speed data according to claim 1, characterized in that: The step of constructing wind resource standard data for the next year based on the monthly data characteristics of the wind speed data for the n years includes: Counting monthly data characteristics of wind speed data for the n years; According to the characteristics of all monthly data in the same month, a two-parameter Weibull distribution is fitted; The data distribution of each month in the wind speed data of the n years is matched with the two-parameter Weibull distribution to obtain the most similar monthly data of each month; The most similar monthly data of each month constitutes the standard wind resource data for the next year.

3. The method for predicting future wind speed based on historical wind speed data according to claim 2, characterized in that: The matching of the two-parameter Weibull distribution to obtain the most similar monthly data for each month includes: Using a preset error calculation method to match the two-parameter Weibull distribution, the preset error calculation method includes a mean absolute error method or a mean absolute percentage error method; According to the array composed of scale parameters and deformation parameters, the corresponding mean absolute error is calculated, and the monthly data with the smallest mean absolute error is counted, and the most similar monthly data of each month is formed by traversing twelve months.

4. The method for predicting future wind speed based on historical wind speed data according to claim 3, characterized in that: The method of obtaining the corresponding mean absolute error and calculating the monthly data with the smallest error from the mean value includes: The Euclidean distance of scale parameter, deformation parameter and mean absolute error value was calculated by cluster analysis, and the compactness of cluster analysis was evaluated based on the silhouette method.

5. The method for predicting future wind speed based on historical wind speed data according to claim 4, characterized in that: Also includes: Cluster analysis is performed based on any combination of the scale parameter, the scale parameter and the deformation parameter, and the scale parameter, the deformation parameter and the average wind speed.

6. The method for predicting future wind speed based on historical wind speed data according to claim 1, characterized in that: Also includes: The wind power generation capacity in the next year is calculated based on the standard wind resource data in the next year and the a priori power curve formula.

7. A device for predicting future wind speed based on historical wind speed data, characterized in that: include: The acquisition module is used to obtain the wind speed data of wind resources for n years; A statistical module, used for collecting monthly data characteristics of the wind speed data of the n years according to the wind speed data of the n years; The construction module is used to construct the wind resource standard data for the next year according to the monthly data characteristics of the wind speed data of the n years.

8. The device for predicting future wind speed based on historical wind speed data according to claim 7, characterized in that: The acquisition module, the statistical module and the construction module are controlled to execute the method for predicting future wind speed based on historical wind speed data as described in any one of claims 1 to 6.

9. An electronic device, characterized in that: include: Communication interface, processor, memory; Wherein, the memory is used to store program instructions, and when the program instructions are executed by the processor that is communicatively connected to the memory through the communication interface, the electronic device implements the method for predicting future wind speed based on historical wind speed data as described in any one of claims 1 to 6.

10. A computer-readable storage medium having program instructions stored thereon, characterized in that: When the program instructions are executed by a computer, the computer implements the method for predicting future wind speed based on historical wind speed data as described in any one of claims 1 to 6.

Images