Wind speed forecast test method, device and equipment

Abstract

The embodiment of the invention provides a wind speed forecast test method, device and equipment. The method comprises the steps that multiple wind speed observation data in a historical time period and wind speed forecast data corresponding to the multiple wind speed observation data are acquired; dividing the wind speed observation data based on the division threshold values of the initial wind speed intervals to obtain wind speed observation data corresponding to different initial wind speed intervals, and performing error calculation on the wind speed observation data in each initial wind speed interval and the corresponding wind speed forecast data to obtain a wind speed forecast result; determining a comprehensive evaluation value corresponding to each initial wind speed interval based on the calculation result; for any initial wind speed interval, if it is determined that the comprehensive evaluation value corresponding to the initial wind speed interval meets the interval adjustment condition, the initial wind speed interval is adjusted based on an adjustment strategy corresponding to the initial wind speed interval; and an inspection report is generated based on the comprehensive evaluation values corresponding to the adjusted wind speed intervals, so that the accuracy and pertinence of the inspection result are improved.

Description

Technical Field

[0001] This application relates to the field of meteorological forecasting technology, and in particular to a wind speed forecasting verification method, apparatus and equipment. Background Technology

[0002] Wind speed forecasting is crucial in various fields such as aviation, shipping, and wind power generation. Accurate wind speed forecasts can significantly improve operational efficiency and safety. Analyzing wind speed forecast data allows for a more detailed understanding of its accuracy, providing a solid basis for improving forecasting models.

[0003] In related technologies, wind speed forecast verification typically employs a full-sample statistical approach, which involves uniformly calculating the error between forecast wind speed data across all times and spaces and their corresponding actual wind speed observation data to obtain an overall forecast performance verification result. However, since high and low wind speeds may exhibit different error distributions in forecasts, and this method ignores the differences in forecast error performance across different wind speed ranges, the verification results are not precise enough and lack specificity. Summary of the Invention

[0004] This invention provides a wind speed forecast verification method, apparatus, and equipment to solve the problems of insufficient accuracy and lack of specificity in the verification results of the prior art.

[0005] In a first aspect, embodiments of this application provide a wind speed forecast verification method, the method comprising: Acquire multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to each of the aforementioned multiple wind speed observation data; Based on the initial wind speed interval division threshold, the above wind speed observation data is divided to obtain wind speed observation data corresponding to different initial wind speed intervals. Error calculations are performed on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval. Based on the calculation results, the comprehensive evaluation value corresponding to each initial wind speed interval is determined. For any initial wind speed range, if it is determined that the comprehensive evaluation value corresponding to the initial wind speed range meets the range adjustment conditions, then the initial wind speed range is adjusted based on the adjustment strategy corresponding to the initial wind speed range. Determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

[0006] The above method, by conducting interval testing of wind speed through a tiered strategy, can reveal the forecast bias of the model under various conditions such as weak wind, moderate wind, strong wind and extreme wind speed, thereby achieving more refined and targeted performance diagnosis and significantly improving the scientific, intelligent and practical nature of wind speed forecast testing.

[0007] Optionally, the aforementioned initial wind speed range includes both a first-type wind speed range and a second-type wind speed range, and the threshold for dividing the aforementioned initial wind speed range is determined based on the following method: The boundary value of the preset wind speed range is determined as the division threshold corresponding to the first type of wind speed range mentioned above. The above wind speed observation data are arranged in order of value size. The value of the wind speed observation data at the preset position in the arranged wind speed observation data is used as the threshold for dividing the second type of wind speed interval.

[0008] The above method, by dividing the initial wind speed range into a first-class wind speed range and a second-class wind speed range, pre-configuring the division threshold for each first-class wind speed range, and dynamically identifying the division threshold for the second-class wind speed range based on historical wind speed observation data, enables the setting of classification standards to be more detailed and accurate.

[0009] Optionally, the above calculation results include the index value of at least one evaluation index; the above calculations of errors for wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval, and the determination of the comprehensive evaluation value corresponding to each initial wind speed interval based on the calculation results, include: For any initial wind speed range, based on the error calculation rules corresponding to each of the above at least one evaluation index, the error calculation is performed on the wind speed observation data and the corresponding wind speed forecast data in the above initial wind speed range to obtain the index value corresponding to the above initial wind speed range. Based on the weight values ​​of at least one of the above evaluation indicators, the indicator values ​​of the above initial wind speed range are weighted to obtain a comprehensive evaluation value corresponding to the above initial wind speed range.

[0010] The above method evaluates the accuracy of wind speed forecasts through multiple evaluation indicators. Based on the error calculation results between the wind speed forecast data and the wind speed observation data after the initial interval division, a comprehensive evaluation value is calculated to provide data support for subsequent interval optimization.

[0011] Optionally, based on the error calculation rules corresponding to each of the above-mentioned at least one evaluation index, the above-mentioned wind speed observation data and corresponding wind speed forecast data in the above-mentioned initial wind speed range are respectively subjected to error calculation to obtain the index value corresponding to the above-mentioned initial wind speed range, including at least one of the following steps: The deviation between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed range is calculated to obtain the first index value; Calculate the root mean square error between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed range to obtain the second index value; The probability that the wind speed forecast data corresponding to each wind speed observation data within the above initial wind speed range falls within the above initial wind speed range is calculated to obtain the third index value.

[0012] The above method, by combining an evaluation mechanism with multiple evaluation indicators such as deviation, root mean square error, and hit rate, can automatically identify wind speed ranges that perform poorly, providing data support for subsequent range optimization.

[0013] Optionally, the above interval adjustment conditions include at least one of the following: The overall score for the current initial wind speed range is higher than the preset threshold; Based on the comprehensive score of the current initial wind speed range and referring to the comprehensive score of the current initial wind speed range in the historical time period, it is determined that the comprehensive score of the current initial wind speed range is on an upward trend; the aforementioned historical time period refers to a preset number of historical time periods before the current historical time period. The numerical relationship between the first average of the comprehensive score values ​​of the current initial wind speed interval and the adjacent initial wind speed intervals and the second average of the comprehensive score values ​​of multiple initial wind speed intervals satisfies the preset relationship.

[0014] The above method improves the dynamic adaptability of the classification test by automatically triggering the adjustment of the wind speed classification interval by introducing a comprehensive score value and automatically reconstructing the wind speed classification interval.

[0015] Optionally, the initial wind speed range is adjusted based on the adjustment strategy corresponding to the initial wind speed range, including: If it is determined that the error distribution between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed interval satisfies the first preset distribution, then the above initial wind speed interval is divided into multiple wind speed intervals; the above first preset distribution is used to characterize that multiple error data are concentrated in multiple regions respectively. If the total number of wind speed observation data and wind speed forecast data within the above initial wind speed interval is determined to be less than a preset number threshold, then the above initial wind speed interval will be merged with the adjacent initial wind speed interval. If it is determined that the error distribution between each wind speed observation data and the corresponding wind speed forecast data within the aforementioned initial wind speed interval satisfies the second preset distribution, then the target sub-interval within the aforementioned initial wind speed interval is merged into the initial wind speed interval adjacent to the aforementioned target sub-interval; the aforementioned second preset distribution is used to characterize that: the error data is abnormal at the boundary of the aforementioned initial wind speed interval, and the aforementioned target sub-interval is: the interval in the corresponding initial wind speed interval where the aforementioned abnormal error data is located.

[0016] The above method, by introducing multiple strategies such as adaptive subdivision, density-driven adjustment, and sample sparsity merging, provides a more reasonable wind speed interval classification based on preliminary results, automatically reconstructs the wind speed classification interval, improves the dynamic adaptability of the classification test, and makes the results more representative.

[0017] Optionally, before determining the comprehensive evaluation value corresponding to each adjusted wind speed interval after adjusting the initial wind speed interval based on the adjustment strategy corresponding to the initial wind speed interval, the method further includes: For each adjusted wind speed interval, the distribution of wind speed observation data in the adjusted wind speed interval is determined to satisfy a third preset distribution; the third preset is used to characterize that the wind speed observation data is concentrated in the central region of the adjusted wind speed interval. If the distribution of wind speed observation data in any adjusted wind speed interval does not satisfy the third preset distribution mentioned above, then the above method further includes: Restore any of the adjusted wind speed ranges to their original state.

[0018] The above method, when implementing the adjustment strategy, uses the distribution of the adjusted wind speed observation data as an adjustment constraint, effectively avoiding situations where the newly divided wind speed intervals completely deviate from the main wind speed distribution center (such as breaking high-density areas into fragments).

[0019] Secondly, embodiments of this application provide a wind speed forecast verification device, the device comprising: The data acquisition module is used to acquire multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to each of the multiple wind speed observation data. The interval division module is used to divide the wind speed observation data based on the initial wind speed interval division threshold, obtain the wind speed observation data corresponding to different initial wind speed intervals, and perform error calculation on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval. Based on the calculation results, the comprehensive evaluation value corresponding to each initial wind speed interval is determined. The interval adjustment module is used to adjust the initial wind speed interval based on the adjustment strategy corresponding to the initial wind speed interval if the comprehensive evaluation value corresponding to the initial wind speed interval is determined to meet the interval adjustment conditions. The report generation module is used to determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

[0020] Thirdly, embodiments of this application provide a wind speed forecast verification device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any step of the wind speed forecast verification method in the first aspect.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer program instructions thereon, which, when executed by a processor, implement any step of the wind speed forecast verification method described in the first aspect above.

[0022] Fifthly, embodiments of this application provide a computer program product, including a computer program stored in a computer-readable storage medium; when a processor of a memory access device reads the computer program from the computer-readable storage medium, the processor executes the computer program, causing the memory access device to perform any step in the wind speed forecast verification method described in the first aspect above.

[0023] For details on each of the second to fifth aspects mentioned above, as well as the technical effects that each aspect may achieve, please refer to the above description of the technical effects that can be achieved for the first aspect or the various possible solutions in the first aspect. These details will not be repeated here. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart of a wind speed forecast verification method provided in this application embodiment; Figure 2 A schematic diagram of a wind speed forecast verification process provided for an embodiment of this application; Figure 3 A schematic diagram of a wind speed forecast verification device provided in an embodiment of this application; Figure 4 This is a schematic diagram of a wind speed forecasting and testing device provided in an embodiment of this application. Detailed Implementation

[0026] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will now be described in further detail with reference to the accompanying drawings.

[0027] The application scenarios described in this application are for the purpose of more clearly illustrating the technical solutions of this application, and do not constitute a limitation on the technical solutions provided in this application. Those skilled in the art will understand that with the emergence of new application scenarios, the technical solutions provided in this application are also applicable to similar technical problems. In the description of this application, unless otherwise stated, "multiple" means two or more.

[0028] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0029] Wind speed forecasting is crucial in various fields such as aviation, shipping, and wind power generation. Accurate wind speed forecasts can significantly improve operational efficiency and safety. In particular, wind speed forecasts at a height of 10 meters, close to the ground, are frequently used to assess the impact on ground facilities, buildings, and people. Verifying wind speed forecast data allows for a more detailed understanding of the forecast's accuracy, providing a valid basis for improving forecasting models.

[0030] In related technologies, wind speed forecast verification typically employs a full-sample statistical approach. This involves uniformly calculating the error between all predicted wind speed data across time and space and their corresponding actual wind speed observation data to obtain the overall forecast performance verification result. Commonly used indicators include bias and root mean square error. This method is simple and intuitive, suitable for quickly evaluating overall forecast performance. However, this method has the following problems: 1. A uniform error test across all samples can mask the model’s performance differences under different wind speed intensities. For example, the model may appear strong in weak winds and weak in strong winds, but the overall error index may still appear good, thus masking the actual risks. For users in industries such as wind disaster prevention and wind power dispatch, it is impossible to provide substantive information on whether the test results for specific wind speed ranges are accurate. 2. Without a hierarchical structure to support it, the test results are difficult to provide feedback to numerical model developers or product creators for targeted corrections; it lacks the ability to optimize in a targeted manner and does not support adjusting weights, parameters, or correcting the model training process separately under different wind speed levels. 3. A few studies use wind speed classification test methods, but they usually adopt manually set static classifications. The classification structure is fixed and lacks flexibility. It is poorly adaptable to different regions (such as coastal or inland), seasons (typhoon season or winter), or special scenarios (wind farms, urban canyon effect). In addition, static classification intervals may also lead to sample size imbalance or the masking of extreme errors. 4. Existing inspection methods mostly provide one-time output results and lack adaptive feedback mechanisms. The inspection results cannot automatically drive hierarchical optimization or be used in the pattern correction process, which limits their role in the business automation system.

[0031] Based on this, this application provides a hierarchical wind speed forecast verification method. By dividing wind speed data into different initial wind speed intervals and calculating the error between the observed data and their corresponding wind speed forecast data, the statistical results for each wind speed interval are determined. Simultaneously, the division of wind speed intervals is adjusted based on the statistical results, and the comprehensive evaluation value for each wind speed interval is recalculated after adjustment, generating a corresponding verification report to comprehensively evaluate the accuracy of the forecast model. By using a hierarchical strategy to perform interval-based verification of wind speeds, the forecast deviations of the model under various conditions such as weak winds, moderate winds, strong winds, and extreme wind speeds can be revealed, thereby achieving more refined and targeted performance diagnosis and significantly improving the scientific rigor, intelligence, and practicality of wind speed forecast verification.

[0032] It should be noted that the wind speed in this application embodiment refers to the horizontal wind speed, that is, the standard layer wind speed most commonly used in meteorological observation and forecasting. This vertical wind speed is often used in various application scenarios such as wind power, urban wind disasters, and aviation meteorology, and is also a regular element in the output of global and regional models.

[0033] The wind speed forecast detection method provided in this application has a wide range of applications and can be used in various scenarios. In addition to being suitable for meteorological forecast verification, it can also be used for quantitative evaluation of wind speed forecast quality in wind power prediction systems, judgment of the reliability of gale level prediction in urban emergency wind disaster response systems, extraction of low wind speed error sensitive sections in agricultural pest and disease transmission models, and scientific research / data fusion fields such as reanalysis data and satellite wind field calibration evaluation.

[0034] Figure 1 A flowchart illustrating a wind speed forecast verification method provided in this application embodiment; as shown Figure 1 As shown in the figure, this application provides a wind speed forecast verification method, including: Step S101: Obtain multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to each of the multiple wind speed observation data; In some embodiments, the aforementioned historical time period can be selected based on demand, and the collection location of the aforementioned wind speed observation data can also be selected based on demand.

[0035] In practice, the wind speed observation data mentioned above refers to the actual wind speed data observed. It can be collected using fixed meteorological stations or other observation sites, or through various methods such as different sensors or devices (e.g., drones, satellite remote sensing). As long as the data source is reliable and can provide effective data for subsequent analysis, it is sufficient.

[0036] The aforementioned wind speed forecast data can first be obtained from wind vector forecast data generated by a weather forecast model, including zonal wind speed. and the meridional wind speed of the wind vector and through Wind speed forecast data is calculated based on wind vector data. .

[0037] In some embodiments, after obtaining wind speed observation data and wind speed forecast data, it is necessary to preprocess the two types of data. Specifically, missing values ​​and outliers (such as abnormally large or negative values) in the wind speed observation data and wind speed forecast data are removed. The criteria for judging outliers can be set based on requirements. All data are processed into a unified format (such as Net CDF) for storage to facilitate subsequent data processing and analysis.

[0038] In some embodiments, after preprocessing the wind speed observation data and wind speed forecast data, the wind speed observation data and wind speed forecast data can be aligned according to timestamps so that they can be compared at the same point in time.

[0039] Step S102: Based on the initial wind speed interval division threshold, the wind speed observation data is divided to obtain the wind speed observation data corresponding to different initial wind speed intervals. Error calculation is performed on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval. Based on the calculation results, the comprehensive evaluation value corresponding to each initial wind speed interval is determined. In one possible implementation, the initial wind speed range includes a first type of wind speed range and a second type of wind speed range. Specifically, the first type of wind speed range is a standard wind speed range, and the second type of wind speed range is an extreme wind speed range. The threshold for defining the initial wind speed range is determined based on the following method: The boundary value of the preset wind speed range is determined as the division threshold corresponding to the first type of wind speed range; The multiple wind speed observation data are arranged in order of their values. The values ​​of the wind speed observation data at the preset positions in the arranged multiple wind speed observation data are used as the threshold for dividing the second type of wind speed interval.

[0040] In specific implementation, the wind speed ranges in this application embodiment include two types. One type is the standard wind speed range, whose grading standard (i.e., the division threshold) can be manually set according to the actual application scenario and regional characteristics. For example, the wind speed can be divided into ranges such as 0-3m / s, 3-5m / s, 5-7m / s, and above 7m / s. The divided wind speed ranges can be adjusted with reference to the statistical distribution of historical data to make the test results of each level more representative.

[0041] Another category is extreme wind speed ranges. Since extreme wind speed weather conditions vary in different regions, manually setting grade ranges is cumbersome and inefficient. In this application, for these extreme wind speed samples, an objective percentile wind speed (i.e., the preset position of multiple wind speed observation data after sorting) is used as the grading standard (division threshold). That is, based on historical wind speed observation data, regional division thresholds are dynamically identified so that the setting of grading standards can be more detailed and accurate. Subsequent error assessment of extreme wind speed ranges also makes the assessment results of extreme wind speed weather more representative.

[0042] This application supports designing different wind speed range division thresholds based on different regions / seasons / application scenarios.

[0043] Specifically, the aforementioned preset position (i.e., percentile) can be set according to requirements, such as the 90th percentile W

[90] , the 50th percentile W

[50] ), etc. Assuming that the 90th percentile of the wind speed calculated based on the wind speed observation data is used as the threshold for dividing the extreme wind speed range, the expression is: ;in, For wind speed observation data, Indicates the observation data Sort in ascending order; Indicates rounding down; Indicates the number of wind speed observation data; square brackets This indicates the value to be taken from the position within the square brackets. A dynamic initial grading structure is generated using quantiles from historical observation samples, replacing the traditional fixed grading template, making it more adaptable to different regional / seasonal / climatic backgrounds.

[0044] In some embodiments, after determining the division threshold corresponding to each initial wind speed interval, the wind speed observation data and wind speed forecast data are processed according to the division threshold to obtain the wind speed observation dataset and wind speed forecast dataset corresponding to each initial wind speed interval. Through the graded processing, the verification of each wind speed interval can be refined, thereby identifying the forecast performance under different wind speed levels.

[0045] Specifically, the wind speed observation data and wind speed forecast data are masked according to the threshold values ​​for each wind speed interval. Wind speeds outside the specified intervals in the wind speed observation data or wind speed forecast data are set to NaN, and the wind speed data after each interval are saved. ; ; in, This represents the wind speed observation data after masking based on the graded n-level wind speed range; This represents the wind speed forecast data after masking based on the graded n-level wind speed range.

[0046] By refining the traditional overall wind speed error evaluation method into independent tests within multiple wind speed levels, it supports high-precision identification of the performance differences of forecasts in different wind speed ranges, and is particularly suitable for evaluating the deviation structure under extreme weather or specific business scenarios.

[0047] In one possible implementation, the calculation result includes the index value of at least one evaluation index; error calculations are performed on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval, and the comprehensive evaluation value corresponding to each initial wind speed interval is determined based on the calculation result, including: For any initial wind speed range, based on the error calculation rules corresponding to at least one evaluation index, the error calculation is performed on the wind speed observation data and the corresponding wind speed forecast data in the initial wind speed range to obtain the index value corresponding to the initial wind speed range. Based on the weight values ​​of at least one evaluation index, the index values ​​of the initial wind speed range are weighted to obtain a comprehensive evaluation value corresponding to the initial wind speed range.

[0048] In some embodiments, the evaluation index corresponding to the above calculation results can be set based on requirements. In this application embodiment, the evaluation index is not limited. For example, although the traditional evaluation indexes such as deviation, root mean square error and hit rate are used in the following embodiments of this application, other indexes (such as percentile error, relative error, etc.) can also be used to replace them to provide analysis from different perspectives.

[0049] In some embodiments, the weight values ​​of each evaluation indicator in this application can be set based on experience or historical data to balance the impact of different indicators on the overall result. For example, if a certain evaluation indicator is more important in the overall evaluation, its corresponding weight value can be set to be relatively larger.

[0050] In one possible implementation, based on the error calculation rules corresponding to at least one evaluation index, error calculations are performed on the wind speed observation data and the corresponding wind speed forecast data in the initial wind speed range to obtain the index value corresponding to the initial wind speed range, including at least one of the following steps: The deviation between the observed wind speed data and the corresponding wind speed forecast data within the initial wind speed range is calculated to obtain the first index value; The root mean square error between the observed wind speed data and the corresponding wind speed forecast data within the initial wind speed range is calculated to obtain the second index value. The probability that the wind speed forecast data corresponding to each wind speed observation data within the initial wind speed range falls within the initial wind speed range is calculated to obtain the third index value.

[0051] In some embodiments, the first indicator is the average deviation, the second indicator is the root mean square error, and the third indicator is the hit rate. By calculating multiple evaluation indicators such as average deviation, root mean square error, and hit rate for each wind speed interval, the accuracy of the forecast can be assessed, providing data support for subsequent model optimization. Specifically, deviation measures the amount of systematic difference between wind speed forecast data and wind speed observation data, revealing the degree of deviation in the forecast data. The closer the deviation is to 0, the closer the forecast data is to the observation data, and the higher the forecast level. Root mean square error assesses the overall magnitude of the error in wind speed forecast data, reflecting the accuracy of the forecast. The smaller the root mean square error, the smaller the error in the wind speed forecast data, and the higher the forecast accuracy. Hit rate assesses the correctness of wind speed forecast data in different wind speed intervals, which is particularly important for forecasting extreme wind speeds (such as strong winds). A higher hit rate indicates a higher forecast accuracy and stronger forecasting capability for that wind speed interval.

[0052] Specifically, errors are calculated for the wind speed observation data and wind speed forecast data for each initial wind speed interval, namely, the deviation, absolute deviation, root mean square error, and hit rate are calculated respectively, as expressed by: deviation: ; Root mean square error: ; True positive: ; False negative: ; Hit rate: ; in, This represents the total number of samples of wind speed observation data within the current initial wind speed range. This indicates the number of data points within the current initial wind speed range that satisfy the subsequent conditions; This indicates the number of samples in which wind speed observation data is located within the current initial wind speed range and the corresponding wind speed forecast data is also located within the current initial wind speed range. This indicates the number of samples where the wind speed observation data is within the current initial wind speed range, but the corresponding wind speed forecast data is not within the current initial wind speed range.

[0053] After obtaining the index values ​​corresponding to the initial wind speed intervals, a comprehensive evaluation value is calculated for each initial wind speed interval based on the corresponding evaluation values, in order to identify wind speed intervals with lower comprehensive evaluation values. The comprehensive evaluation value is calculated using the following formula:

[0054] in, This is a comprehensive evaluation value; and These are the maximum values ​​of the deviation and root mean square error of each wind speed observation data within the current initial wind speed range, used for normalization to ensure that each evaluation index is on the same scale. , , , which is a weighting coefficient used to adjust the relative importance of different evaluation indicators.

[0055] Specifically, through the The absolute value is normalized to ensure it is between 0 and 1; the larger the deviation, the higher the comprehensive evaluation value, indicating a large systematic error in the wind speed forecast data; by normalizing the absolute value of the absolute value, the wind speed forecast data ... Normalization is performed to ensure the value is between 0 and 1, representing the degree of deviation between forecast and actual observation data. By subtracting the hit rate from 1, a larger value indicates a worse forecast, facilitating comprehensive calculation. The aforementioned weights... , , The specific values ​​can be set and adjusted based on experience or historical data to balance the impact of different indicators on the overall result. For example, if In the overall evaluation, it is more important to consider that... Set the value to be larger.

[0056] In some embodiments, this application may also employ a model to utilize machine learning algorithms to learn the relationship between wind speed and observed values ​​from historically acquired wind speed observation data and wind speed forecast data, thereby dividing wind speed intervals. However, this method requires the collection of a large amount of historical data as samples to train the model in order to ensure the accuracy of its interval division.

[0057] Step S103: For any initial wind speed interval, if it is determined that the comprehensive evaluation value corresponding to the initial wind speed interval meets the interval adjustment conditions, then the initial wind speed interval is adjusted based on the adjustment strategy corresponding to the initial wind speed interval. In some embodiments, the interval adjustment conditions in this application can be set based on requirements. Optionally, the interval adjustment conditions may include three types: threshold judgment, trend judgment, and multi-interval average judgment.

[0058] In one possible implementation, the above-mentioned interval adjustment conditions include at least one of the following: Threshold judgment: The overall score of the current initial wind speed range is higher than the preset threshold; Trend Judgment: Based on the comprehensive score of the current initial wind speed range and referring to the comprehensive score of the current initial wind speed range in the historical time period, it is determined that the comprehensive score of the current initial wind speed range is on an upward trend; the aforementioned historical time period refers to a preset number of historical time periods before the current historical time period. Multi-interval averaging judgment: The numerical relationship between the first average of the comprehensive score values ​​of the current initial wind speed interval and the adjacent initial wind speed intervals and the second average of the comprehensive score values ​​of multiple initial wind speed intervals satisfies the preset relationship.

[0059] In practice, the above-mentioned preset threshold can be set according to needs. Optionally, the preset threshold can be set to 0.5 or determined according to historical data. When the comprehensive score of the initial wind speed range is higher than the preset threshold, the system will automatically determine that the initial wind speed range needs to be readjusted. The aforementioned trend judgment is based on the error change trend of the current wind speed range over multiple detection cycles to determine whether to adjust it. In practice, this application can pre-set the detection cycle, execute the above-mentioned wind speed forecast verification process in each detection cycle, and record the comprehensive score value corresponding to each wind speed range each time. When determining the conditions for range adjustment, the comprehensive score value corresponding to the current initial wind speed range is obtained in the previous preset number of detection cycles (i.e., the previous preset number of historical time periods). The change trend of the comprehensive score value in multiple detection cycles determines whether to trigger wind speed range adjustment. Specifically, the preset number can be determined based on needs, and the comprehensive score value can also be one or more evaluation index values. The aforementioned upward trend indicates that the error of the current wind speed range is on an increasing trend. The specific determination principle can be set based on needs. For example, in the preset number of detection cycles, there is a first number (less than the preset number) of detection cycles where the comprehensive score value is on an upward trend, or a second consecutive number (less than the preset number) of detection cycles where the comprehensive score value is on an upward trend, etc.; for example, three consecutive detection cycles with poor performance (such as an increase in RMSE and a decrease in HR), etc.

[0060] In the above multi-interval averaging judgment, the adjacent initial wind speed intervals can be wind speed intervals adjacent to either side, or wind speed intervals adjacent to both sides. The preset relationship can be a preset multiple by which the first average value is greater than the second average value. The specific value of this preset multiple can be set according to requirements. That is, if the combined average value of two adjacent levels is significantly higher than the overall average value, the interval adjustment mechanism is triggered.

[0061] In some embodiments, when the present application determines that the current initial wind speed range meets the range adjustment conditions based on the above method, it adjusts it based on the corresponding adjustment strategy.

[0062] In one possible implementation, the adjustment of the initial wind speed range based on the adjustment strategy corresponding to the initial wind speed range includes: Error density subdivision method: If it is determined that the error distribution between each wind speed observation data and the corresponding wind speed forecast data within the above initial wind speed interval meets the first preset distribution, then the above initial wind speed interval is divided into multiple wind speed intervals; the above first preset distribution is used to characterize that multiple error data are concentrated in multiple regions respectively. Sample density-driven merging method: If the total number of wind speed observation data and wind speed forecast data in the above initial wind speed interval is determined to be lower than a preset number threshold, then the above initial wind speed interval is merged with the adjacent initial wind speed interval. Error gradient driven adjustment method: If it is determined that the error distribution of each wind speed observation data and the corresponding wind speed forecast data in the above initial wind speed interval meets the second preset distribution, then the target sub-interval in the above initial wind speed interval is merged into the initial wind speed interval adjacent to the above target sub-interval; the above second preset distribution is used to characterize: the error data has anomalies at the boundary of the above initial wind speed interval, and the above target sub-interval is: the interval in the corresponding initial wind speed interval where the above-mentioned abnormal error data is located.

[0063] In some embodiments, the error values ​​(i.e., error data) corresponding to each wind speed observation data within the initial wind speed range are defined as follows: Distribution analysis is performed. When the error data in the initial wind speed range is concentrated but the level span is large, it is not possible to effectively distinguish the local performance differences. That is, the error data in the initial wind speed range is concentrated in multiple areas of the initial wind speed range. In this case, the initial wind speed range is divided based on the error density subdivision method. Specifically, the mid-latitude or peak position can be used as the subdivision point to divide the initial wind speed range into two wind speed ranges (for example, 5-7m / s is divided into 5-6m / s and 6-7m / s).

[0064] Optionally, the aforementioned peak position can be determined using the kernel density estimation (KDE) method, that is, the peak value in the probability distribution curve simulated based on kernel density estimation is taken as the aforementioned peak position.

[0065] In some embodiments, the error distribution between the wind speed observation data and the corresponding wind speed forecast data within the initial wind speed range can be determined in various ways, and no limitation is made in this embodiment; for example, probability density analysis is performed on the error data to obtain a probability density curve, and the specific strategy to be executed subsequently is determined based on the number of peaks and the data distribution in the probability density curve.

[0066] It should be noted that if the current initial wind speed range simultaneously meets the above multiple adjustment strategies, one of the adjustment strategies can be randomly selected for execution, or multiple adjustment strategies can be executed separately to determine the comprehensive evaluation value of the adjusted wind speed range corresponding to the multiple adjustment strategies, so as to select the final adjustment method.

[0067] In some embodiments, if the number of valid samples (i.e., the total number of wind speed observation data and wind speed forecast data) within the initial wind speed interval is determined to be lower than a preset number threshold (this threshold can be set based on requirements, for example, it can be set to 30), then the current initial wind speed interval is merged with the adjacent initial wind speed intervals using the sample density-driven merging method. This method is applicable to scenarios where there are insufficient samples in extreme wind speed intervals (e.g., >12m / s) but they are independently divided, resulting in a high degree of randomness in the test results.

[0068] Specifically, since there are two adjacent initial wind speed intervals in the current initial wind speed interval, it is necessary to determine the initial wind speed interval to which the current initial wind speed interval will be merged (hereinafter referred to as the first initial wind speed interval for ease of explanation) when merging. When determining the first initial wind speed interval, it can be determined based on the error of the initial wind speed interval, and priority should be given to merging into initial wind speed intervals with similar error levels (such as Bias interpolation < 0.5 m / s); that is, calculate the average error value (such as Bias) of the samples in the current initial wind speed interval and its two adjacent initial wind speed intervals respectively, and take the initial wind speed interval with the smallest difference between the average error value and the average error value of the current initial wind speed interval as the aforementioned first initial wind speed interval.

[0069] In some embodiments, the error values ​​(i.e. error data) corresponding to each wind speed observation data within the initial wind speed interval are analyzed for distribution. When the error data within the initial wind speed interval is abnormal at the boundary of the initial wind speed interval, i.e., a significant error jump occurs at a certain boundary (such as a sudden increase in the RMSE slope), the initial wind speed interval is adjusted based on the error gradient-driven adjustment method. This method is suitable for scenarios where the wind speed interval boundary is set unreasonably and falls exactly in the performance transition interval.

[0070] Specifically, the root mean square error between the observed wind speed data and the corresponding wind speed forecast data within the initial wind speed range is calculated. And calculate the root mean square error relative to the wind speed observation data respectively. derivative If the derivative of the wind speed observation data at a certain boundary of the initial wind speed interval is significantly different from the derivative of other wind speed observation data (e.g., much larger), this judgment method can be set according to requirements. For example, the derivative at the boundary can be set to a preset multiple of the average derivative value of the entire interval, and the preset multiple can also be set according to requirements. In this case, it can be determined that the error data in the initial wind speed interval is abnormal at the boundary of the initial wind speed interval (hereinafter referred to as the first boundary for ease of explanation). At this time, the first boundary of the initial wind speed interval is moved inward, that is, the interval within the preset range at the first boundary is merged into the initial wind speed interval adjacent to the first boundary; the value of the preset range can be set according to requirements, and this application embodiment does not impose any restrictions.

[0071] By introducing multiple strategies such as score triggering, adaptive subdivision, density-driven adjustment, and sample sparsity merging, the wind speed classification intervals are automatically reconstructed, thereby improving the dynamic adaptability of the classification test.

[0072] In some embodiments, the above-mentioned adjustment strategies of this application can be listed in a list form to facilitate user selection and viewing; and after the above-mentioned adjustment strategies are implemented, the comprehensive evaluation value calculation process is re-executed to form a closed-loop structure of "forecast → verification → scoring → feedback → re-division → re-verification" to achieve continuous optimization of wind speed forecast verification and service quality.

[0073] In some embodiments, this application may determine whether to implement the above adjustment strategy based on the principle that the sum of the comprehensive evaluation values ​​corresponding to multiple adjusted wind speed ranges is the largest.

[0074] Step S104: Determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

[0075] In some embodiments, after obtaining the comprehensive evaluation values ​​corresponding to each adjusted wind speed range, this application analyzes the comprehensive evaluation values ​​and generates corresponding charts and test reports to intuitively display the test results of each wind speed range, making it easier for users to understand and apply.

[0076] Specifically, by comprehensively analyzing the comprehensive evaluation values ​​of each wind speed range and the index values ​​of each evaluation indicator, the results are presented in various visualization methods such as pictures and tables, generating a preliminary inspection report to help users understand the performance of different wind speed levels.

[0077] It should be noted that the process of generating the corresponding charts and inspection reports can be performed after each calculation of the evaluation index values ​​for each wind speed range. That is, this operation can be performed before and after each range adjustment. This application embodiment does not limit its execution time.

[0078] In one possible implementation, after adjusting the initial wind speed range based on the adjustment strategy corresponding to the initial wind speed range, and before determining the comprehensive evaluation value corresponding to each adjusted wind speed range, the method further includes: For each adjusted wind speed interval, the distribution of wind speed observation data in the adjusted wind speed interval is determined to satisfy a third preset distribution; the third preset is used to characterize that the wind speed observation data is concentrated in the central region of the adjusted wind speed interval. If the distribution of wind speed observation data in any adjusted wind speed interval does not satisfy the third preset distribution mentioned above, then the above method further includes: Restore any of the adjusted wind speed ranges to their original state.

[0079] In some embodiments, after executing the adjustment strategy corresponding to the initial wind speed range, this application also needs to determine whether the adjusted initial wind speed range meets the requirements, that is, whether the distribution of wind speed observation data in the adjusted wind speed range has not deviated from the central area of ​​the wind speed range; if it is determined that the third preset distribution is not met, it means that the adjusted wind speed range is not suitable, and at this time it is restored to the unadjusted state, that is, the initial wind speed range.

[0080] Specifically, the distribution of observed data within the adjusted wind speed range can be determined using the wind speed probability density function (PDF). The wind speed probability density function is a mathematical model used to describe the probability of wind speed occurring at different wind speed values. It is of great significance in meteorology, wind energy utilization, architectural design, and environmental protection. Common wind speed probability density functions include the Weibull distribution, Rayleigh distribution, and log-normal distribution; this application does not limit the type of wind speed probability density function used in its embodiments.

[0081] It should be noted that the scope of the central area of ​​the above-mentioned wind speed range is not limited in the embodiments of this application. In specific implementation, it can be set according to the needs, such as setting it to a region 1m / s away from the center of the wind speed range.

[0082] In some embodiments, this application supports a rollback mechanism if, after a wind speed range adjustment, the overall evaluation value of the adjusted wind speed range does not significantly improve (e.g., ...). Automatic rollback can be enabled to restore the original wind speed range setting, ensuring dynamism while maintaining system stability.

[0083] In this embodiment of the application, the above-described wind speed forecast verification method can be applied to the following various application scenarios: Weather forecast model optimization assistance: The poor performance level in the test can be located in the section where the model bias is concentrated, and targeted adjustment suggestions can be made by associating with the physical parameter library (such as boundary layer parameterization scheme, surface flux setting, turbulence closure scheme); Product credibility label output: Each wind speed level outputs the corresponding hit rate and root mean square error, which can be used as a confidence factor to add "credibility level" to products such as wind disaster warning and wind power dispatch to enhance customer perception. Correction model weight feedback: The test results can be used as weighting terms during the training phase. Higher error levels have higher weights in the training of the correction model, which improves the correction model's ability to focus on "weak segments". Multi-source forecast fusion optimization: The wind speed forecasts of multiple models are checked for level, the performance of each wind speed range is ranked, the optimal model is dynamically selected as the output source of that level range, and a "level-driven model hybrid system" is constructed.

[0084] The following combination Figure 2 The above wind speed forecast verification process will be explained in detail with a specific example. The process includes the following steps: Step S201: Obtain wind speed data within a historical time period. The wind speed data includes: multiple wind speed observation data and wind speed forecast data corresponding to the multiple wind speed observation data respectively. Specifically, wind speed observation data from meteorological stations and wind speed forecast data from meteorological forecast sources are collected to ensure the accuracy and completeness of the data.

[0085] Step S202: The collected wind speed data is initially classified according to the set standard (i.e., the classification threshold) to obtain the wind speed data corresponding to each initial wind speed interval. The threshold for dividing the data can be set manually or by extracting a specific percentile from the wind speed observation data.

[0086] Step S203: For each wind speed range, calculate the index value of the evaluation index corresponding to the wind speed range. Evaluation metrics include mean deviation, root mean square error, and hit rate, to assess the accuracy of wind speed forecasts.

[0087] Step S204: Analyze the index values ​​of the evaluation indicators corresponding to each wind speed range, and generate a preliminary test report and visualization charts to show the forecast performance of different wind speed ranges. Specifically, the index values ​​of the wind speed range can be weighted based on the weight values ​​of each evaluation index to obtain a comprehensive evaluation value corresponding to the wind speed range.

[0088] Step S205: For any wind speed range, if it is determined that the comprehensive evaluation value corresponding to the wind speed range meets the range adjustment conditions, then the wind speed range is adjusted based on the adjustment strategy corresponding to the wind speed range, and the process returns to step S205. Specifically, if the comprehensive evaluation value corresponding to the wind speed range does not meet the range adjustment conditions, the process ends.

[0089] Specifically, the effectiveness of the preliminary interval classification is evaluated based on the results of error calculation. If the error is found to be large in some wind speed intervals, or if there is a conflict between the conclusions of different statistics, the data in these intervals are reclassified according to different purposes and strategies to optimize the test results.

[0090] Based on the same disclosed concept, this application also provides a wind speed forecast verification device. Since this device is the same as the device in the method of this application, and the principle of the device in solving the problem is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0091] Figure 3 Please refer to the schematic diagram of a wind speed forecast verification device provided in the embodiments of this application. Figure 3 This application provides a wind speed forecast verification device, which includes: The data acquisition module 301 is used to acquire multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to the multiple wind speed observation data. The interval division module 302 is used to divide the wind speed observation data based on the division threshold of the initial wind speed interval, obtain the wind speed observation data corresponding to different initial wind speed intervals, and perform error calculation on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval, and determine the comprehensive evaluation value corresponding to each initial wind speed interval based on the calculation results. The interval adjustment module 303 is used to adjust the initial wind speed interval based on the adjustment strategy corresponding to the initial wind speed interval if the comprehensive evaluation value corresponding to the initial wind speed interval is determined to meet the interval adjustment conditions. The report generation module 304 is used to determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

[0092] Optionally, the aforementioned initial wind speed range includes both a first-type wind speed range and a second-type wind speed range, and the threshold for dividing the aforementioned initial wind speed range is determined based on the following method: The boundary value of the preset wind speed range is determined as the division threshold corresponding to the first type of wind speed range mentioned above. The above wind speed observation data are arranged in order of value size. The value of the wind speed observation data at the preset position in the arranged wind speed observation data is used as the threshold for dividing the second type of wind speed interval.

[0093] Optionally, the above calculation results include the index value of at least one evaluation indicator; the above interval division module 302 is specifically used for: For any initial wind speed range, based on the error calculation rules corresponding to each of the above at least one evaluation index, the error calculation is performed on the wind speed observation data and the corresponding wind speed forecast data in the above initial wind speed range to obtain the index value corresponding to the above initial wind speed range. Based on the weight values ​​of at least one of the above evaluation indicators, the indicator values ​​of the above initial wind speed range are weighted to obtain a comprehensive evaluation value corresponding to the above initial wind speed range.

[0094] Optionally, the interval division module 302 described above shall specifically perform at least one of the following steps: The deviation between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed range is calculated to obtain the first index value; Calculate the root mean square error between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed range to obtain the second index value; The probability that the wind speed forecast data corresponding to each wind speed observation data within the above initial wind speed range falls within the above initial wind speed range is calculated to obtain the third index value.

[0095] Optionally, the above interval adjustment conditions include at least one of the following: The overall score for the current initial wind speed range is higher than the preset threshold; Based on the comprehensive score of the current initial wind speed range and referring to the comprehensive score of the current initial wind speed range in the historical time period, it is determined that the comprehensive score of the current initial wind speed range is on an upward trend; the aforementioned historical time period refers to a preset number of historical time periods before the current historical time period. The numerical relationship between the first average of the comprehensive score values ​​of the current initial wind speed interval and the adjacent initial wind speed intervals and the second average of the comprehensive score values ​​of multiple initial wind speed intervals satisfies the preset relationship.

[0096] Optionally, the aforementioned interval adjustment module 303 is specifically used for: If it is determined that the error distribution between the observed wind speed data and the corresponding wind speed forecast data within the above initial wind speed interval satisfies the first preset distribution, then the above initial wind speed interval is divided into multiple wind speed intervals; the above first preset distribution is used to characterize that multiple error data are concentrated in multiple regions respectively. If it is determined that the number of wind speed observation data within the above initial wind speed interval is less than a preset number threshold, then the above initial wind speed interval will be merged with the adjacent initial wind speed interval. If it is determined that the error distribution between each wind speed observation data and the corresponding wind speed forecast data within the aforementioned initial wind speed interval satisfies the second preset distribution, then the target sub-interval within the aforementioned initial wind speed interval is merged into the initial wind speed interval adjacent to the aforementioned target sub-interval; the aforementioned second preset distribution is used to characterize that: the error data is abnormal at the boundary of the aforementioned initial wind speed interval, and the aforementioned target sub-interval is: the interval in the corresponding initial wind speed interval where the aforementioned abnormal error data is located.

[0097] Optionally, the aforementioned interval adjustment module 303 is also used for: For each adjusted wind speed interval, the distribution of wind speed observation data in the adjusted wind speed interval is determined to satisfy a third preset distribution; the third preset is used to characterize that the wind speed observation data is concentrated in the central region of the adjusted wind speed interval. If the distribution of wind speed observation data in any adjusted wind speed interval does not meet the third preset distribution mentioned above, then the adjusted wind speed interval will be restored to its state before adjustment.

[0098] Based on the same disclosed concept, this application also provides a wind speed forecast verification device. Since this device is the same as the device in the method of this application, and the principle of the device in solving the problem is similar to that of the method, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0099] Those skilled in the art will understand that various aspects of this application can be implemented as a system, method, or program product. Therefore, various aspects of this application can be specifically implemented in the following forms: a completely hardware implementation, a completely software implementation (including firmware, microcode, etc.), or a combination of hardware and software implementations, collectively referred to herein as a "circuit," "module," or "system."

[0100] In some possible implementations, the device according to this application may include at least one processor and at least one memory. The memory stores program code that, when executed by the processor, causes the processor to perform the steps in the wind speed forecast verification method according to the various exemplary embodiments of this application described above.

[0101] The following reference Figure 4 This application describes a wind speed forecasting and testing device 400 according to this embodiment. Figure 4 The device 400 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0102] like Figure 4 As shown, device 400 is presented in the form of a general-purpose device. The components of device 400 may include, but are not limited to: at least one processor 401, at least one memory 402, and a bus 403 connecting different system components (including memory 402 and processor 401). The memory stores program code, which, when executed by the processor, causes the processor to perform the steps described in the wind speed forecast verification method.

[0103] Bus 403 represents one or more of several bus structures, including a memory bus or memory controller, peripheral bus, processor, or a local bus using any of the various bus structures.

[0104] The memory 402 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 4021 and / or cache memory 4022, and may further include read-only memory (ROM) 4023.

[0105] The memory 402 may also include a program / utility 4025 having a set (at least one) of program modules 4024, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0106] Device 400 can also communicate with one or more external devices 404 (e.g., keyboard, pointing device, etc.), and with one or more devices that enable a user to interact with device 400, and / or with any device that enables device 400 to communicate with one or more other devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 405. Furthermore, device 400 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 406. As shown, network adapter 406 communicates with other modules used with device 400 via bus 403. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with device 400, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0107] This application also provides a computer-readable storage medium storing computer-executable instructions required to execute the processor, including a program required to execute the processor.

[0108] In some possible implementations, various aspects of the wind speed forecast verification method provided in this application can also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to cause the computer device to perform the steps in the wind speed forecast verification method according to the various exemplary embodiments of this application described above.

[0109] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0110] The monitoring program product of the embodiments of this application can be a portable compact disc read-only memory (CD-ROM) and include program code, and can run on a device. However, the program product of this application is not limited to this. In this document, the readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0111] A readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying readable program code. This propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A readable signal medium may also be any readable medium other than a readable storage medium, capable of sending, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0112] The program code contained on the readable medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0113] Program code for performing the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device or server. In cases involving remote devices, the remote device can be connected to the user device via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external device (e.g., via the Internet using an Internet service provider).

[0114] It should be noted that although several units or sub-units of the device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this application, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units.

[0115] Furthermore, although the operations of the method of this application are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0116] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0117] This application is described with reference to flowchart illustrations and block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block and / or segment of the flowchart illustrations and block diagrams, as well as combinations of blocks and segments in the flowchart illustrations and block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart. Figure 1 One or more processes and boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0118] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and boxes Figure 1 The steps of the function specified in one or more boxes.

[0120] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0121] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for verifying wind speed forecasts, characterized in that, The method includes: Acquire multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to each of the multiple wind speed observation data; Based on the initial wind speed interval division threshold, the wind speed observation data is divided to obtain wind speed observation data corresponding to different initial wind speed intervals. Error calculations are performed on the wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval. Based on the calculation results, the comprehensive evaluation value corresponding to each initial wind speed interval is determined. For any initial wind speed range, if it is determined that the comprehensive evaluation value corresponding to the initial wind speed range meets the range adjustment conditions, then the initial wind speed range is adjusted based on the adjustment strategy corresponding to the initial wind speed range. Determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

2. The method according to claim 1, characterized in that, The initial wind speed range includes a first type of wind speed range and a second type of wind speed range, and the threshold for dividing the initial wind speed range is determined based on the following method: The boundary value of the preset wind speed range is determined as the division threshold corresponding to the first type of wind speed range; The multiple wind speed observation data are arranged in order of their values. The values ​​of the wind speed observation data at the preset positions in the arranged multiple wind speed observation data are used as the threshold for dividing the second type of wind speed interval.

3. The method according to claim 1, characterized in that, The calculation results include the index value of at least one evaluation index; the step of calculating the error of wind speed observation data and corresponding wind speed forecast data in each initial wind speed interval, and determining the comprehensive evaluation value corresponding to each initial wind speed interval based on the calculation results, includes: For any initial wind speed range, based on the error calculation rules corresponding to each of the at least one evaluation index, the error calculation is performed on the wind speed observation data and the corresponding wind speed forecast data in the initial wind speed range to obtain the index value corresponding to the initial wind speed range. Based on the weight values ​​of each of the at least one evaluation index, the index values ​​of the initial wind speed range are weighted to obtain a comprehensive evaluation value corresponding to the initial wind speed range.

4. The method according to claim 3, characterized in that, The step of calculating the error of wind speed observation data and corresponding wind speed forecast data in the initial wind speed range based on the error calculation rules corresponding to each of the at least one evaluation index, and obtaining the index value corresponding to the initial wind speed range, includes at least one of the following steps: The deviation between the observed wind speed data and the corresponding wind speed forecast data within the initial wind speed range is calculated to obtain the first index value; The root mean square error between the observed wind speed data and the corresponding wind speed forecast data within the initial wind speed range is calculated to obtain the second index value. The probability that the wind speed forecast data corresponding to each wind speed observation data within the initial wind speed range is located within the initial wind speed range is calculated to obtain the third index value.

5. The method according to claim 1, characterized in that, The interval adjustment conditions include at least one of the following: The overall score for the current initial wind speed range is higher than the preset threshold; Based on the comprehensive score of the current initial wind speed range and by referring to the comprehensive score of the current initial wind speed range in historical time periods, it is determined that the comprehensive score of the current initial wind speed range is on an upward trend. The reference historical time period is: a preset number of historical time periods preceding the current historical time period; The numerical relationship between the first average of the comprehensive score values ​​of the current initial wind speed interval and the adjacent initial wind speed intervals and the second average of the comprehensive score values ​​of multiple initial wind speed intervals satisfies the preset relationship.

6. The method according to claim 1, characterized in that, The adjustment of the initial wind speed range based on the adjustment strategy corresponding to the initial wind speed range includes: If it is determined that the error distribution between each wind speed observation data and the corresponding wind speed forecast data within the initial wind speed interval satisfies a first preset distribution, then the initial wind speed interval is divided into multiple wind speed intervals; the first preset distribution is used to characterize that multiple error data are respectively concentrated in multiple regions; If the total number of wind speed observation data and wind speed forecast data within the initial wind speed interval is determined to be less than a preset number threshold, then the initial wind speed interval is merged with the adjacent initial wind speed interval. If it is determined that the error distribution between each wind speed observation data and the corresponding wind speed forecast data within the initial wind speed interval satisfies the second preset distribution, then the target sub-interval within the initial wind speed interval is merged into the initial wind speed interval adjacent to the target sub-interval; the second preset distribution is used to characterize that: the error data is abnormal at the boundary of the initial wind speed interval, and the target sub-interval is: the interval in the corresponding initial wind speed interval where the abnormal error data is located.

7. The method according to any one of claims 1 to 6, characterized in that, Before determining the comprehensive evaluation value corresponding to each adjusted wind speed interval after adjusting the initial wind speed interval based on the adjustment strategy corresponding to the initial wind speed interval, the method further includes: For each adjusted wind speed interval, the distribution of wind speed observation data in the adjusted wind speed interval is determined to satisfy a third preset distribution; the third preset is used to characterize that the wind speed observation data is concentrated in the central region of the adjusted wind speed interval. If the distribution of wind speed observation data in any adjusted wind speed interval does not satisfy the third preset distribution, then the method further includes: Restore any adjusted wind speed range to its original state before adjustment.

8. A wind speed forecast verification device, characterized in that, The device includes: The data acquisition module is used to acquire multiple wind speed observation data within a historical time period, as well as wind speed forecast data corresponding to the multiple wind speed observation data. The interval division module is used to divide the wind speed observation data based on the division threshold of the initial wind speed interval, to obtain the wind speed observation data corresponding to different initial wind speed intervals, and to calculate the error of the wind speed observation data and the corresponding wind speed forecast data in each initial wind speed interval, and to determine the comprehensive evaluation value corresponding to each initial wind speed interval based on the calculation results. The interval adjustment module is used to adjust the initial wind speed interval based on the adjustment strategy corresponding to the initial wind speed interval if the comprehensive evaluation value corresponding to the initial wind speed interval is determined to meet the interval adjustment conditions. The report generation module is used to determine the comprehensive evaluation value corresponding to each adjusted wind speed range, and generate an inspection report based on the determined comprehensive evaluation value.

9. A wind speed forecast verification device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When executed by a processor, the computer program instructions implement the steps of the method according to any one of claims 1 to 7.

Images