Method and device for predicting extended period of solar shortwave radiation

Abstract

The application discloses a method and device for predicting the extension period of solar shortwave radiation, and belongs to the technical field of solar radiation prediction. The method comprises the following steps: statistically analyzing the ground-air temperature difference in the Qinghai-Tibet region, and then classifying the similar years by standard deviation to form a similar year list; further statistically analyzing the solar shortwave radiation data of a target region to be predicted, and then identifying the dominant spatial distribution category of each month of each year to form a similar year type library; when predicting the extension period of the target region, the similar year is determined by using the category of the target region in the current month and the similar year case library, so that the dominant spatial distribution type of the extension period month can be determined by using the determined similar year and the similar year type library, and then the solar shortwave radiation prediction result of the extension period is obtained through the synthesis logic. The application can realize the extension period prediction of the solar shortwave radiation.

Description

Technical Field

[0001] This invention relates to the field of solar radiation prediction technology, and in particular to a method and apparatus for predicting the extension period of solar shortwave radiation. Background Technology

[0002] Solar shortwave radiation is a direct reflection of the amount of solar energy received by the Earth's surface, and its prediction accuracy is crucial for photovoltaic power generation forecasting, climate change research, and energy policy formulation. On extended timescales (10-30 days), accurate solar radiation predictions are of critical indicative significance for energy dispatch and grid security.

[0003] Currently, solar radiation prediction methods mainly focus on short-term (0-72 hours) and ultra-short-term (0-6 hours) forecasts. Related technologies are primarily based on medium- and short-term forecasting methods using multi-source data fusion and complex artificial intelligence models. The core technology of these methods lies in using various data sources and complex models to capture the statistical relationship between meteorological elements and solar radiation. While they have shown some effectiveness in short-term nowcasting, the models are highly dependent on input data (real-time satellite cloud images, ground observation data, etc.) and have high computational complexity, making them difficult to directly apply to extended-term forecasting operations with long data update cycles and greater uncertainty.

[0004] Therefore, there is an urgent need to provide a method for predicting the extension of solar shortwave radiation. Summary of the Invention

[0005] This invention provides a method and apparatus for predicting the extension period of solar shortwave radiation. The technical solution is as follows: On the one hand, a method for predicting the extension of solar shortwave radiation is provided, the method comprising: Obtain a dataset of the Qinghai-Tibet region over the past few years; the dataset includes monthly historical geothermal temperature difference sequences. Based on the standard deviation of the historical geothermal temperature difference series for each month, each month is divided into corresponding categories within multiple categories; different categories correspond to different standard deviations. A similar year case library is formed based on the classification results; the similar year case library includes a list of similar years for each category in each month; Obtain monthly historical data of solar shortwave radiation in the target area over the past few years to identify the dominant spatial distribution type of the target area on a monthly basis. Based on the dominant spatial distribution type of the target region each month, a similar year type library is formed; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; Based on the temperature difference between the ground and air in the Qinghai-Tibet region in the current month, the target category of the current month is determined, and based on the similar year case library, a list of similar years corresponding to the target category in the current month is determined; Based on the similar year type library, determine the dominant spatial distribution type of each year in the target similar year list for the next month; Based on the dominant spatial distribution type of each year in the target similar year list for the next month, the solar shortwave radiation prediction results for the next month are generated according to the synthesis logic.

[0006] On the other hand, a device for predicting the extension of solar shortwave radiation is provided, the device comprising: The acquisition unit is used to acquire a dataset of the Qinghai-Tibet region over the past few years; the dataset includes a monthly historical geothermal temperature difference sequence. The partitioning unit is used to divide each month into corresponding categories among multiple categories based on the standard deviation of the historical geothermal temperature difference series; different categories correspond to different standard deviation sizes; The first forming unit is used to form a similar year case library based on the division results; the similar year case library includes a list of similar years for each category in each month; The identification unit is used to acquire monthly solar shortwave radiation history data of the target area over the past few years in order to identify the dominant spatial distribution type of the target area on a monthly basis. The second forming unit is used to form a similar year type library based on the dominant spatial distribution type of the target area on a monthly basis; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; The first determining unit is used to determine the target category of the current month based on the temperature difference between the ground and air in the Qinghai-Tibet region in the current month, and to determine the target similar year list corresponding to the target category in the current month based on the similar year case library; The second determining unit is used to determine the dominant spatial distribution type of each year in the target similar year list for the next month based on the similar year type library; The prediction unit is used to generate the solar shortwave radiation prediction results for the next month based on the dominant spatial distribution type corresponding to each year in the target similar year list in the next month, according to the synthesis logic.

[0007] On the other hand, a computer device is provided, the computer device including a memory and a processor, the memory for storing computer programs, and the processor for executing the computer programs stored in the memory to implement the steps of the above-described method for predicting the extension of solar shortwave radiation.

[0008] On the other hand, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the steps of the above-described method for predicting the extension of solar shortwave radiation.

[0009] On the other hand, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described method for predicting the extension of solar shortwave radiation.

[0010] The technical solution provided by this invention can bring at least the following beneficial effects: This invention employs a climate statistics-based similarity forecasting approach. Because the temperature difference between the Tibetan Plateau and the Earth's surface is particularly sensitive to global warming, the temperature difference is statistically analyzed, and then categorized using standard deviation to create a list of similar years. Further, solar shortwave radiation data for the target region is statistically analyzed to identify the dominant spatial distribution categories for each year and month, forming a similar year type library. When forecasting the extended period for the target region, the category of the target region in the current month and the similar year case library are used to determine similar years. These similar years and the similar year type library are then used to determine the dominant spatial distribution type for the extended period months, and finally, the extended period solar shortwave radiation forecast results are obtained through synthesis logic. Therefore, this scheme, by combining the temperature difference between the Tibetan Plateau and the Earth's surface with the solar shortwave radiation in China, can achieve extended period forecasting of solar shortwave radiation. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a method for predicting the extension of solar shortwave radiation according to an embodiment of the present invention; Figure 2 This is a structural diagram of a solar shortwave radiation extension prediction device provided in an embodiment of the present invention; Figure 3 This is a hardware architecture diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0014] Existing methods primarily focus on short- and ultra-short-term forecasting windows, which imposes time constraints on decision-making and deployment. Extended-range forecasting, on the other hand, allows for better advance deployment and decision-making. This timeliness limitation stems from the fact that these methods are technically "numerical extrapolations based on the current initial state." Although they continuously integrate various real-time observational data, the forecasting logic heavily relies on the statistical patterns of data changes, thus failing to meet the requirements for longer forecast periods. In meteorology, as the forecast time extends, small errors in the initial field are rapidly amplified by nonlinear dynamic processes, rendering the forecast results completely worthless on extended-range timescales. Therefore, the theoretical foundation of techniques relying on precise initial fields inherently does not support accurate forecasting over long periods.

[0015] Furthermore, it has stringent requirements for input data quality and real-time performance, resulting in insufficient robustness. Model performance is highly dependent on the completeness, accuracy, and high-frequency updates of input data (such as real-time satellite cloud images and ground observation station data). In cases of missing data, delayed updates, or poor data quality, model performance will significantly degrade. In real-world applications, especially for medium- to long-term forecasts, it is difficult to guarantee a continuous, stable, and uninterrupted flow of high-quality data. If any problem occurs in the data input chain (such as satellite malfunctions, data transmission delays, or site maintenance), the entire forecasting system will malfunction or produce unreliable results, limiting the system's robustness and applicability.

[0016] Poor adaptability to extreme and abrupt climate scenarios is also a problem with existing mainstream methods. While existing artificial intelligence models are powerful, their core relies on learning statistical patterns from vast amounts of historical data to establish a mapping relationship between input and output. In the context of climate change, the frequent occurrence of extreme weather and climate events means that weather and climate change patterns are experiencing abnormal fluctuations. When models encounter features that exceed the statistical range of their training data, their lack of a true "understanding" of the underlying physical mechanisms affects their accuracy, ultimately leading to prediction failures.

[0017] By analyzing existing methods, this invention departs from the mainstream approach of numerical extrapolation based on initial states and proposes a similarity forecasting approach based on climate statistics. Considering that the temperature difference between the Tibetan Plateau and the land surface is particularly sensitive to the background of global warming, this invention uses this characteristic as a starting point to combine the temperature difference between the Tibetan Plateau and the land surface with the solar shortwave radiation in China to predict the extended period of solar shortwave radiation.

[0018] The specific implementation of the above concept is described below.

[0019] Please refer to Figure 1 This invention provides a method for predicting the extension period of solar shortwave radiation, the method comprising: Step 100: Obtain a dataset of the Qinghai-Tibet region over the past few years; the dataset includes monthly historical geothermal temperature difference sequences. Step 102: Based on the standard deviation of the historical geothermal temperature difference series for each month, each month is divided into corresponding categories among multiple categories; different categories correspond to different standard deviations. Step 104: Form a similar year case library based on the segmentation results; the similar year case library includes a list of similar years for each category in each month; Step 106: Obtain the monthly solar shortwave radiation history data of the target area over the past few years to identify the dominant spatial distribution type of the target area on a monthly basis. Step 108: Based on the dominant spatial distribution type of the target area each month, form a similar year type library; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; Step 110: Determine the target category of the current month based on the ground temperature difference of the target area in the current month, and determine the target similar year list corresponding to the target category in the current month based on the similar year case library; Step 112: Based on the similar year type library, determine the dominant spatial distribution type of each year in the target similar year list for the next month; Step 114: Based on the dominant spatial distribution type of each year in the target similar year list for the next month, generate the solar shortwave radiation prediction results for the next month according to the synthesis logic.

[0020] In this embodiment of the invention, a similarity forecasting approach based on climate statistics is adopted. Since the temperature difference between the Tibetan Plateau and the surrounding landmass is particularly sensitive to global warming, the temperature difference between the two regions is statistically analyzed, and then categorized by standard deviation to form a list of similar years. Furthermore, solar shortwave radiation data for the target area to be predicted is statistically analyzed to identify the dominant spatial distribution type of the target area each month, forming a similar year type library. When predicting the extended period for the target area, the category of the target area in the current month and the similar year case library are used to determine similar years. The dominant spatial distribution type of the extended period months can then be determined using the identified similar years and the similar year type library, and the extended period solar shortwave radiation prediction result is obtained through synthesis logic. Therefore, this scheme, by combining the temperature difference between the Tibetan Plateau and the surrounding landmass with the solar shortwave radiation in China, can achieve extended period prediction of solar shortwave radiation.

[0021] The following description Figure 1 The execution method for each step is shown.

[0022] First, regarding steps 100 to 104.

[0023] In this embodiment of the invention, the Qinghai-Tibet region includes Qinghai and Tibet, and these two regions can be merged into one area to obtain the monthly historical geothermal temperature difference sequence.

[0024] In one embodiment of the present invention, the method for obtaining the monthly historical geothermal temperature difference sequence is as follows: Obtain daily historical temperature and geothermal data for the Qinghai-Tibet region over the past few years; Based on historical air temperature data and historical ground temperature data, the historical ground temperature difference sequence for each month in the Qinghai-Tibet region over the past few years was calculated. The historical geothermal temperature difference of each month in the Qinghai-Tibet Plateau over the past few years was averaged using the grid area weighted average method, resulting in a series of historical geothermal temperature differences for each month after the regional average.

[0025] Among them, historical temperature data is obtained by measuring the temperature at a meteorological station at a specified distance from the ground, such as 1.5m, 2m, etc.

[0026] In one implementation, the number of years in the "recent years" should be at least 10. This is necessary to provide more data for the subsequent formation of a database of similar year cases. Taking the period from 1981 to 2024 as an example, daily historical temperature data and historical ground temperature data are obtained, and then the monthly historical ground temperature difference sequence of the Qinghai-Tibet Plateau is calculated.

[0027] To achieve extended forecasting, in this embodiment of the invention, the last few days of each calendar month can be included in the forecast for the next calendar month. This approach allows time for data updates, calculations, and publication for the forecast of the next calendar month. In one implementation, these few days can be the last 2-6 days of the calendar month, preferably the last 5 days. For example, the monthly statistical period is from the 25th of the previous calendar month to the 24th of the current calendar month.

[0028] It is understandable that there are multiple meteorological stations in the Qinghai-Tibet Plateau region, each corresponding to a monthly historical geothermal temperature difference sequence. To calculate the regional average monthly historical geothermal temperature difference sequence, a grid area-weighted average method can be used to average the monthly historical geothermal temperature differences in the Qinghai-Tibet Plateau region over recent years. This grid area-weighted average method can be the internationally recognized Jones grid area-weighted average method. In this way, a regional average monthly geothermal temperature difference sequence from 1981 to 2024 can be obtained. Standardizing this sequence yields a normalized monthly geothermal temperature difference sequence.

[0029] In this embodiment of the invention, the standard deviation of the historical land and sea temperature difference series over each month can be used to classify categories, with different categories corresponding to different standard deviation values. In one implementation, the categories can be divided into five types: significantly higher, higher, normal, lower, and significantly lower.

[0030] Specifically, the historical geothermal temperature difference series for each month is traversed month by month, and the standard deviation for that month is calculated. If the standard deviation σ of the current month is greater than or equal to 1, then the current month will be classified into the category of significantly higher. If the standard deviation of the month is 1 > σ ≥ 0.5, then the month will be classified into the higher category. If the standard deviation for the month is 0.5 > σ > -0.5, then the month will be classified into the normal category. If the standard deviation of the month is -0.5 ≥ σ > -1, then the month will be classified into the lower category. If the standard deviation of the month is -1 ≥ σ, then the month will be classified into the category of significantly low.

[0031] It is understandable that the normal category represents the average climate state or slight fluctuations, which belongs to the background field; the above or below category represents a clear climate anomaly signal, which is usually associated with a specific atmospheric circulation pattern; the significantly above or significantly below category represents strong extreme climate anomaly events, which correspond to very typical and powerful climate driving factors.

[0032] By categorizing all months over several years, subsequent predictions do not seek a single most similar year, but rather a similar year plan. By searching by category, multiple historical similar cases can be found within the same category. Through set analysis, more stable and probabilistic prediction results can be provided.

[0033] In this embodiment of the invention, a similar year case library can be formed based on the division results. This similar year case library includes a list of similar years for each category in each month.

[0034] When forming a similar year case library based on the partitioning results, the specific details may include: For each category, perform the following: For each month, determine if the month exists in the current category; if it does, count the years that the month is in the current category, and form a list of similar years for the current category in that month.

[0035] For example, this database of similar year cases can be shown in Table 1 below: Table 1: In one embodiment of the present invention, the similar year case library is dynamically updated. The update frequency is either monthly or annually.

[0036] In one implementation, the dynamic update method for the similar year case library is as follows: Whenever an update node is reached, months that were not included in the dataset are added to the dataset, and the oldest month is removed from the dataset, wherein the number of months removed is the same as the number of months added; Update the similar year case library using the added dataset.

[0037] In subsequent business applications, the case library is dynamically updated so that the model can continuously track climate change and ensure that the historically similar cases used are always the most relevant to the recent climate situation.

[0038] Next, steps 106 and 108 will be explained.

[0039] In this embodiment of the invention, any region of China can be used as the target region for prediction, or the entire region of China can be used as the target region for prediction. When it is necessary to perform extended-period prediction on the target region, it is first necessary to obtain the monthly historical data of solar shortwave radiation of the target region over the past few years.

[0040] When identifying dominant spatial distribution types, for each identical month in recent years, the following steps can be performed: REOF analysis is conducted on a first set of historical solar shortwave radiation data for the current month to determine a second set of dominant spatial distribution types based on the rotational mode variance contribution rate; the matching degree between the radiation field of each year in the current month and each of the second set of dominant spatial distribution types is calculated, and each year is assigned to the dominant spatial distribution type with the highest matching degree for the current month. Here, the first set of data is equal to the number of years in recent years.

[0041] The second quantity can be a specified number, or it can be the number of variance contribution rates that exceed a specified value.

[0042] For example, taking January as an example for a target region, we can obtain historical solar shortwave radiation data for 44 January data points from 1981 to 2024. During REOF (Rotated Empirical Orthogonal Function) analysis, a grid matrix can be generated from these 44 January solar shortwave radiation data points, and then the grid matrix can be decomposed using the REOF. Assuming that after analysis, the variance contribution rate of the first three rotating modes (RPCs) exceeds 70% and their physical meaning is clear, then these three modes can be extracted as the dominant spatial distribution types for January. This can be understood as the first number being 44, and the second number being 3.

[0043] After performing REOF analysis, each mode corresponds to a spatial load vector map, which is used to show the typical distribution of "same-direction" or "opposite-direction" radiation changes in the target area under that type. Based on the spatial load vector map corresponding to each mode, its dominant spatial distribution type can be determined.

[0044] For example, for January, we can obtain the following three dominant spatial distribution types: Type 1 (Nationwide Consistently Strong Type): The load value is positive in most parts of the country, indicating that when this type is dominant, the solar radiation across the country is generally strong.

[0045] Type 2 (South Strong, North Weak): The load values ​​are positive in the south and negative in the north, showing a clear north-south opposite distribution.

[0046] Type 3 (High Plateau Strong - Eastern Weak): The load value is positive in the Qinghai-Tibet region, but negative in the central and eastern regions.

[0047] Then, for the radiation field of January in each year from 1981 to 2024, the matching degree with the three dominant spatial distribution types mentioned above is calculated, and it is assigned to the dominant spatial distribution type with the highest matching degree. Thus, the similar year type library for January can be obtained as shown in Table 2 below.

[0048] Table 2: In this way, a database of similar years for geothermal temperature differences and a database of similar years for solar shortwave radiation patterns can be obtained, providing a data foundation for the final prediction.

[0049] Finally, steps 110 and 114 will be explained.

[0050] In this embodiment of the invention, after the similar year case library and similar year type library have been prepared, the business operation phase can begin.

[0051] The following examples illustrate this section.

[0052] Assuming the current month is January 31, 2025, and the extended forecast is for the next month, February 2025, the standardized value for January can be determined as +1.2σ based on the geothermal temperature difference in the Qinghai-Tibet Plateau region in January. According to the classification criteria (σ≥1 indicates significantly higher values), January can be classified as significantly higher. Furthermore, all historical years corresponding to January being significantly higher can be determined using a similar year case library. These are assumed to be 1999, 2002, and 2007. These three similar years are determined based on the similarity of the geothermal temperature difference in January; however, the prediction object of this embodiment is the solar radiation field of these similar years in the next month (February).

[0053] Next, using a similar year type database, the dominant spatial distribution type of solar radiation in February for these three similar years is determined. Assuming February 1999 is assigned to Type I (nationwide uniformly strong type), February 2002 to Type II (strong in the south, weak in the north), and February 2007 to Type I (nationwide uniformly strong type), then, according to the synthesis logic, we can conclude that in the three similar years, the February radiation field is nationwide uniformly strong in two years (1999, 2007), and in one year (2002) it is strong in the south and weak in the north.

[0054] Finally, in generating the prediction results, one implementation can generate a probability prediction map and mark the prediction results on the map. For example, on the probability prediction map for February 2025, it can be marked that the probability of stronger solar radiation in most parts of the country is about 67% (2 / 3), of which the probability of stronger solar radiation in the south and weaker solar radiation in the north is about 33% (1 / 3).

[0055] Another approach is to generate a spatial distribution prediction map of radiation anomalies. For example, by taking the arithmetic mean of the actual radiation anomaly percentage fields for February in three similar years, a spatial distribution prediction map of radiation anomalies for February 2025 can be created. This map, used to comprehensively represent the characteristics of the two distribution patterns, might show "most parts of the country are experiencing stronger radiation, but the stronger radiation is more pronounced in the south."

[0056] When outputting prediction results, the prediction results can be output along with necessary textual descriptions, indicating high-confidence areas and main reference types.

[0057] The embodiments of this invention are dynamic, extended-period solar radiation prediction methods based on the similarity of land and temperature differences in key regions. They borrow traditional weather and climate prediction ideas (climate similarity) and apply them to solar shortwave radiation prediction. They innovatively select key prediction factors (land and temperature differences in the Qinghai-Tibet Plateau) and a core update mechanism (dynamic similarity field), forming a set of operational implementation plans.

[0058] In terms of timeliness, this invention enables extended-term forecasts with strong business applicability. Regarding stability, it eliminates reliance on fragile real-time data chains, constructing a highly robust, low-cost, and easily deployable business system. In terms of adaptability, the selection and dynamic update mechanism of the plateau temperature difference enable the model to self-evolve with climate change, exhibiting superior forecast stability and reliability in the face of the new normal of extreme climate conditions.

[0059] Please refer to Figure 2 This invention provides a device for predicting the extension of solar shortwave radiation, the device comprising: Acquisition unit 200 is used to acquire a dataset of the Qinghai-Tibet region over the past few years; the dataset includes a monthly historical geothermal temperature difference sequence. Division unit 202 is used to divide each month into corresponding categories among multiple categories based on the standard deviation of the historical geothermal temperature difference series for each month; different categories correspond to different standard deviation sizes; The first forming unit 204 is used to form a similar year case library based on the division results; the similar year case library includes a list of similar years for each category in each month; The identification unit 206 is used to acquire the monthly solar shortwave radiation history data of the target area in the past few years, so as to identify the dominant spatial distribution type of the target area on a monthly basis. The second forming unit 208 is used to form a similar year type library based on the dominant spatial distribution type of the target area on a monthly basis; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; The first determining unit 210 is used to determine the target category of the current month based on the ground temperature difference in the Qinghai-Tibet region in the current month, and to determine the target similar year list corresponding to the target category in the current month according to the similar year case library; The second determining unit 212 is used to determine the dominant spatial distribution type of each year in the target similar year list in the next month based on the similar year type library; Prediction unit 214 is used to generate solar shortwave radiation prediction results for the next month based on the dominant spatial distribution type corresponding to each year in the target similar year list in the next month, according to the synthesis logic.

[0060] In one embodiment of the present invention, the method for obtaining the monthly historical geothermal temperature difference sequence is as follows: Obtain daily historical temperature and geothermal data for the Qinghai-Tibet region over the past few years; Based on historical air temperature data and historical ground temperature data, the historical ground temperature difference sequence for each month in the Qinghai-Tibet region over the past few years was calculated. The historical geothermal temperature difference of each month in the Qinghai-Tibet Plateau over the past few years was averaged using the grid area weighted average method, resulting in a series of historical geothermal temperature differences for each month after the regional average.

[0061] In one embodiment of the present invention, the step of forming a similar year case library based on the division results includes: For each category, perform the following: For each month, determine if the month exists in the current category; if it does, count the years that the month is in the current category, and form a list of similar years for the current category in that month.

[0062] In one embodiment of the present invention, the similar year case library is dynamically updated.

[0063] In one embodiment of the present invention, the similar year case database is updated as follows: Whenever an update node is reached, months that were not included in the dataset are added to the dataset, and the oldest month is removed from the dataset, wherein the number of months removed is the same as the number of months added; Update the similar year case library using the added dataset.

[0064] In one embodiment of the present invention, identifying the dominant spatial distribution type of the target region month by month includes: For each identical month in recent years, the following steps are performed: REOF analysis is conducted on the first number of historical solar shortwave radiation data for the current identical month to determine the second number of dominant spatial distribution types based on the rotational mode variance contribution rate; the radiation field of each year in the current identical month is calculated, and the matching degree between it and each of the second number of dominant spatial distribution types is determined, and each year is assigned to the dominant spatial distribution type with the highest matching degree in the current identical month; the first number is equal to the number of years in recent years.

[0065] It should be noted that the solar shortwave radiation extension prediction device provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the solar shortwave radiation extension prediction device and the solar shortwave radiation extension prediction method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0066] Embodiments of this application also provide a computer device, please refer to... Figure 3 The computer device includes a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, at least one program, code set or instruction set being loaded and executed by the processor to implement the solar shortwave radiation extension prediction method provided in the above method embodiments.

[0067] Embodiments of this application also provide a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the solar shortwave radiation extension prediction method provided in the above-described method embodiments.

[0068] Embodiments of this application also provide a computer program product, which includes a computer program. A processor of a computer device reads the computer program from a computer-readable storage medium and executes the computer program, causing the computer device to perform the solar shortwave radiation extension prediction method described in any of the above embodiments.

[0069] For ease of description, the above systems or devices are described separately as various modules or units based on their functions. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware components.

[0070] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0071] Finally, it should be noted that in this document, relational terms such as target, second, third, and fourth are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0072] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for predicting the extension period of solar shortwave radiation, characterized in that, The method includes: Obtain a dataset of the Qinghai-Tibet region over the past few years; the dataset includes monthly historical geothermal temperature difference sequences. Based on the standard deviation of the historical geothermal temperature difference series for each month, each month is divided into corresponding categories within multiple categories; different categories correspond to different standard deviations. A similar year case library is formed based on the classification results; the similar year case library includes a list of similar years for each category in each month; Obtain monthly historical data of solar shortwave radiation in the target area over the past few years to identify the dominant spatial distribution type of the target area on a monthly basis. Based on the dominant spatial distribution type of the target region each month, a similar year type library is formed; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; Based on the temperature difference between the ground and air in the Qinghai-Tibet region in the current month, the target category of the current month is determined, and based on the similar year case library, a list of similar years corresponding to the target category in the current month is determined; Based on the similar year type library, determine the dominant spatial distribution type of each year in the target similar year list for the next month; Based on the dominant spatial distribution type of each year in the target similar year list for the next month, the solar shortwave radiation prediction results for the next month are generated according to the synthesis logic.

2. The method according to claim 1, characterized in that, The method for obtaining the monthly historical geothermal temperature difference series is as follows: Obtain daily historical temperature and geothermal data for the Qinghai-Tibet region over the past few years; Based on historical air temperature data and historical ground temperature data, the historical ground temperature difference sequence for each month in the Qinghai-Tibet region over the past few years was calculated. The historical geothermal temperature difference of each month in the Qinghai-Tibet Plateau over the past few years was averaged using the grid area weighted average method, resulting in a series of historical geothermal temperature differences for each month after the regional average.

3. The method according to claim 1, characterized in that, The process of forming a similar year case library based on the segmentation results includes: For each category, perform the following: For each month, determine if the month exists in the current category; if it does, count the years that the month is in the current category, and form a list of similar years for the current category in that month.

4. The method according to claim 1, characterized in that, The database of similar year cases is dynamically updated.

5. The method according to claim 4, characterized in that, The update method for the similar year case database is as follows: Whenever an update node is reached, months that were not included in the dataset are added to the dataset, and the oldest month is removed from the dataset, wherein the number of months removed is the same as the number of months added; Update the similar year case library using the added dataset.

6. The method according to any one of claims 1-5, characterized in that, The identification of the dominant spatial distribution type of the target region on a monthly basis includes: For each identical month in recent years, the following steps are performed: REOF analysis is conducted on the first number of historical solar shortwave radiation data for the current identical month to determine the second number of dominant spatial distribution types based on the rotational mode variance contribution rate; the radiation field of each year in the current identical month is calculated, and the matching degree between it and each of the second number of dominant spatial distribution types is determined, and each year is assigned to the dominant spatial distribution type with the highest matching degree in the current identical month; the first number is equal to the number of years in recent years.

7. A device for predicting the extension period of solar shortwave radiation, characterized in that, The device includes: The acquisition unit is used to acquire a dataset of the Qinghai-Tibet region over the past few years; the dataset includes a monthly historical geothermal temperature difference sequence. The partitioning unit is used to divide each month into corresponding categories among multiple categories based on the standard deviation of the historical geothermal temperature difference series; different categories correspond to different standard deviation sizes; The first forming unit is used to form a similar year case library based on the division results; the similar year case library includes a list of similar years for each category in each month; The identification unit is used to acquire monthly solar shortwave radiation history data of the target area over the past few years in order to identify the dominant spatial distribution type of the target area on a monthly basis. The second forming unit is used to form a similar year type library based on the dominant spatial distribution type of the target area on a monthly basis; the similar year type library includes a list of similar years for each month under each dominant spatial distribution type; The first determining unit is used to determine the target category of the current month based on the temperature difference between the ground and air in the Qinghai-Tibet region in the current month, and to determine the target similar year list corresponding to the target category in the current month based on the similar year case library; The second determining unit is used to determine the dominant spatial distribution type of each year in the target similar year list for the next month based on the similar year type library; The prediction unit is used to generate the solar shortwave radiation prediction results for the next month based on the dominant spatial distribution type corresponding to each year in the target similar year list in the next month, according to the synthesis logic.

8. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to execute the computer programs stored in the memory to implement the steps of the method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, implements the steps of the method according to any one of claims 1-6.

Images