A regional wind resource assessment and prediction method and system based on interpolation and SARIMA, a storage medium and a processor

Abstract

This invention relates to the field of wind resource assessment and prediction technology, and particularly to a regional wind resource assessment and prediction method, system, storage medium, and processor based on interpolation and SARIMA. The method includes acquiring regional wind speed data; interpolating the data to obtain interpolated wind speed data; calculating the annual available hours of wind resources for each spatial region based on the interpolated wind speed data; selecting the region with the greatest wind resource utilization potential as the target region for wind speed prediction; establishing and training a SARIMA wind speed prediction model; and using the trained SARIMA prediction model to predict the wind speed of the selected region with the greatest wind resource utilization potential, obtaining the wind speed prediction data for that region. This invention combines planar interpolation with a SARIMA model. Planar interpolation is used to reasonably fill in spatial data and assess the utilization potential of wind resources, while the SARIMA model is combined to dynamically predict time series data, overcoming the limitations of traditional methods.

Description

Technical Field

[0001] This invention relates to the field of wind resource assessment and prediction technology, and in particular to a regional wind resource assessment and prediction method, system, storage medium and processor based on interpolation and SARIMA. Background Technology

[0002] With the booming development of the renewable energy industry, the proportion of renewable energy in the energy structure has increased significantly. Among them, wind power generation accounts for a growing share of total electricity consumption year by year. Taking 2023 as an example, my country's newly installed wind power capacity reached 75 gigawatts, accounting for approximately 65% ​​of the global newly installed wind power capacity. Of this, offshore wind power accounted for 6.3 gigawatts, representing 71% of the global newly installed offshore wind power. With the rapid development of the wind power industry and the planned construction of numerous wind farms in the future, obtaining large amounts of high-resolution wind speed data to assist in wind farm site selection is crucial for the efficient utilization of wind energy.

[0003] Traditional wind resource assessment methods, such as field measurements and assessments based on numerical weather prediction (NWP) models, often suffer from difficulties in data acquisition, high costs, and insufficient spatial coverage. Planar interpolation methods, by interpolating multiple observation points within a region, can geographically expand the assessment scope of wind resources, making them suitable for data-sparse areas, and exhibiting good spatial adaptability, especially in complex terrain conditions. The SARIMA model, as a time-series forecasting model, can capture the seasonality and trends of wind speed data, making it suitable for dynamic forecasting of historical wind speed data.

[0004] Therefore, it is necessary to study a regional wind resource assessment and prediction method based on interpolation and SARIMA to solve the problems of high data acquisition difficulty, high cost and insufficient spatial coverage in current wind resource assessment. Summary of the Invention

[0005] To address the problems in existing technologies, this invention provides a regional wind resource assessment and prediction method, system, storage medium, and processor based on interpolation and SARIMA. The specific technical solution is as follows:

[0006] A regional wind resource assessment and prediction method based on interpolation and SARIMA includes the following steps:

[0007] Step S1: Obtain regional wind speed data, classify the obtained data according to time series and spatial series, and preprocess the obtained raw wind speed data to remove obviously abnormal data.

[0008] Step S2: Preserve the original temporal resolution, perform interpolation on the classified spatial sequence data to obtain the interpolated spatial sequence, divide the wind speed data into various spatial regions, and arrange the interpolated wind speed data according to the time series to obtain the interpolated wind speed data.

[0009] Step S3: Calculate the annual available hours of wind resources for each spatial region based on the interpolated wind speed data, and select the region with the greatest potential for wind resource utilization as the target region for wind speed prediction.

[0010] Step S4: Establish and train the SARIMA wind speed prediction model;

[0011] Step S5: Use the trained SARIMA prediction model to predict the wind speed in the selected area with the greatest wind resource utilization potential, and obtain the wind speed prediction data for that area.

[0012] Preferably, the interpolation processing of the classified spatial sequence data in step S2 specifically includes interpolating the longitude direction of the spatial sequence data and interpolating the latitude direction of the spatial sequence data.

[0013] Preferably, the longitude direction of the spatial sequence data is interpolated using the following formula:

[0014] f(x)=a(xx i ) 3 +b(xx i ) 2 +c(xx i )+d

[0015] Where x is the longitude and azimuth of the location being sought, x i Let be the longitude and azimuth of the i-th region near the target location, where a, b, c, and d are coefficients, specifically calculated using the following formula:

[0016]

[0017] Where, f(x) i f'(x) represents the wind speed value of the i-th region with respect to longitude and azimuth. i f(x) represents the partial derivative of the wind speed in the i-th region with respect to longitude; i+1 f'(x) represents the wind speed value of the (i+1)th region with respect to longitude and azimuth. i+1 ) represents the partial derivative of the wind speed in the (i+1)th region with respect to the longitude direction.

[0018] Preferably, the interpolation of the latitudinal direction of the spatial sequence data is performed using the following formula:

[0019]

[0020] Where f(y) is the wind speed value at latitude y of the target location, y is the latitude of the target point, y1 is the wind speed value of the region one latitude lower than y, and y2 is the wind speed value of the region one latitude higher than y.

[0021] Preferably, the calculation of the annual available hours of wind resources for each spatial region based on the interpolated data in step S3 is as follows:

[0022]

[0023] Among them, H a This refers to the annual available hours of the wind turbine; W y P represents the region's annual power generation. r This refers to the rated power of a single fan.

[0024] Preferably, step S4 specifically includes the following steps:

[0025] First, select data from the interpolated wind speed data to divide it into a test set and a training set;

[0026] Determine the amount of data in the training set. If the amount of data in the training set is more than 10 times the amount of data in the test set, proceed directly with model training; otherwise, interpolate the time series data to increase the amount of data in the training set.

[0027] The SARIMA wind speed prediction model is established as follows:

[0028]

[0029] θ(B)=1-θ1B-…-θqB q

[0030] Φ(B s )=1-ΦB s -…-Φ p B Ps

[0031] Θ(B s )=1-Θ1B-…-Θ Q B Qs ;

[0032] Among them, Y t This represents the time series of wind speed data for the target area, where d represents the time series of the Y sequence. t The number of ordinary differencing operations is used to eliminate non-seasonal trends; D represents the number of seasonal differencing operations performed on the time series to eliminate seasonal costs; (1-B) d (1-B s ) D Yt The time series is a stationary series after differencing; B represents the lag operator, (1-B) represents the difference operator, and (1-B) represents the stationary time series after differencing. s The lag operator B shifts the time series forward by one time step (s). For seasonal autoregressive models, This represents a p-th order autoregressive polynomial. For non-seasonal autoregressive parameters; Φ(B) s ) represents a p-order seasonal autoregressive polynomial, Φ1,Φ2,…,Φ p For p-order seasonal autoregressive parameters; θ(B)Θ(B) s Let θ(B) represent the seasonal moving average model, where θ(B) represents the q-th order moving average polynomial, θ1, θ2, ..., θ3. q For non-seasonal moving average parameters, Θ(B) s ) represents the seasonal moving average polynomial, Θ1,Θ2,…,Θ3 Q Let ε be the parameter of the Q-order seasonal moving average. t The noise is Gaussian, and the s-parameters represent the seasonal period.

[0033] The wind speed data of the test set is subjected to a stationarity test to determine whether the wind speed data of the test set is a stationary sequence. If not, the wind speed data of the test set is differentially processed.

[0034] After the wind speed data of the test set stabilizes, the SARIMA wind speed prediction model is identified and its parameters are determined based on the autocorrelation coefficient plot (ACF) and partial autocorrelation coefficient plot (PACF).

[0035] Preferably,

[0036] Model identification and parameter order determination based on the autocorrelation coefficient graph (ACF) and partial autocorrelation coefficient graph (PACF) involves first determining the range of values ​​for parameters p and q based on the ACF and PACF, and then selecting and determining the values ​​of model parameters p, d, q, P, D, and Q according to the Akaike information content criterion.

[0037] A regional wind resource assessment and prediction system based on interpolation and SARIMA, applied to the method described, includes:

[0038] The data acquisition and processing module is used to acquire regional wind speed data, classify the acquired data according to time series and spatial series, and preprocess the acquired raw wind speed data to remove obviously abnormal data.

[0039] The data interpolation module is used to preserve the original temporal resolution, perform interpolation processing on the classified spatial sequence data to obtain the interpolated spatial sequence, divide the wind speed data into various spatial regions, and arrange the interpolated wind speed data according to the time series to obtain the interpolated wind speed data.

[0040] The target area selection module is used to calculate the annual available hours of wind resources for each spatial area based on the interpolated wind speed data, and select the area with the greatest potential for wind resource utilization as the target area for wind speed prediction.

[0041] The model building module is used to build and train the SARIMA wind speed prediction model;

[0042] The wind speed prediction module is used to predict the wind speed of the selected area with the greatest wind resource utilization potential using a trained SARIMA prediction model, and obtain the wind speed prediction data for that area.

[0043] A computer-readable storage medium includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to perform the regional wind resource assessment and prediction method based on interpolation and SARIMA.

[0044] A processor for running a program, wherein the program executes the regional wind resource assessment and prediction method based on interpolation and SARIMA.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] This invention combines planar interpolation with the SARIMA (Seasonal Autoregressive Integral Moving Average) model. Planar interpolation is used to appropriately fill in spatial data and assess the utilization potential of wind resources. Combined with the SARIMA model, dynamic prediction of time series data is performed, overcoming the limitations of traditional methods. This method improves the accuracy and timeliness of wind resource assessment without increasing additional measurement costs, and is particularly suitable for wind resource assessment and wind power generation potential prediction in areas with sparse wind speed measurement points or complex terrain. This method can improve the scientific rigor of wind farm site selection and significantly reduce the cost of wind resource assessment. Attached Figure Description

[0047] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0048] Figure 1 This is a flowchart of the method of the present invention.

[0049] Figure 2 This is a wind speed-wind power curve.

[0050] Figure 3 The annual wind speed curve for the example area is shown.

[0051] Figure 4 The annual power curve for the example region is shown.

[0052] Figure 5 The diagram shows the ACF and PACF.

[0053] Figure 6 This is a wind speed data chart for January 1, 2023.

[0054] Figure 7 This is a system schematic diagram of the present invention. Detailed Implementation

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0057] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0058] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0059] Example 1:

[0060] like Figure 1 As shown, this embodiment provides a regional wind resource assessment and prediction method based on interpolation and SARIMA, including the following steps:

[0061] Step S1: Obtain regional wind speed data, classify the obtained data according to time series and spatial series, and preprocess the obtained raw wind speed data to remove obviously abnormal data.

[0062] The obtained data is classified into time series and spatial series to facilitate the extraction of the required data according to the interpolation requirements during subsequent interpolation algorithms.

[0063] This embodiment selects hourly wind speed data from 2012 to 2023 for the South my country Sea region (3° to 23° north latitude, 109° to 121° east longitude), with a spatial resolution of 31km × 31km.

[0064] Step S2: Preserving the original temporal resolution, interpolate the categorized spatial sequence data to obtain the interpolated spatial sequence. Divide the wind speed data into various spatial regions and arrange the interpolated wind speed data according to the time series to obtain the interpolated wind speed data. The interpolation processing of the categorized spatial sequence data specifically includes interpolating the longitude direction and the latitude direction of the spatial sequence data.

[0065] The categorized data is interpolated to preserve the original temporal resolution while improving its spatial resolution, thereby obtaining more spatial wind speed data.

[0066] Two interpolation methods, Bicubic and Cubic, were used for planar interpolation. The method with higher accuracy for this region was selected through MSE comparison, avoiding overfitting or underfitting during interpolation. This embodiment found that the Cubic interpolation method performed better on the selected original data in the South China Sea.

[0067] Cubic interpolation is an extension of one-dimensional cubic polynomial interpolation. In this embodiment, the longitude direction of spatial sequence data is interpolated using the following formula:

[0068] f(x)=a(xx i ) 3 +b(xx i ) 2 +c(xx i )+d

[0069] Where x is the longitude and azimuth of the location being sought, x i Let be the longitude and azimuth of the i-th region near the target location, where a, b, c, and d are coefficients, specifically calculated using the following formula:

[0070]

[0071] Where, f(x) i f'(x) represents the wind speed value of the i-th region with respect to longitude and azimuth. i f(x) represents the partial derivative of the wind speed in the i-th region with respect to longitude; i+1 f'(x) represents the wind speed value of the (i+1)th region with respect to longitude and azimuth. i+1 ) represents the partial derivative of the wind speed in the (i+1)th region with respect to the longitude direction.

[0072] The following formula is used to interpolate the latitudinal direction of spatial sequence data:

[0073]

[0074] Where f(y) is the wind speed value at latitude y of the target location, y is the latitude of the target point, y1 is the wind speed value of the region one latitude lower than y, and y2 is the wind speed value of the region one latitude higher than y. The latitude region is the region in the latitude direction of the previously divided spatial regions.

[0075] This embodiment doubles the spatial resolution of the original wind speed data in the South my country Sea, increasing it to 15km × 15km. The interpolated wind speed data is then arranged according to a time series, including all wind speed data from the South China Sea at each given time point. Because this embodiment doubles the original spatial resolution, the original single spatial region containing wind speed data is increased to four spatial regions containing wind speed data, thus expanding the amount of wind speed data across the spatial range and achieving higher spatial resolution.

[0076] Step S3: Calculate the annual available hours of wind resources for each spatial region based on the interpolated wind speed data, and select the region with the greatest potential for wind resource utilization as the target region for wind speed prediction.

[0077] First, the wind turbine model is selected. The required wind turbine model is chosen, and its power curve is plotted using its relevant parameters and the relationship between wind speed and power. The annual wind speed variation curve for each small area is then combined with the wind turbine's power curve to calculate the annual power generation for that area. Based on the interpolated data, the annual available hours of wind resources for each spatial area are calculated as follows:

[0078]

[0079] Among them, H a This refers to the annual available hours of the wind turbine; W y P represents the region's annual power generation. r This refers to the rated power of a single fan.

[0080] In step S2, spatial interpolation is used to improve the spatial resolution of the selected area and generate more small areas. By calculating the annual available hours of wind resources in the small areas, the wind resource distribution of the entire large area can be evaluated. Areas with an annual available hour of more than 3,000 hours can be identified for further analysis, avoiding wind speed prediction for all areas and significantly reducing the computational load of the subsequent SARIMA algorithm.

[0081] In this embodiment, the wind speed data selected is the interpolated wind speed data for the South my country Sea in 2023. An 8MW Vestes V 164-8.0 wind turbine is selected for calculating the annual available hours in the region. The specific power calculation can be performed according to the following algorithm:

[0082] When v <v i At this time, the wind speed is less than the wind speed at which the wind turbine starts, so the wind turbine does not start and P = 0;

[0083] When v i ≤v≤v r At this time, the output power gradually increases with the increase of wind speed. The relationship between output power and wind speed is shown in the attached figure. Figure 2 As shown;

[0084] When v r <v<v t At this time, the output power remains stable, P = P r ;

[0085] When v≥v t At that time, to protect the fan, the fan stops operating, and P = 0.

[0086] Among them, v i For the cut-in wind speed of the fan, v r v is the rated wind speed of the fan. t P is the cut-out velocity of the fan. r This represents the rated power of the wind turbine. Using the above algorithm, the output power of the wind turbine at each moment can be obtained. Adding up the output power at all moments throughout the year yields the annual cumulative power generation. This example selects the region with the greatest wind resource utilization potential at 18.93°N, 120.52°E for a detailed assessment of wind resources and prediction of future wind speeds. The annual wind speed curve for this region is attached. Figure 3 As shown in the attached figure, the annual power curve is as follows. Figure 4 As shown, the overall wind speed in the selected area with the greatest wind resource utilization potential is in the range of 5-15 m / s, and the wind speed can reach the rated wind speed of the wind turbine for a large part of the time, thus better improving the capacity factor of the wind turbine.

[0087] Step S4: Establish and train the SARIMA wind speed prediction model. This includes the following steps:

[0088] Step S41: First, select data from the interpolated wind speed data to divide it into a test set and a training set; specifically, select data from the interpolated wind speed data that have relatively stable time and spatial intervals to divide it into a test set and a training set.

[0089] Step S42: Determine the amount of data in the training set. If the amount of data in the training set is more than 10 times the amount of data in the test set, directly train the model; otherwise, interpolate the time series data to increase the amount of data in the training set.

[0090] Step S43, establish the SARIMA wind speed prediction model, as follows:

[0091]

[0092] θ(B)=1-θ1B-…-θqB q

[0093] Φ(B s )=1-ΦB s -…-Φ p B Ps

[0094] Θ(B s )=1-Θ1B-…-Θ Q B Qs ;

[0095] Among them, Y t This represents the time series of wind speed data for the target area, where d represents the time series of the Y sequence. t The number of ordinary differencing operations is used to eliminate non-seasonal trends; D represents the number of seasonal differencing operations performed on the time series to eliminate seasonal costs; (1-B) d (1-B s ) D Y t The time series is a stationary series after differencing; B represents the lag operator, (1-B) represents the difference operator, and (1-B) represents the stationary time series after differencing. s The lag operator B shifts the time series forward by one time step (s). For seasonal autoregressive models, This represents a p-th order autoregressive polynomial. For non-seasonal autoregressive parameters; Φ(B) s ) represents a p-order seasonal autoregressive polynomial, Φ1,Φ2,…,Φ p For p-order seasonal autoregressive parameters; θ(B)Θ(B) sLet θ(B) represent the seasonal moving average model, where θ(B) represents the q-th order moving average polynomial, θ1, θ2, ..., θ3. q For non-seasonal moving average parameters, Θ(B) s ) represents the seasonal moving average polynomial, Θ1,Θ2,…,Θ3 Q Let ε be the parameter of the Q-order seasonal moving average. t The noise is Gaussian, and the s-parameter represents the seasonal period.

[0096] Step S44: Perform a stationarity test on the wind speed data of the test set to determine whether the wind speed data of the test set is a stationary series. If not, perform differencing on the wind speed data of the test set. Specifically, use the ADF test to determine whether the input time series is a stationary series. When the p-value < 0.05, the series can be determined to be a stationary series. The formula for performing D-order differencing is shown below:

[0097]

[0098] in This indicates that for the time series Y t The results after performing d non-seasonal differencing. This indicates that for the time series Y t The results after performing d-1 non-seasonal differencing. Y represents t-1 The results after performing d-2 non-seasonal differencing.

[0099] Step S45: After the wind speed data of the test set stabilizes, the SARIMA wind speed prediction model is identified and its parameters are determined based on the autocorrelation coefficient map (ACF) and the partial autocorrelation coefficient map (PACF).

[0100] Furthermore, model identification and parameter order determination are performed based on the autocorrelation coefficient plot (ACF) and partial autocorrelation coefficient plot (PACF). By observing the changing trends of the two curves in the partial autocorrelation plot and the autocorrelation plot, such as tailing and truncation, it is determined whether the input time series data has obvious seasonal trends. The value range of SARIMA model parameters p and q is initially determined, and then the p, d, q, P, D, and Q parameters of the model are screened and determined using the Akaike information content criterion.

[0101] The formula for calculating the Akaike Information Content Criterion is as follows:

[0102] AIC = -2ln(L) + 2k;

[0103] Where L is the maximum likelihood estimate of the input time series, and k is the number of variables in the SARIMA model.

[0104] In this embodiment, interpolated wind speed data from January 1st of each year from 2012 to 2022, representing the area with the greatest wind resource utilization potential in the South my country Sea, is used as the training set. Wind speed data from January 1st, 2023, is used as the test set. Since the training set data volume is much larger than the test set data volume, secondary interpolation is not required. The ACF and PACF are shown in the attached figure. Figure 5 As shown, both the ACF and PACF plots exhibit a certain degree of tailing, indicating a degree of seasonality. The wind speed data predicted by the SARIMA model for January 1, 2023, is attached. Figure 6 As shown, the predicted wind speed data and the actual wind speed data have similar trends, and the daily average error between the predicted and actual wind speeds is less than 15%, demonstrating good accuracy in regional wind speed prediction and making it suitable for wind resource assessment in larger areas. The parameters in this example are p=1, d=0, q=2, P=0, D=0, Q=2, s=24. The AIC under this model is 526.977, indicating that the model strikes a good balance between fitting error and complexity, demonstrating good performance in interpreting data. An AIC of around 500 indicates that the selected data volume is moderate, avoiding excessive complexity and computational difficulty, and no overfitting was observed.

[0105] Step S5: Use the trained SARIMA prediction model to predict the wind speed in the selected area with the greatest wind resource utilization potential, and obtain the wind speed prediction data for that area.

[0106] The more raw wind speed data obtained by the method of this invention, the higher the accuracy of wind speed prediction after interpolation and SARIMA model training. For different geographical regions, it is necessary to select an appropriate interpolation method based on the MSE value. For cases with a small amount of data, the dataset can also be expanded according to the interpolation algorithm to facilitate the training of the SARIMA model.

[0107] This invention combines interpolation and SARIMA model prediction methods. Interpolation handles missing or noisy data, resulting in smoother and more complete wind speed data that more accurately describes wind speed changes. This high-precision data, input into the SARIMA model, allows for more accurate prediction of future wind speeds, improving the reliability of wind resource assessment and effectively supporting the scientific planning and layout of wind farms. The SARIMA model, based on stable and complete time-series data, has a strong ability to capture trend and periodic changes in wind speed. Improved accuracy in predicted power output better supports grid load management, reducing the uncertainty brought by wind power grid connection. Accurate wind speed prediction also helps wind farms make flexible power generation decisions, especially during periods of fluctuating market electricity prices, allowing for increased output during peak wind speed periods, reducing costs and increasing revenue. Combining interpolation and SARIMA wind speed prediction significantly reduces the cost of acquiring high-resolution wind speed data. Predicting future wind speed changes through interpolation and the SARIMA model also helps operation and maintenance teams anticipate wind speed peaks and troughs, optimizing wind farm power generation scheduling and unit operation strategies.

[0108] Example 2:

[0109] like Figure 7 As shown, based on the same inventive concept as Embodiment 1, this embodiment provides a regional wind resource assessment and prediction system based on interpolation and SARIMA, applied to the method described, including:

[0110] The data acquisition and processing module is used to acquire regional wind speed data, classify the acquired data according to time series and spatial series, and preprocess the acquired raw wind speed data to remove obviously abnormal data.

[0111] The data interpolation module is used to preserve the original temporal resolution, perform interpolation processing on the classified spatial sequence data to obtain the interpolated spatial sequence, divide the wind speed data into various spatial regions, and arrange the interpolated wind speed data according to the time series to obtain the interpolated wind speed data.

[0112] The target area selection module is used to calculate the annual available hours of wind resources for each spatial area based on the interpolated wind speed data, and select the area with the greatest potential for wind resource utilization as the target area for wind speed prediction.

[0113] The model building module is used to build and train the SARIMA wind speed prediction model;

[0114] The wind speed prediction module is used to predict the wind speed of the selected area with the greatest wind resource utilization potential using a trained SARIMA prediction model, and obtain the wind speed prediction data for that area.

[0115] Example 3:

[0116] Based on the same inventive concept as Embodiment 1, this embodiment provides a computer-readable storage medium, which includes a stored program, wherein, when the program is executed, it controls the device where the computer-readable storage medium is located to execute the regional wind resource assessment and prediction method based on interpolation and SARIMA.

[0117] Example 4:

[0118] Based on the same inventive concept as in Embodiment 1, this embodiment provides a processor for running a program, wherein the program executes the regional wind resource assessment and prediction method based on interpolation and SARIMA.

[0119] Those skilled in the art will recognize that the modules of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of the examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of the invention.

[0120] In the embodiments provided by this invention, it should be understood that the division of modules is only a logical functional division. In actual implementation, there may be other division methods, such as multiple modules can be combined into one module, one module can be split into multiple modules, or some features can be ignored.

[0121] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0122] If the integrated module is implemented as a software functional module 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 invention, in essence, 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions 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, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A regional wind resource assessment and prediction method based on interpolation method and SARIMA, characterized in that, The method comprises the following steps: Step S1, obtaining regional wind speed data, classifying the obtained data according to time series and spatial series, and pre-processing the obtained original wind speed data to eliminate obviously abnormal data; Step S2, preserving the original time resolution, interpolating the classified spatial series data to obtain interpolated spatial series, dividing the wind speed data into various spatial regions, and arranging the wind speed data after interpolation according to time series to obtain the interpolated wind speed data; Step S3, calculating the annual available hours of wind resources of each spatial region according to the interpolated wind speed data, and selecting a region with the largest wind resource utilization potential as a target region for wind speed prediction; Step S4, establishing and training a SARIMA wind speed prediction model; Step S5, using the trained SARIMA prediction model to predict the wind speed of the region with the largest wind resource utilization potential, and obtaining the wind speed prediction data of the region; The interpolation of the spatial series data in the longitude direction is specifically performed by using the following formula: The interpolation of the spatial series data in the latitude direction is specifically performed by using the following formula: wherein is the longitude of the position sought, is the longitude of the area near the target position, i is the longitude of the area near the target position, , , , is a coefficient, calculated in particular by the following formula: ； wherein, is the wind speed value for the i-th region with respect to the longitudinal direction, i is the partial derivative of the wind speed with respect to the longitudinal direction for the i-th region; is the wind speed value for the i+1-th region with respect to the longitudinal direction, i is the partial derivative of the wind speed with respect to the longitudinal direction for the i+1-th region; is the wind speed value for the i-th region with respect to the longitudinal direction, is the partial derivative of the wind speed with respect to the longitudinal direction for the i-th region; The annual available hours of wind resources of each spatial region are calculated according to the interpolated data in the step S3 as follows: ； in, The desired target location is at y Wind speed values ​​at latitude Let the latitude and azimuth of the target point be defined. For comparison y The wind speed value in the region one latitude lower than the current location. For comparison y The wind speed value of the region at a higher latitude.

2. The regional wind resource assessment and prediction method based on interpolation method and SARIMA according to claim 1, characterized in that, The step S4 specifically comprises the following steps: ； wherein, is the annual available hours of the wind turbine; is the annual energy production of the region; is the rated power of the individual wind turbine. 3.The regional wind resource assessment and prediction method based on interpolation and SARIMA according to claim 1, wherein, First, the data after interpolation is selected to divide into a test set and a training set; The data amount of the training set is determined, and when the data amount of the training set is greater than 10 times the data amount of the test set, the model is directly trained; otherwise, the time series is interpolated to expand the data amount of the training set; The SARIMA wind speed prediction model is established as follows: The wind speed data of the test set is subjected to stationarity test to determine whether the wind speed data of the test set is a stationary sequence, and if not, the wind speed data of the test set is subjected to difference processing; ； ； wherein, a time series of wind speed data representing a target area, denotes the number of ordinary differences to the sequence to eliminate non-seasonal trends; D denotes the number of seasonal differences to the time series, for eliminating seasonal costs; is a stationary time series after differencing; denotes a lag operator, denotes a difference operator, denotes a lag operator moves the time series back by one time step s; is a seasonal autoregressive model, denotes a p-th order autoregressive polynomial, is a non-seasonal autoregressive parameter; denotes a p-th order seasonal autoregressive polynomial, is a p seasonal autoregressive parameter; denotes a seasonal moving average model, wherein denotes q a Q-th order moving average polynomial, is a non-seasonal moving average parameter, denotes a seasonal moving average polynomial, is a Q-th order seasonal moving average parameter, is a Gaussian noise, s the parameter denotes a seasonal period; After the wind speed data of the test set is stationary, the SARIMA wind speed prediction model is identified and parameterized according to the autocorrelation coefficient graph ACF and the partial autocorrelation coefficient graph PACF. The model identification and parameterization according to the autocorrelation coefficient graph ACF and the partial autocorrelation coefficient graph PACF are specifically as follows: first, the value ranges of the parameters p and q are determined according to the autocorrelation coefficient graph ACF and the partial autocorrelation coefficient graph PACF, and then the value of the model parameters p, d, q, P, D and Q is screened and determined according to the Akaike information criterion.

4. The regional wind resource assessment and prediction method based on interpolation method and SARIMA according to claim 3, characterized in that, The method is applied to any one of claims 1 to 4, comprising:

5. A regional wind resource assessment and forecasting system based on interpolation method and SARIMA, characterized in that, A data acquisition and processing module is configured to obtain regional wind speed data, classify the obtained data according to time series and spatial series, and pre-process the obtained original wind speed data to eliminate obviously abnormal data; A data interpolation module is configured to preserve the original time resolution, interpolate the classified spatial series data to obtain interpolated spatial series, divide the wind speed data into various spatial regions, and arrange the wind speed data after interpolation according to time series to obtain the interpolated wind speed data; ​ The target area selection module is configured to calculate annual available hours of wind resources of each spatial area according to the interpolated wind speed data, and select an area with the largest potential of wind resource utilization as a target area for wind speed prediction. The model establishment module is configured to establish and train the SARIMA wind speed prediction model. The wind speed prediction module is configured to use the trained SARIMA prediction model to predict the wind speed of the area with the largest potential of wind resource utilization, and obtain wind speed prediction data of the area.

6. A computer-readable storage medium, characterized in that, The computer readable storage medium comprises a stored program, wherein the program controls the device where the computer readable storage medium is located to execute the interpolation method and SARIMA-based regional wind resource evaluation and prediction method in any one of claims 1 to 4 when the program is running.

7. A processor, comprising: The processor is configured to run a program, wherein the program executes the interpolation method and SARIMA-based regional wind resource evaluation and prediction method in any one of claims 1 to 4 when the program is running.

Images