Regional sub-seasonal wind energy resource assessment method based on dynamic climate mode, electronic equipment, storage medium and computer program product

Through quantile regression correction technology, the deviation problem of the dynamic climate model in wind speed forecast is solved, the operation efficiency of wind farms and power grids is improved, and high-precision evaluation and prediction of wind energy resources are achieved.

CN119067263AActive Publication Date: 2024-12-03STATE QIHOU CENT

Patent Information

Application Number
CN202411232018.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-12-03
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

The power climate model has limited wind speed forecasting capabilities on the sub-seasonal scale of wind energy resources, and the forecast accuracy has significantly decreased with time, and there are systemic deviations, which is difficult to meet the operating needs of wind power systems.

Method used

The error correction technology based on quantile regression method is adopted, combined with historical wind speed observation data and pattern return data, and the mode deviation is analyzed and corrected, and a quantile regression correction model is established to improve the accuracy of wind speed forecasting.

Benefits of technology

It significantly reduces the system error of wind speed forecasting, especially in low wind speeds and high wind speeds, improves forecast accuracy, improves the operating efficiency of wind farms and power grids, and meets the needs of high-precision wind energy resource evaluation and prediction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067263A_ABST
    Figure CN119067263A_ABST
Patent Text Reader

Abstract

The invention discloses a regional sub-seasonal wind energy resource assessment method based on a dynamic climate mode, electronic equipment, a storage medium and a computer program product, and aims to improve the accuracy of sub-seasonal assessment or prediction of wind energy resources, especially to reduce the error of the dynamic climate mode in the assessment or prediction of the height and wind speed of a fan hub. According to the method, deviation characteristics output by a mode are analyzed, and error correction is carried out by adopting a quantile regression technology, so that accurate evaluation or prediction of the wind speed within 1-45 days in the future is realized. When the method is implemented, the evaluation or prediction capability of the mode on the wind speed of the target area and the deviation characteristics of the sub-seasonal scale for different levels of wind speeds are firstly analyzed; and secondly, aiming at the deviation characteristic of the mode for different grades of wind speeds, a correction scheme capable of effectively improving the mode deviation is selected, so that the evaluation or prediction capability of regional sub-seasonal scale wind energy resources is improved, and the evaluation or prediction accuracy of future evolution of the regional wind energy resources is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of wind energy resource assessment, and relates to a sub-seasonal time scale (next 1 to 45 days) assessment of wind energy resources. Specifically, it is a regional sub-seasonal wind energy resource percentile regression assessment correction technology based on a dynamic climate model. By analyzing the historical observation data and model return data of wind turbine hub height wind speed in the target assessment area, the quantile regression method is used to correct the model deviation, thereby effectively improving its prediction accuracy and achieving accurate assessment of regional sub-seasonal wind energy resources. Background Art

[0002] Against the backdrop of global climate change and energy transformation, the proposal of the "dual carbon" goal (carbon peak and carbon neutrality) makes the transformation of traditional fossil energy to clean and renewable energy imperative. Compared with traditional fossil energy, the biggest feature of renewable energy, especially wind energy resources, is that its power generation is intermittent and volatile. Especially in the short term, changes in wind speed will directly affect the power generation capacity of wind power. In the future, with the large-scale and high-proportion access to the power grid, it will inevitably bring serious challenges to the safe and stable operation of the power grid. The safe and stable operation of the energy system and changes in energy supply and demand are closely related to climate conditions. Especially under seasonal changes and extreme weather conditions, the stability of power supply depends more on the accurate assessment or prediction of wind energy resources. Therefore, how to make accurate assessments or predictions of the availability of wind energy resources in the short to medium term has become an important means to ensure the energy security strategy.

[0003] Dynamic climate model is the most important technical means for sub-seasonal forecasting business at present. Its principle is to simulate and calculate the atmospheric dynamics and thermodynamic processes to obtain the climate conditions in the future. This type of model has achieved certain results in the evaluation or prediction of atmospheric circulation, temperature, precipitation, etc., and is widely used in climate change research. However, the current dynamic climate model has very limited ability to predict near-surface wind speed. Even the most advanced global climate model in the world has a forecast skill of only about 1 week, and the forecast effect decreases significantly with the increase of forecast time. Forecast results exceeding 10 days are no longer of reference and application value. Therefore, the accuracy of directly predicting near-surface wind speed or average wind turbine hub height wind speed on a sub-seasonal scale (1 to 45 days in the future) based on climate models is difficult to meet the needs of the new energy power industry. With the increase in the number of wind farms and the increase in the proportion of wind power connected to the grid, model evaluation or prediction error problems will bring potential risks and uncertainties to the operation of the power system.

[0004] In summary, in the context of the current global energy structure transformation, facing the growing scale of wind power grid connection and the need for wind energy resource assessment or prediction, the limitations of dynamic climate models in wind energy resource assessment or prediction are becoming increasingly prominent. In particular, the sub-seasonal wind speed assessment or prediction faces a large deviation problem, resulting in greater uncertainty in the dispatch and operation of wind power systems in the power grid. Therefore, it is urgent to develop wind speed forecasting technology that can optimize the dynamic model for the average wind turbine hub height, improve the sub-seasonal assessment or prediction accuracy and service level of regional wind energy resources, and thus meet the urgent need to ensure the safe and stable operation of the energy system. Summary of the invention

[0005] 1. Purpose of the invention Aiming at the limitations of existing dynamic climate models in wind energy resource assessment or prediction, especially the limited ability to predict wind speed at the hub height of wind turbines on a sub-seasonal scale, the significant decrease in prediction accuracy over time, and the systematic deviation of the model for different levels of wind speed, in order to solve at least one of the above and other technical problems in the prior art, the present invention aims to propose a regional sub-seasonal wind energy resource assessment or prediction method based on a dynamic climate model, an electronic device, a storage medium and a computer program product. Although the direct prediction ability of the climate dynamic model for wind speed is low, the model deviation in different regions has certain regularity and distribution characteristics, which provides the possibility for model-based prediction correction, thereby further improving the prediction accuracy of regional wind speed. For the target evaluation or prediction area, the prediction ability of the dynamic model for the average wind speed at the hub height of wind turbines in the target area and the deviation characteristics of the model for different levels of wind speed in the sub-seasonal evaluation or prediction are first analyzed. Secondly, based on the deviation characteristics of the target area model for different levels of wind speed, a correction scheme that can effectively improve the model deviation is selected, thereby improving the assessment or prediction ability of regional sub-seasonal wind energy resources, improving the assessment or prediction accuracy of the future evolution of regional wind energy resources, and enhancing the ability to ensure the safe and stable operation of the power grid.

[0006] (II) Technical solution In order to achieve the purpose of the invention and solve the technical problems, the present invention adopts the following technical solutions: The first invention object of the present invention is to provide a regional sub-seasonal wind energy resource assessment or prediction method based on a dynamic climate model, which is used to improve the accuracy of wind energy resource assessment or prediction, especially to reduce the dynamic climate model error in the wind turbine hub height wind speed forecast in the target assessment or prediction area, and to achieve accurate assessment or prediction of wind energy resources within a sub-seasonal scale (1 to 45 days in the future). The method comprises the following steps when implemented: SS1. Acquisition and compilation of historical wind speed observation data Obtain historical wind speed observation data at the hub height of wind turbines from several representative wind farms in the target evaluation or prediction area within a set past time range, remove outliers and missing data through cleaning, and convert them into daily average wind speed data as the basic data set for subsequent evaluation or prediction; SS2. Acquisition and organization of daily output data from the model Obtain daily output data of the model in the target assessment or prediction area within the set past time range under the high-resolution dynamical climate model, including daily rolling forecast data of the corresponding sub-seasonal scale for the next 1 to 45 days generated by the feedback method within the set past time range, and make it match the historical wind speed observation data at least in space, height and time through spatial interpolation and time warping; SS3. Analysis of the characteristics of forecast bias of dynamical climate models Based on the collated historical wind speed observation data and the corresponding model daily output data, the daily root mean square error (RMSE) was selected as the test and evaluation indicator to evaluate the forecasting ability and deviation level of the dynamic climate model for wind turbine hub height wind speed at different lead times (LT) in the sub-seasonal scale, i.e., within the next 1 to 45 days. On this basis, the distribution characteristics of model deviations under different lead times, different wind farms, different months and / or different wind speed levels were further analyzed. SS4. Construction and training of quantile regression bias correction model Based on the analysis results of the forecast deviation characteristics of the dynamic climate model, the daily output data of the model within the set past time range is used as the independent variable, and the corresponding historical wind speed observation data is used as the dependent variable. The quantile regression method (Quantile Regression, QR) is selected and combined with the machine learning algorithm to establish and train quantile regression deviation correction models for different advance forecast days, different wind farms, different months and / or different wind speed levels. SS5. Error correction and application of model real-time forecast data The original real-time forecast data of the dynamic climate model is obtained, and the corresponding quantile regression deviation correction model is selected according to its advance forecast days, target wind farm station, forecast month and / or forecast wind speed level. The quantile regression correction result is obtained by inputting the original real-time forecast data of the model into the selected model for error correction. The quantile regression correction result is weightedly fused with the original forecast result of the model to obtain the final sub-seasonal scale model-corrected wind speed forecast result of the target evaluation or prediction area, so as to provide decision support for wind energy resource evaluation or prediction and actual operation and scheduling of wind farms.

[0007] The second invention object of the present invention is to provide an electronic device, comprising at least a processor and a memory communicatively connected to the processor, wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the above-mentioned regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model.

[0008] The third invention object of the present invention is to provide a computer program product, including computer instructions, for executing the above-mentioned regional sub-seasonal wind energy resource assessment or prediction method based on dynamic climate model.

[0009] The fourth invention object of the present invention is to provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model is implemented.

[0010] (III) Technical Effect Compared with the prior art, the present invention has the following beneficial and significant technical effects: (1) The present invention fully considers the forecasting ability of the dynamic climate model for the average wind turbine hub height wind speed and the deviation characteristics of different levels of wind speed. By adopting the model-based error correction technology, especially by introducing the quantile regression method, and adopting the scheme of giving different regression models with different forecast durations, the error correction of the model direct output results can be effectively corrected for the forecast deviation of different levels of wind speed. Through the comprehensive analysis of model feedback data and historical observation data, the present invention can significantly reduce the systematic error in the forecast, especially in the problem of low wind speed (such as less than 6m / s) being too large and high wind speed (such as more than 12m / s) being too small.

[0011] (2) Quantile regression technology has strong adaptability and robustness. Its significant advantage is that it avoids the problem of random disturbance term distribution assumptions in traditional regression models and effectively handles heteroskedasticity. Since quantile regression can be performed at different quantile levels, the model can better reduce the impact of outliers on the prediction results and enhance the stability of the forecast model. In addition, using this method, quantile regression models can be established based on different forecast days, different wind farms, and historical data from different months, which makes the present invention more adaptable and can perform accurate forecasts under diverse meteorological conditions.

[0012] (3) The present invention establishes a model-based objective sub-seasonal assessment or prediction model for wind energy resources by correcting the errors of the direct output results of the model. The model combines the simulation capability of the dynamic model with the error correction advantage of quantile regression, which can not only improve the accuracy of wind speed prediction, but also make more accurate predictions of wind speed evolution on a sub-seasonal scale of 1 to 45 days in the future. The application of this technology will significantly improve the operating efficiency of wind farms and regional power grids, and meet the needs of the energy and power industry for high-precision assessment or prediction of wind energy resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is an implementation flow chart of the regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model of the present invention; Figure 2 This is a schematic diagram of the average RMSE (vertical axis) of the selected 9 wind power sites in October 2021 (yellow) and February 2022 (blue) versus the number of days of advance forecast LT (horizontal axis); Figure 3 This is a schematic diagram of the deviation of the model forecast wind speed relative to the site observed wind speed in October 2021 and February 2022 under different advance forecast days LT. The horizontal axis in the figure is the observed wind speed range, and the vertical axis is the size of the model deviation; Figure 4 The figure shows the growth rate of the extended-period average RMSE of nine stations after correction by the QR50 scheme and the control schemes LR and OTS relative to the CFS model forecast in (a) October 2022 and (b) October 2023. In the figure, LR is the linear regression correction scheme, and OTS is the optimal TS score scheme. Figure 5 Schematic diagram of the growth rate of the extended-period average RMSE of nine stations relative to the CFS model forecast after QR50, LR, and OTS corrections in (a) February 2023 and (b) February 2024. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical scheme and advantages of the implementation of the present invention clearer, the technical scheme in the embodiment of the present invention will be described in more detail below in conjunction with the drawings in the embodiment of the present invention. The described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The embodiments described below with reference to the drawings are exemplary and are intended to be used to explain the present invention, and should not be understood as limiting the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0015] In view of the limitations of existing dynamic climate models in wind energy resource assessment or prediction, especially the limited ability to assess or predict wind speed on a sub-seasonal scale, the significant decrease in assessment or prediction accuracy over time, and the existence of systematic deviations in the model for wind speeds of different levels, the present invention proposes a regional sub-seasonal wind energy resource assessment or prediction method, electronic equipment, storage medium and computer program product based on a dynamic climate model.

[0016] Example 1 As a specific example, Figure 1 The figure shows a flow chart of the implementation of the objective sub-seasonal evaluation or prediction method of regional wind energy resources based on the dynamic climate model of the present invention. Figure 1 As shown, the present invention is mainly divided into two stages, namely, preliminary data regularization and analysis stage and modeling and correction stage, according to the actual operation process of the technology. The first stage includes the sorting of measured wind speed data within the available period of the selected typical wind farm station, the extraction and processing of the power model feedback data of the corresponding period, and the analysis of the model error characteristics. After integrating the required data, a suitable correction method is selected according to the distribution characteristics of the model error, and a specific implementation example based on percentile regression is selected. This method can effectively reduce the deviation between the model forecast wind speed and the observed wind speed under different levels of wind speed thresholds. The method specifically includes the following steps when implemented: SS1. Acquisition and compilation of historical wind speed observation data The historical wind speed observation data at the height of the wind turbine hub of several representative wind farms in the target evaluation or prediction area are obtained within the set past time range, and the outliers and missing data are removed through cleaning processing, and the data are converted into daily average wind speed data as the basic data set for subsequent evaluation or prediction.

[0017] In some preferred embodiments, the set past time range can be the past 3-10 years. The historical wind speed observation data can be based on the wind tower of the wind farm station, the measurement data of the wind turbine SCADA system, the meteorological station, the lidar and the satellite remote sensing data, or it can be obtained by the fusion of these multi-source data. The time interval of wind speed observation is minute level, and when the wind speed observation data is cleaned, an adaptive outlier removal algorithm is used to identify and remove outliers. At the same time, the observation days with a daily missing rate exceeding the set threshold are processed as whole-day missing and the missing values ​​are supplemented by polynomial interpolation. According to the needs of sub-seasonal scale forecast evaluation or prediction, the data of each observation day is averaged to obtain the daily average wind speed observation results, and finally a historical wind speed observation data set is generated.

[0018] SS2. Acquisition and organization of daily output data from the model The daily output data of the model in the target assessment or prediction area within the set past time range under the high-resolution dynamic climate model are obtained, including the daily return data within the set past time range and the daily rolling forecast data for the next 1 to 45 days on the corresponding sub-seasonal scale, and through spatial interpolation and time warping, they are matched with the historical wind speed observation data at least in space, height and time.

[0019] In some preferred embodiments, the forecast duration of the selected dynamic climate model must be equal to or greater than 45 days to ensure that the model can provide meteorological information within a sufficient time range and support sub-seasonal wind energy resource assessment or prediction, and the model's daily report data contains a variety of meteorological variable parameters and at least includes latitudinal wind speed and meridional wind speed at an altitude of 10 meters, ground air pressure, potential heights of multiple isobaric layers (such as 925hPa, 850hPa, 700hPa, 500hPa), air temperature, latitudinal wind speed and meridional wind speed, and the model data is high-resolution grid data containing geographic longitude and latitude information to facilitate subsequent spatial interpolation and matching, thereby ensuring that the model output is consistent with the wind farm data in the target assessment or prediction area.

[0020] Further preferably, when the daily output data of the model is sorted: first, the height of each isobaric layer from the ground is calculated by the air temperature and potential height of each isobaric layer in combination with the ground pressure; then, the longitude and latitude wind speeds of each isobaric layer and the longitude and latitude wind speeds at an altitude of 10m are vertically interpolated to the height of the wind turbine hub by a linear interpolation method; then, the grid data output by the model are spatially interpolated to the locations of each selected wind farm station by a bilinear interpolation method; finally, the forecast time of the model output data is sorted, and the rolling forecast data output by the model on a daily basis is obtained by intercepting the forecast results of 1 to 45 days after each reporting date within the set past time range.

[0021] SS3. Analysis of the characteristics of forecast bias of dynamical climate models Based on the collated historical wind speed observation data and the corresponding model daily output data, the root mean square error (RMSE) was selected as the test and evaluation indicator to evaluate the forecasting ability and deviation level of the dynamic climate model for wind speed at wind turbine hub height under different lead times (LT) in the sub-seasonal scale, i.e. within the next 1 to 45 days. On this basis, the distribution characteristics of model deviations under different lead times, different wind farms, different months and / or different wind speed levels were further analyzed.

[0022] In some preferred embodiments, the dynamical climate model is evaluated for different lead-time forecast days at sub-seasonal scales. LT When the wind speed forecast deviation is 0.004, the root mean square error (RMSE) is used to characterize the test evaluation index, which is used to quantify the deviation between the predicted wind speed and the actual observed wind speed. The mathematical calculation formula is: in, LT Forecast days in advance and LT ≤45 days, F i For the i The forecast daily average wind speed for the day, O i is the average wind speed observed on the ith day, 1≤ i ≤ LT .

[0023] In other preferred embodiments, when evaluating the wind speed forecast deviation of the dynamic climate model, multi-dimensional evaluation indicators such as bias and correlation coefficient can be further introduced to comprehensively analyze the spatiotemporal distribution characteristics of the model forecast deviation. At the same time, machine learning algorithms (such as random forests or gradient boosting trees, etc.) can be used to identify the key factors affecting the model forecast deviation, providing a scientific basis for the construction of the subsequent deviation correction model.

[0024] SS4. Construction and training of quantile regression bias correction model Based on the results of the analysis of the forecast deviation characteristics of the dynamic climate model, the daily output data of the model within a set past time range was used as the independent variable, and the corresponding historical wind speed observation data was used as the dependent variable. The quantile regression method (Quantile Regression, QR) was selected and combined with the machine learning algorithm to establish and train quantile regression deviation correction models for different advance forecast days, different wind farms, different months and / or different wind speed levels.

[0025] In some preferred embodiments, when constructing a quantile regression bias correction model, an adaptive quantile selection strategy can be adopted to automatically adjust the quantile value (such as 25%, 50%, 75% quantile) according to the error distribution characteristics of different advance forecast days, different wind farms, different months and / or different wind speed levels. At the same time, deep learning technology (such as long short-term memory network LSTM or attention mechanism Attention Mechanism, etc.) is introduced to capture the long-term dependence and periodic characteristics of wind speed time series, improve the model's ability to fit complex nonlinear relationships, and automatically tune the model hyperparameters through the Bayesian optimization algorithm to obtain the best model structure and training strategy.

[0026] In other preferred embodiments, when training the quantile regression bias correction model, adaptive learning rate and dynamic regularization strategy can be introduced to achieve differentiated training of different quantile regression models, thereby ensuring training convergence and avoiding model overfitting, thereby improving the correction accuracy and applicability of the model under different advance forecast days.

[0027] SS5. Error correction and application of model real-time forecast data The original real-time forecast data of the dynamic climate model is obtained, and the corresponding quantile regression deviation correction model is selected according to its advance forecast days, target wind farm station, forecast month and / or forecast wind speed level. The quantile regression correction result is obtained by inputting the original real-time forecast data of the model into the selected model for error correction. The quantile regression correction result is weightedly fused with the original forecast result of the model to obtain the final sub-seasonal scale model-corrected wind speed forecast result of the target evaluation or prediction area, so as to provide decision support for wind energy resource evaluation or prediction and actual operation and scheduling of wind farms.

[0028] SS6. Verify and optimize model-corrected wind speed assessment or forecast results At least the daily root mean square error (RMSE) is used as an evaluation indicator to evaluate the model-corrected wind speed forecast results obtained in step SS5. The accuracy of the model quantile regression correction results is verified by comparing them before and after correction and / or by comparing them with the results of other model correction schemes. Then, based on the verification results, the key parameters in the quantile regression bias correction model are optimized to further improve the correction accuracy and robustness of the model. The model performance is continuously improved through an iterative verification and optimization process to ensure the reliability and practicality of the model-corrected wind speed assessment or prediction results.

[0029] In some preferred embodiments, when optimizing the key parameters in the quantile regression bias correction model, the accuracy and stability of the model are improved by adjusting the quantile threshold, optimizing the hyperparameters of the machine learning algorithm (such as learning rate, maximum depth, regularization coefficient, etc.) and combining an improved weighted fusion strategy (such as dynamic weighting or a weighting method based on model uncertainty).

[0030] Example 2 On the basis of the above-mentioned embodiment 1, this embodiment 2 takes the wind speed at the hub height of the wind turbines of 9 representative wind farms selected in a certain area as the research object, and uses the historical observation data of the wind speed at the hub height of the wind turbines of the wind farms and the return data of the power model to evaluate the forecasting skills of the power model for the wind speed at the hub height of the wind turbines on a sub-seasonal scale, and focuses on analyzing the model deviation characteristics of wind speeds of different levels; on this basis, through comparative screening, select correction schemes that can effectively improve the deviations of wind speeds of different levels, thereby establishing an objective sub-seasonal assessment or prediction technology of wind energy resources based on model error correction, and the daily root mean square error of wind speed in the next 45 days is significantly lower than that of the direct output of the power model. Specifically: SS1. Acquisition and compilation of historical wind speed observation data Step 1.1 Screening and reading of raw data. The observation data of the wind farm station is obtained based on the observation of the wind tower of the station. The raw data without quality control is stored in the server corresponding to the wind tower in text file format (.txt). The first column is the observation time (format is 20210101001500), and the second column is the observed wind speed (in m / s). The observation time resolution is 15 minutes, and the daily observation results are stored as a file. First, the wind speed data set of the hub height of the wind turbine (taking 70m height as an example) measured in 9 typical wind farms in a certain area in the past three years is screened and read from the server. There are 10,953 text files in total. The file name is named after the observation time and the wind farm station serial number (such as 20220101_01.txt).

[0031] Step 1.2 Cleaning and sorting of wind speed data at the site. First, clean the 15-minute observation data, remove outliers, and treat the observation days with a daily missing rate greater than 30% as missing days (missing value is defined as -999). After that, average the data of each observation day to obtain the daily average wind speed observation results, and merge the daily observation data of each wind tower to obtain a total of 10 text files named with the wind farm station number (such as 01.txt), where the first column is the observation time (format is 20210101), and the second column is the daily average wind speed (unit is m / s).

[0032] SS2. Acquisition and organization of daily output data from the model Step 2.1 Obtain and store daily forecast data of the model. Obtain the return data and real-time forecast data of the high-resolution dynamic model (here, the NCEP_CFSv2 model is used as an example) for the past three years. The data is the forecast data for the next 1 to 45 days on a daily basis. The realtime forecast time period is from 20220101 to 20240228, and the parameters include 10 metre u-velocity, 10 metre v-velocity, U-velocity, V-velocity, Geopotential height, Surface pressure, and Temperature. The model data is grid data, which contains longitude and latitude information and is stored in grib2 format.

[0033] Step 2.2 Spatial interpolation and time regularization of model data. After preprocessing the model results using Python, the model data is first spatially interpolated. Specifically, the height of each isobaric layer from the ground is calculated using the temperature and potential height of each isobaric layer and the ground pressure; then, the wind speed of each isobaric layer and the 10m wind speed are interpolated to the height of the wind tower (70m is taken as an example here) by linear interpolation; finally, the grid data is interpolated to the positions of 9 observation stations using the bilinear interpolation method, and the 70m wind speed forecast results at the 9 stations can be obtained.

[0034] Secondly, the forecast time of the model is sorted, and the results of 1 to 45 days after each reporting start date are intercepted to obtain a 45×9 array (for example, if the reporting start date is January 1, 2023, the forecast results from January 2 to February 15, 2023 are intercepted), and the processed results are stored as a txt file with the naming convention of 20230101.txt.

[0035] SS3. Analysis of the characteristics of forecast bias of dynamical climate models Step 3.1 Evaluation of the model's ability to forecast wind speed during the extended period. Using the processed historical observation data of 9 wind farms and the corresponding feedback data of the power model, evaluate the model's ability to forecast wind speed at the hub height of wind turbines for the next 1 to 45 days. The root mean square error (RMSE) is used to characterize the test evaluation index, and the root mean square error of the next 45 days is averaged. The smaller the RMSE value, the better the forecasting ability of the model. The calculation formula for RMSE is: (1) In formula (1), N is the total number used for calculation, which is set to 45 days here. F i Is the number of days i The predicted daily average wind speed is Oi Is the number of days i The average wind speed on the observed day.

[0036] Through analysis, it is found that when the lead time (LT) does not exceed 7 days, the RMSE shows a significant increasing trend as the lead time LT increases. When the lead time LT is within about one week, the RMSE is lower than 2.0 m / s; when the lead time LT exceeds one week, the error increases to more than 2.0 m / s. Subsequently, as the lead time LT increases, the RMSE basically fluctuates around an average level higher than 2.0 m / s ( Figure 2 ). Overall, the error level of the model forecast of the wind speed during the extended period of the selected station exceeds 2.0 m / s, and there is a relatively large room for optimization of the forecast effect.

[0037] Step 3.2 Analysis of the distribution characteristics of model forecast errors. Taking 2 m / s as a grade, the deviation range of model forecast results at different lead times is classified. It can be found that when the actual wind speed is below 6 m / s, the average error of the model in the five forecast time periods is positive. For wind speeds above 6 m / s, the model shows a negative deviation, and the degree of negative deviation increases significantly with the increase of wind speed and the extension of forecast time ( Figure 3 ). When LT=1d, the average forecast error of the model for wind speeds of 6-8 m / s is -1.2 m / s, and the average forecast error for wind speeds of 12-14 m / s is -4.6 m / s. When LT=30d, the average forecast error of the model for wind speeds of 6-8 m / s is -1.2 m / s, and the average forecast error for wind speeds of 12-14 m / s can reach -8.8 m / s. Therefore, the model's forecast of 70 m wind speed has the error characteristics of "light wind forecast is too large, and strong wind forecast is obviously too small".

[0038] SS4. Build and train a quantile regression bias correction model

[0039] According to the error characteristics, after screening and analyzing multiple schemes, the quantile regression method was finally selected to correct the error. Quantile regression (QR) is to regress the prediction factor X through the conditional quantile of the prediction quantity Y, that is, Q(τ|X=x)=x′β(τ). Similar to the traditional linear regression model, the quantile regression model can be expressed as: (2) Among them, Y is the forecast quantity, ; X is the prediction factor, ; βτ is the parameter vector, ; ετ is the corresponding error vector, ; τ satisfies 0< τ <1, which is a specific quantile. Parameter estimation in quantile regression is done by using the quantile of the given forecast quantity Y τ , and then use the weighted residual minimum to estimate the parameters, specifically: (3) in is the loss function, I ( μ ) is an indicative function, and the indicative function satisfies: (4) This scheme adopts τ =50%, that is, the 50th percentile (median) is used as the correction for quantile regression, and the following formula is used for correction: (5) Among them, Correct QR is the percentile regression correction result, QR τ=50% is the quantile regression result with a percentile of 50, and is the model forecast result. The Quantile Regressor model in sklearn in Python is used to train historical data (the quantile is selected as the 50th percentile, that is, the median). The QR τ=50% Through training, we can obtain quantile regression models for different observation stations with different advance forecast days.

[0040] SS5. Corrected estimates or forecasts based on the corrected quantile regression model.

[0041] Here, we take October and February as examples, and establish quantile regression models for different days of advance forecast in these two months, that is, 45 models are established for each month, a total of 90. The quantile regression correction results account for 50% and the original forecast accounts for 50%, that is, according to the calculation method of formula (5), as the corrected model results.

[0042] SS6. Further validation and evaluation of model returns and real-time forecasts The RMSE of the model correction results was calculated and compared with the results of the other two model correction schemes (such as Figure 4 , 5 The evaluation or prediction results obtained by the model-based quantile regression correction model have lower RMSE at all stations to varying degrees compared with the model direct output results, and the overall improvement effect is more significant than the other two control correction schemes.

[0043] Example 3 Based on the above-mentioned embodiment 1, this embodiment 3 further provides an electronic device, comprising at least a processor and a memory communicatively connected to the processor, wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor so that the processor can execute the regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model shown in the above-mentioned embodiment 1.

[0044] Example 4 On the basis of the above-mentioned embodiment 1, this embodiment 4 further provides a computer program product, including computer instructions for executing the regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model shown in the above-mentioned embodiment 1.

[0045] Example 5 On the basis of the above-mentioned embodiment 1, this embodiment 5 further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, the regional sub-seasonal wind energy resource assessment or prediction method based on the dynamic climate model shown in the above-mentioned embodiment 1 is implemented.

[0046] Through the above embodiments, the purpose of the present invention is fully and effectively achieved. Those skilled in the art can understand that the present invention includes but is not limited to the contents described in the drawings and the above specific embodiments. Although the present invention has been described with respect to the most practical and preferred embodiments currently considered, it should be understood that the present invention is not limited to the disclosed embodiments, and any modification that does not deviate from the functional and structural principles of the present invention will be included in the scope of the claims.

Claims

1. A regional sub-seasonal wind energy resource assessment method based on dynamic climate model, It is characterized in that The method comprises at least the following steps when implemented: SS1. Acquisition and compilation of historical wind speed observation data Obtain historical wind speed observation data at the hub height of wind turbines from several representative wind farms in the target assessment area within a set past time range, and organize them into daily average wind speed data after cleaning; SS2. Acquisition and organization of daily output data from the model Obtain daily output data of the model in the target assessment area under the high-resolution dynamical climate model within the set past time range, including daily rolling forecast data of the corresponding sub-seasonal scale for the next 1 to 45 days generated by the feedback method within the set past time range, and make it match the historical wind speed observation data at least in space, height and time through spatial interpolation and time warping; SS3. Analysis of the characteristics of forecast bias of dynamical climate models Based on the collated historical wind speed observation data and the corresponding daily output data of the model, the daily root mean square error (RMSE) was selected as the test and evaluation indicator to evaluate the deviation level of the dynamic climate model for wind speed at the hub height of wind turbines under different advance forecast days in the sub-seasonal scale. On this basis, the distribution characteristics of model deviation under different advance forecast days, different wind farms, different months and / or different wind speed levels were further analyzed. SS4. Construction and training of quantile regression bias correction model Based on the analysis results of the forecast deviation characteristics of the dynamic climate model, the daily output data of the model is used as the independent variable and the corresponding historical wind speed observation data is used as the dependent variable. The quantile regression method is selected and combined with the machine learning algorithm to establish and train quantile regression deviation correction models for different forecast days, different wind farms, different months and / or different wind speed levels. SS5. Error correction and application of model real-time forecast data The original real-time forecast data of the dynamic climate model is obtained, and the corresponding quantile regression deviation correction model is selected according to its advance forecast days, target wind farm station, forecast month and / or forecast wind speed level. The quantile regression correction result is obtained by inputting the original real-time forecast data of the model into the selected model for error correction. The quantile regression correction result is weightedly fused with the original forecast result of the model to obtain the final sub-seasonal scale model corrected wind speed forecast result of the target assessment area, which provides decision support for wind energy resource assessment and actual operation and scheduling of wind farms.

2. The regional sub-seasonal wind energy resource assessment method based on a dynamic climate model according to claim 1, It is characterized in that In the above step SS1, the past time range is set to the past 3-10 years. The historical wind speed observation data is based on the wind tower of the wind farm station, the measurement data of the wind turbine SCADA system, the meteorological station, the lidar and the satellite remote sensing data or is obtained through multi-source data fusion. The time interval of wind speed observation is in minutes. When the wind speed observation data is cleaned, an adaptive outlier removal algorithm is used to identify and remove outliers. At the same time, the observation days with a daily missing rate exceeding the set threshold are processed as whole-day missing and the missing values ​​are supplemented by the polynomial interpolation method. According to the needs of sub-seasonal scale forecast evaluation, the data of each observation day are averaged to obtain the daily average wind speed observation results, and finally a historical wind speed observation data set is generated.

3. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 1, It is characterized in that In the above step SS2, the forecast duration of the selected dynamic climate model must be equal to or greater than 45 days to ensure that the model can provide meteorological information within a sufficient time range and support sub-seasonal wind energy resource assessment, and the model daily report data contains a variety of meteorological variable parameters and at least includes 10-meter altitude zonal wind speed and meridional wind speed, ground pressure, potential heights of multiple isobaric layers, air temperature, zonal wind speed and meridional wind speed, and the model data is high-resolution grid data containing geographic longitude and latitude information to facilitate subsequent spatial interpolation and matching.

4. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 3, It is characterized in that In the above step SS2, when the daily output data of the model are sorted: first, the height of each isobaric layer from the ground is calculated by the air temperature and potential height of each isobaric layer and combined with the ground pressure; then, the longitude and latitude wind speeds of each isobaric layer and the longitude and latitude wind speeds at a height of 10m are vertically interpolated to the height of the wind turbine hub by the linear interpolation method; then, the grid data output by the model are spatially interpolated to the locations of each selected wind farm station by the bilinear interpolation method; finally, the forecast time of the model output data is sorted, and the rolling forecast data output by the model on a daily basis is obtained by intercepting the forecast results of 1 to 45 days after each reporting date within the set past time range.

5. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 1, It is characterized in that In step SS3 above, the dynamical climate model is evaluated for different lead forecast days at sub-seasonal scales. LT When the wind speed forecast deviation is 0.004, the root mean square error (RMSE) is used to characterize the test evaluation index, which is used to quantify the deviation between the predicted wind speed and the actual observed wind speed. The mathematical calculation formula is: in, LT Forecast days in advance and LT ≤45 days, F i For the i The forecast daily average wind speed for the day, O i is the average wind speed observed on the ith day, 1≤ i ≤ LT .

6. The regional sub-seasonal wind energy resource assessment method based on a dynamic climate model according to claim 1, It is characterized in that In the above step SS3, when evaluating the wind speed forecast deviation of the dynamic climate model, the deviation and / or correlation coefficient evaluation index is introduced to comprehensively analyze the spatiotemporal distribution characteristics of the model forecast deviation, and the machine learning algorithm is used to identify the key factors affecting the model forecast deviation.

7. The regional sub-seasonal wind energy resource assessment method based on a dynamic climate model according to claim 1, It is characterized in that In the above step SS4, the constructed quantile regression bias correction model adopts τ =50% median for error correction, and regress the forecast factor X through the conditional quantile of the forecast quantity Y. The mathematical expression is: Among them, Y is the forecast quantity, ; X is the prediction factor, ; β τ is the parameter vector, ; ε τ is the corresponding error vector, ; τ is the quantile and satisfies 0< τ <1, by the quantile of the given forecast quantity Y τ , and estimate the parameters of the quantile regression by minimizing the weighted residuals : in, is the loss function and through the indicative function I ( μ ) is determined, and the indicative function I ( μ )satisfy ; In the practical application of the quantile regression bias correction model, the following formula is used for error correction: Among them, Correct QR is the percentile regression correction result, QR τ=50% for τ =50% quantile regression result, Ori The model forecast results.

8. The regional sub-seasonal wind energy resource assessment method based on a dynamic climate model according to claim 1, It is characterized in that In the above step SS4, when constructing the quantile regression bias correction model, an adaptive quantile selection strategy is adopted to automatically adjust the quantile value according to the error distribution characteristics of different advance forecast days, different wind farms, different months and / or different wind speed levels. At the same time, deep learning technology is introduced to capture the long-term dependence and periodicity characteristics of the wind speed time series, and the model hyperparameters are automatically tuned through the Bayesian optimization algorithm.

9. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 1, It is characterized in that In the above step SS4, when training the quantile regression bias correction model, an adaptive learning rate and a dynamic regularization strategy are introduced to achieve differentiated training of different quantile regression models, thereby ensuring training convergence and avoiding model overfitting.

10. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 1, It is characterized in that The regional sub-seasonal wind energy resource assessment method further includes step SS6 of verifying the model-corrected wind speed assessment result and optimizing the model, wherein: The model-corrected wind speed forecast results obtained in step SS5 are evaluated using at least the daily root mean square error (RMSE) as an evaluation indicator. The accuracy of the model quantile regression correction results is verified by comparing the results before and after correction and / or by comparing with the results of other model correction schemes. Then, the key parameters in the quantile regression bias correction model are optimized based on the verification results, and the model performance is continuously improved through an iterative verification and optimization process.

11. The regional sub-seasonal wind energy resource assessment method based on dynamic climate model according to claim 10, It is characterized in that In the above step SS6, when optimizing the key parameters in the quantile regression bias correction model, the accuracy and stability of the model are improved by adjusting the quantile threshold, optimizing the hyperparameters of the machine learning algorithm and combining the improved weighted fusion strategy.

12. An electronic device comprising at least a processor and a memory connected to the processor in communication, in, The memory stores instructions that can be executed by the processor, and the instructions are executed by the processor so that the processor can execute the regional sub-seasonal wind energy resource assessment method based on the dynamic climate model as described in any one of claims 1 to 11 above.

13. A computer program product, comprising computer instructions for executing the regional sub-seasonal wind energy resource assessment method based on a dynamic climate model as described in any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the regional sub-seasonal wind energy resource assessment method based on a dynamic climate model as described in any one of claims 1 to 11.

Citation Information

Patent Citations

  • Forecast wind speed correction method and system based on multi-scale features

    CN112132365A

  • Refined wind speed prediction method for power transmission line

    CN114021858A

  • Complex terrain-oriented wind energy resource assessment method and device and storage medium

    CN116306026A

  • Wind speed correction model training method, wind speed prediction method, equipment and medium

    CN116595394A

  • Medium and long term weather prediction method and system based on variable mesh model, and medium

    CN117076738A

Cited By

  • Method for optimizing climate mode simulation composite extreme wind and rain event frequency based on deep learning

    CN119442923A

  • Optimization method for simulating the frequency of compound extreme wind and rain events in climate models based on deep learning

    CN119442923B

  • Flood forecasting precision improving method based on multi-model data fusion driving

    CN120069246A

  • Regional wind speed meteorological reference station representativeness evaluation method based on correlation function field, medium and program product

    CN121303674A

  • A method, medium, and program product for representative evaluation of regional wind speed weather reference stations based on a correlation function field

    CN121303674B