Sunlight radiation intensity simulation method and system based on artificial intelligence

Abstract

The present application relates to the field of data processing technology, and in particular to a method and system for simulating sunlight radiation intensity based on artificial intelligence. The method includes: obtaining a first data set including solar radiation values ​​of a target monitoring point within a target time period, and a second data set including surface temperature values ​​of the target monitoring point within a target time period; determining a first fluctuation value of the first data set and a second fluctuation value of the second data set to determine an evaluation value of the solar radiation value; determining an abnormal solar radiation value and a non-abnormal solar radiation value according to the evaluation value of the solar radiation value; correcting the abnormal solar radiation value to obtain a first solar radiation value; and performing simulation prediction based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at the second moment. Through the above technical solution, the predicted target solar radiation value at the second moment can be made more accurate.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method and system for simulating sunlight radiation intensity based on artificial intelligence. Background Art

[0002] Solar energy is a clean and renewable energy source. There are two main ways to use solar energy: one is photothermal conversion, which is to collect sunlight through equipment such as solar collectors and convert solar energy into thermal energy for heating, hot water and other scenarios requiring thermal energy; the other is photoelectric conversion, which is to convert sunlight into electrical energy through equipment such as solar panels, which is used to power electrical equipment or charge energy storage devices to store electrical energy.

[0003] Before actually laying solar thermal collectors or solar photovoltaic panels, it is usually necessary to select the location for laying the solar thermal collectors or solar photovoltaic panels, and the solar radiation value is one of the important reference indicators for site selection. Therefore, the solar radiation value of the target location can be simulated and predicted to determine whether to lay the solar thermal collectors or solar photovoltaic panels at the target location.

[0004] In the related art, for example, a Chinese patent application document with publication number CN115545338A provides a solar radiation prediction and correction method based on a BP neural network, including: obtaining a predicted value of solar radiation, inputting a pre-constructed solar radiation prediction and correction model based on a BP neural network; and outputting a corrected predicted value of solar radiation through the solar radiation prediction and correction model based on the BP neural network.

[0005] However, the solar radiation value set may include some abnormal solar radiation values, making it difficult for the predicted values ​​obtained based on the abnormal solar radiation values ​​to reflect the actual situation of solar radiation. The predicted values ​​that are difficult to reflect the actual situation of solar radiation cannot provide valuable reference for the site selection of solar collectors or solar photovoltaic panels. Summary of the invention

[0006] In order to overcome the problem that the predicted values ​​obtained in the related art cannot provide valuable reference for the site selection of solar collectors or solar photovoltaic panels, the present application provides a sunlight radiation intensity simulation method and system based on artificial intelligence.

[0007] According to a first aspect of an embodiment of the present application, a method for simulating sunlight radiation intensity based on artificial intelligence is provided, including: obtaining a first data set including solar radiation values ​​of a target monitoring point within a target time period, and a second data set including surface temperature values ​​of the target monitoring point within the target time period; determining a first fluctuation value of the solar radiation values ​​in the first data set, and determining a second fluctuation value of the surface temperature values ​​in the second data set, and determining an evaluation value of the solar radiation value at a first moment in the target time period based on a first product of the first fluctuation value and the second fluctuation value; determining an abnormal solar radiation value and a non-abnormal solar radiation value from a plurality of solar radiation values ​​corresponding to a plurality of first moments based on the evaluation value of the solar radiation value; correcting the abnormal solar radiation value based on the non-abnormal solar radiation value adjacent to the moment corresponding to the abnormal solar radiation value to obtain a first solar radiation value; performing simulation prediction based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at a second moment, wherein the second moment is a future moment of the first moment.

[0008] In this way, by determining the evaluation value of the solar radiation value, and determining the abnormal solar radiation value and the non-abnormal solar radiation value from multiple solar radiation values ​​according to the evaluation value, the abnormal solar radiation value is corrected to obtain the first solar radiation value, so as to simulate and predict the target solar radiation value at the second moment, which can eliminate the influence of the abnormal solar radiation value on the prediction result, thereby making the predicted target solar radiation value at the second moment more accurate, and providing a valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0009] At the same time, the evaluation value of the solar radiation value is determined based on the first fluctuation value of the solar radiation value and the second fluctuation value of the surface temperature value. The evaluation value can effectively reflect the abnormal conditions of the solar radiation value and the surface temperature value of the target monitoring point. Therefore, compared with determining whether the solar radiation value at the first moment is abnormal only by the size of the solar radiation value, it is possible to determine whether the solar radiation value at the first moment is abnormal in combination with the surface temperature value of the target monitoring point, and the abnormal solar radiation value can be determined more effectively to eliminate the influence of the abnormal solar radiation value on the predicted result of the solar radiation value at the second moment.

[0010] Optionally, determining the first fluctuation value of the solar radiation value in the first data set includes: determining, based on the first data set, a first mean, information entropy and median of multiple solar radiation values ​​of the target monitoring point within the target time period; determining a first difference value between the first mean and the median; and using the first difference value and the second product of the information entropy as the first fluctuation value of the solar radiation value of the target monitoring point within the target time period.

[0011] In this way, the information entropy can reflect the complexity of multiple solar radiation values ​​in the first data set to characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The first difference value can also better characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The first difference value and the second product of the information entropy are used as the first fluctuation value of the solar radiation value of the target monitoring point within the target time period. The first fluctuation value can be used to better characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period, so as to determine whether there are abnormal solar radiation values ​​among the multiple solar radiation values ​​within the target time period.

[0012] Optionally, determining the second fluctuation value of the surface temperature value in the second data set includes: determining, based on the second data set, a first change in the surface temperature value of the target monitoring point within a target time period and a target time interval; determining a second mean of multiple first change amounts corresponding to multiple target time intervals, and using the second mean as the second fluctuation value of the surface temperature value in the second data set.

[0013] In this way, the second mean is used as the second fluctuation value of the surface temperature value in the second data set. The second fluctuation value can better characterize the fluctuation degree of the surface temperature value of the target monitoring point within the target time period, so as to determine whether there are abnormal solar radiation values ​​among the multiple solar radiation values ​​within the target time period.

[0014] Optionally, determining the evaluation value of the solar radiation value at the first moment in the target time period based on the first product of the first fluctuation value and the second fluctuation value includes: determining a third mean of multiple solar radiation values ​​of the target monitoring point in the target time period based on the first data set; determining a second difference value between the solar radiation value and the third mean, and determining a third product of the first product and the second difference value; obtaining the correlation coefficient between the surface temperature value and the solar radiation value, and using a first ratio of the third product to the correlation coefficient as the evaluation value of the solar radiation value at the first moment in the target time period.

[0015] In this way, the first product value can comprehensively reflect the fluctuation degree of the surface temperature value of the target monitoring point within the target time period and the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The third product value is obtained by multiplying the first product value and the second difference value. The third product value can better reflect the possible abnormalities in the solar radiation value of the target monitoring point.

[0016] Optionally, the correlation coefficient is obtained in the following manner: determine the first percentile of the solar radiation value at the first moment in the first data set, and determine the second percentile of the surface temperature value at the first moment in the second data set; determine a third difference value between the first percentile and the second percentile corresponding to the same first moment, and determine the average difference value of multiple third difference values ​​corresponding to multiple first moments; for a target moment in a target time period, determine a fourth difference value between the third difference value corresponding to the target moment and the average difference value; use an exponential function to perform an exponential operation on the fourth difference value to obtain a first exponential operation value, and use the inverse of the first exponential operation value as the correlation coefficient of the target moment.

[0017] Optionally, according to the evaluation value of the solar radiation value, abnormal solar radiation values ​​and non-abnormal solar radiation values ​​are determined from multiple solar radiation values ​​corresponding to multiple first moments, including: determining an average evaluation value of the multiple evaluation values, and performing an exponential operation on the average evaluation value using an exponential function to obtain a second exponential operation value; determining the upper quartile and the lower quartile of the multiple solar radiation values, and determining the abnormal upper limit value and the abnormal lower limit value of the box plot according to the inverse of the second exponential operation value, the upper quartile and the lower quartile, respectively; and using the abnormal upper limit value and the abnormal lower limit value to determine abnormal solar radiation values ​​and non-abnormal solar radiation values ​​from multiple solar radiation values ​​corresponding to multiple first moments.

[0018] In this way, since the second index operation value is equal to the average evaluation value of multiple evaluation values, the average evaluation value can characterize the overall abnormality of the solar radiation values ​​in the first data set. The abnormal upper limit value and the abnormal lower limit value of the box plot are determined according to the reciprocal, upper quartile and lower quartile of the second index operation value, respectively. The abnormal upper limit value and the abnormal lower limit value can be determined in combination with the overall abnormality of the solar radiation values ​​in the first data set, which helps to more accurately determine abnormal solar radiation values ​​and non-abnormal solar radiation values ​​from the first data set.

[0019] Optionally, the abnormal solar radiation value is corrected according to the non-abnormal solar radiation value adjacent to the moment corresponding to the abnormal solar radiation value to obtain a first solar radiation value, including: obtaining a first average change in solar radiation values ​​at adjacent moments in the first data set, and determining a second average change in surface temperature values ​​at adjacent moments in the second data set; determining a second ratio of the first average change to the second average change; for a first abnormal radiation value among multiple abnormal solar radiation values, determining a fifth difference between the surface temperature value at a third moment in a target time period and the surface temperature value at a moment adjacent to the third moment; obtaining a fourth product of the second ratio and the fifth difference, and subtracting the fourth product from the non-abnormal solar radiation value adjacent to the first abnormal radiation value at the corresponding moment to obtain the first solar radiation value.

[0020] Optionally, a simulation prediction is performed based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at the second moment, including: inputting the first solar radiation value and the non-abnormal solar radiation value into a pre-trained prediction model to obtain the target solar radiation value at the second moment output by the prediction model; wherein the prediction model is obtained by training a pre-constructed initial model using a training set and a test set, the training set includes the solar radiation values ​​of the first time period, and the test set includes the solar radiation values ​​of the second time period, the second time period being later than the first time period.

[0021] Optionally, the method also includes: constructing a target data set including the first solar radiation value and the non-abnormal solar radiation value; inputting the target data set into an animation generation model to obtain a target animation output by the animation generation model, wherein the images at different moments in the target animation correspond to different solar radiation values; and sending the target animation to a terminal device so that the terminal device displays the target animation.

[0022] In this way, the images at different times in the target animation correspond to different solar radiation values. By sending the target animation to the terminal device, the user of the terminal device can more intuitively view the change process of the solar radiation value of the target monitoring point within the target time period through the terminal device.

[0023] According to the second aspect of an embodiment of the present application, there is provided a sunlight radiation intensity simulation system based on artificial intelligence, comprising: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the steps of the sunlight radiation intensity simulation method based on artificial intelligence provided in the first aspect of the present application are implemented.

[0024] The technical solution provided by the embodiments of the present application may include the following beneficial effects: by determining an evaluation value of a solar radiation value, and determining an abnormal solar radiation value and a non-abnormal solar radiation value from a plurality of solar radiation values ​​according to the evaluation value, the abnormal solar radiation value is corrected to obtain a first solar radiation value, so as to simulate and predict the target solar radiation value at the second moment, thereby eliminating the influence of the abnormal solar radiation value on the prediction result, thereby making the predicted target solar radiation value at the second moment more accurate, and providing a valuable reference for the site selection of a solar collector or a solar photovoltaic panel.

[0025] At the same time, the evaluation value of the solar radiation value is determined based on the first fluctuation value of the solar radiation value and the second fluctuation value of the surface temperature value. The evaluation value can effectively reflect the abnormal conditions of the solar radiation value and the surface temperature value of the target monitoring point. Therefore, compared with determining whether the solar radiation value at the first moment is abnormal only by the size of the solar radiation value, it is possible to determine whether the solar radiation value at the first moment is abnormal in combination with the surface temperature value of the target monitoring point, and the abnormal solar radiation value can be determined more effectively to eliminate the influence of the abnormal solar radiation value on the predicted result of the solar radiation value at the second moment.

[0026] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0028] Figure 1 is a flow chart of a method for simulating sunlight radiation intensity based on artificial intelligence according to an exemplary embodiment;

[0029] Figure 2 It is a structural schematic diagram of a sunlight radiation intensity simulation system based on artificial intelligence according to an exemplary embodiment. DETAILED DESCRIPTION

[0030] Here, exemplary embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application.

[0031] First, the application scenario of the embodiment of the present application is briefly introduced. In the application scenario of the present application, before the actual laying of the solar collector or solar photovoltaic panel, it is usually necessary to select the laying location of the solar collector or solar photovoltaic panel, and the solar radiation value is one of the important reference indicators for site selection. Therefore, the solar radiation value of the target location can be simulated and predicted to determine whether to lay the solar collector or solar photovoltaic panel at the target location.

[0032] However, currently, when simulating and predicting the solar radiation value of the target location, the prediction is mainly made through multiple solar radiation values ​​of the target location in the historical time period, and there may be some abnormal solar radiation values ​​in the multiple solar radiation values ​​of the historical time period. For example, the light at the target location is temporarily blocked and caused by an abnormality, or there is a temporary abnormality in the radiation measuring device.

[0033] Predicting the solar radiation value of the target location based on the abnormal solar radiation value will affect the prediction accuracy of the solar radiation value of the target location, so that the obtained prediction results cannot provide valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0034] In response to the above technical problems, the embodiments of the present application provide a sunlight radiation intensity simulation method and system based on artificial intelligence, which determines the evaluation value of the solar radiation value, and determines the abnormal solar radiation value and the non-abnormal solar radiation value from multiple solar radiation values ​​according to the evaluation value, thereby correcting the abnormal solar radiation value to obtain the first solar radiation value, so as to simulate and predict the target solar radiation value at the second moment, thereby eliminating the influence of the abnormal solar radiation value on the prediction result, thereby making the predicted target solar radiation value at the second moment more accurate, and providing a valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0035] At the same time, the evaluation value of the solar radiation value is determined based on the first fluctuation value of the solar radiation value and the second fluctuation value of the surface temperature value. The evaluation value can effectively reflect the abnormal conditions of the solar radiation value and the surface temperature value of the target monitoring point. Therefore, compared with determining whether the solar radiation value at the first moment is abnormal only by the size of the solar radiation value, it is possible to determine whether the solar radiation value at the first moment is abnormal in combination with the surface temperature value of the target monitoring point, and the abnormal solar radiation value can be determined more effectively to eliminate the influence of the abnormal solar radiation value on the predicted result of the solar radiation value at the second moment.

[0036] Figure 1 is a flow chart of a method for simulating sunlight radiation intensity based on artificial intelligence according to an exemplary embodiment. Figure 1 As shown, the method includes the following steps.

[0037] In step S101, a first data set including solar radiation values ​​of a target monitoring point within a target time period and a second data set including ground surface temperature values ​​of the target monitoring point within a target time period are acquired.

[0038] In the embodiment of the present application, the solar radiation value may be the total solar radiation, which refers to the sum of the solar radiation energy reaching the earth's surface, and the total solar radiation may include direct radiation and scattered radiation. By installing a radiation measurement device at the target monitoring point, such as setting a radiation meter at the target monitoring point, the total solar radiation of the target monitoring point may be measured, thereby obtaining a first data set, which includes multiple solar radiation values ​​of the target monitoring point at multiple times within a target time period.

[0039] The target monitoring point may be a monitoring position among a plurality of preset monitoring positions, and different preset monitoring positions are located at different spatial positions.

[0040] Changes in surface temperature values ​​will also be affected by solar radiation values. For example, the larger the solar radiation value of the target monitoring point, the more intense the sun's radiation on the target monitoring point, and the higher the surface temperature of the target monitoring point is generally. On the contrary, the smaller the solar radiation value of the target monitoring point, the weaker the sun's radiation on the target monitoring point, and the lower the surface temperature of the target monitoring point is generally. The surface temperature value of the target monitoring point may refer to the temperature value of the surface at the same horizontal position as the target monitoring point.

[0041] The surface temperature value of the target monitoring point can be acquired through the temperature sensor, and the second data set may include multiple surface temperature values ​​of the target monitoring point within the target time period.

[0042] In step S102, a first fluctuation value of the solar radiation value in the first data set is determined, and a second fluctuation value of the surface temperature value in the second data set is determined, and an evaluation value of the solar radiation value at the first moment in the target time period is determined based on a first product of the first fluctuation value and the second fluctuation value.

[0043] The first fluctuation value can characterize the degree of fluctuation of the solar radiation value of the target monitoring point within the target time period; the larger the first fluctuation value, the greater the degree of fluctuation of the solar radiation value of the target monitoring point within the target time period; conversely, the smaller the first fluctuation value, the smaller the degree of fluctuation of the solar radiation value of the target monitoring point within the target time period.

[0044] The second fluctuation value can characterize the degree of fluctuation of the surface temperature value of the target monitoring point within the target time period; the larger the second fluctuation value, the greater the degree of fluctuation of the surface temperature value of the target monitoring point within the target time period; on the contrary, the smaller the first fluctuation value, the smaller the degree of fluctuation of the surface temperature value of the target monitoring point within the target time period.

[0045] In this way, the evaluation value of the solar radiation value at the first moment in the target time period is determined according to the first product of the first fluctuation value and the second fluctuation value, which can comprehensively consider the fluctuation degree of the surface temperature value of the target monitoring point in the target time period and the fluctuation degree of the solar radiation value of the target monitoring point in the target time period, and help to better determine whether there are abnormal solar radiation values ​​in the first data set, so as to provide a more valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0046] In one embodiment, determining the first fluctuation value of the solar radiation value in the first data set includes: determining, based on the first data set, a first mean, information entropy and median of multiple solar radiation values ​​of the target monitoring point within the target time period; determining a first difference value between the first mean and the median; and using the first difference value and the second product of the information entropy as the first fluctuation value of the solar radiation value of the target monitoring point within the target time period.

[0047] The first mean of multiple solar radiation values ​​within the target time period can better reflect the average level of solar radiation values ​​at the target monitoring point within the target time period; the median of multiple solar radiation values ​​within the target time period can better reflect the general level of solar radiation values ​​at the target monitoring point within the target time period.

[0048] The first difference value between the first mean and the median can better reflect the fluctuation of the solar radiation value of the target monitoring point within the target time period. For example, the larger the first difference value between the first mean and the median, the greater the fluctuation of the solar radiation value of the target monitoring point within the target time period. On the contrary, the smaller the first difference value between the first mean and the median, the smaller the fluctuation of the solar radiation value of the target monitoring point within the target time period. The first difference value can better characterize the fluctuation of the solar radiation value of the target monitoring point within the target time period.

[0049] The information entropy of multiple solar radiation values ​​of the target monitoring point within the target time period can reflect the complexity of the multiple solar radiation values ​​in the first data set; for example, in chronological order, the multiple solar radiation values ​​within the target time period are T1, T2, T3, T4, T1, T1, T2, T2, T3 and T3, among which T1, T2, T3 and T4 are solar radiation values ​​with different values. The information entropy can effectively characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period.

[0050] In this way, the information entropy can reflect the complexity of multiple solar radiation values ​​in the first data set to characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The first difference value can also better characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The first difference value and the second product of the information entropy are used as the first fluctuation value of the solar radiation value of the target monitoring point within the target time period. The first fluctuation value can be used to better characterize the fluctuation degree of the solar radiation value of the target monitoring point within the target time period, so as to determine whether there are abnormal solar radiation values ​​among the multiple solar radiation values ​​within the target time period.

[0051] In one embodiment, determining the second fluctuation value of the surface temperature value in the second data set includes: determining, based on the second data set, a first change in the surface temperature value of the target monitoring point within a target time period and a target time interval; determining a second mean of multiple first change amounts corresponding to multiple target time intervals, and using the second mean as the second fluctuation value of the surface temperature value in the second data set.

[0052] The first change in the surface temperature value of the target monitoring point at the target time interval within the target time period can reflect the change frequency or change amplitude of the surface temperature value of the target monitoring point, and the second average of multiple first change amounts corresponding to multiple target time intervals can comprehensively reflect the overall change frequency or change amplitude of the surface temperature value of the target monitoring point within the target time period.

[0053] In this way, the second mean is used as the second fluctuation value of the surface temperature value in the second data set. The second fluctuation value can better characterize the fluctuation degree of the surface temperature value of the target monitoring point within the target time period, so as to determine whether there are abnormal solar radiation values ​​among the multiple solar radiation values ​​within the target time period.

[0054] In one embodiment, an evaluation value of the solar radiation value at the first moment in the target time period is determined based on a first product of the first fluctuation value and the second fluctuation value, including: determining a third mean of multiple solar radiation values ​​of the target monitoring point in the target time period based on the first data set; determining a second difference value between the solar radiation value and the third mean, and determining a third product of the first product and the second difference value; obtaining a correlation coefficient between the surface temperature value and the solar radiation value, and using a first ratio of the third product to the correlation coefficient as the evaluation value of the solar radiation value at the first moment in the target time period.

[0055] The third mean of multiple solar radiation values ​​of the target monitoring point within the target time period can reflect the average level of the solar radiation value of the target monitoring point within the target time period. The second difference value between the solar radiation value and the third mean can reflect the difference between the solar radiation value and the average level. The second difference value can, for example, be the absolute value of the difference between the solar radiation value and the third mean. The second difference value can characterize the degree of difference between the solar radiation value and the third mean.

[0056] Here, the first product value can comprehensively reflect the fluctuation degree of the surface temperature value of the target monitoring point within the target time period and the fluctuation degree of the solar radiation value of the target monitoring point within the target time period. The third product value is obtained by multiplying the first product value and the second difference value. The third product value can better reflect the possible abnormalities in the solar radiation value of the target monitoring point.

[0057] Since the correlation coefficient can characterize the degree of correlation between the solar radiation value and the surface temperature value, by obtaining the correlation coefficient between the surface temperature value and the solar radiation value, and taking the third product value and the first ratio of the correlation coefficient as the evaluation value of the solar radiation value at the first moment in the target time period, it is possible to better determine whether there is an abnormality in the solar radiation value at the first moment based on the evaluation value.

[0058] For example, when the third multiplication value is the same, the larger the correlation coefficient between the surface temperature value and the solar radiation value, the higher the correlation between the surface temperature value and the solar radiation value at the first moment, so that the first ratio of the third multiplication value to the correlation coefficient is smaller, and the probability that the solar radiation value at the first moment in the target time period is abnormal is lower; on the contrary, when the third multiplication value is the same, the smaller the correlation coefficient between the surface temperature value and the solar radiation value, the lower the correlation between the surface temperature value and the solar radiation value at the first moment, so that the larger the first ratio of the third multiplication value to the correlation coefficient is, the higher the probability that the solar radiation value at the first moment in the target time period is abnormal.

[0059] In one embodiment, the correlation coefficient is obtained in the following manner: determining the first percentile of the solar radiation value at the first moment in the first data set, and determining the second percentile of the surface temperature value at the first moment in the second data set; determining a third difference value between the first percentile and the second percentile corresponding to the same first moment, and determining the average difference value of multiple third difference values ​​corresponding to multiple first moments; for a target moment in a target time period, determining a fourth difference value between the third difference value corresponding to the target moment and the average difference value; performing an exponential operation on the fourth difference value using an exponential function to obtain a first exponential operation value, and taking the reciprocal of the first exponential operation value as the correlation coefficient of the target moment.

[0060] Percentiles are used to describe the relative position of a value in a data set in the overall data. Percentiles refer to a specific percentage of data values ​​in a data set that are lower than or equal to a certain value. For example, for a specific value in a data set corresponding to the 80th percentile, it means that 80% of the data in the data set are lower than or equal to this specific value, while 20% of the data in the data set are higher than this specific value. Percentiles can be calculated, for example, in the following way: sort the data in the data set in descending order to obtain a sorted sequence, determine the order of the values ​​of the target data in the data set in the sorted sequence, and thus determine the percentile corresponding to the target data.

[0061] The first percentile of the solar radiation value at the first moment in the first data set can reflect the difference in numerical value between the solar radiation value at the first moment and other solar radiation values ​​in the first data set. For example, when the first percentile is 80%, it means that the solar radiation value at the first moment is higher than 80% of other solar radiation values ​​in the first data set.

[0062] The second percentile of the surface temperature value at the first moment in the second data set can reflect the difference in numerical value between the surface temperature value at the first moment and other surface temperature values ​​in the second data set. For example, when the second percentile is 80%, it means that the surface temperature value at the first moment is higher than 80% of other surface temperature values ​​in the second data set.

[0063] The third difference value between the first percentile and the second percentile corresponding to the same first moment can reflect the correlation between the surface temperature value and the solar radiation value of the target monitoring point at the same moment. The average difference value of multiple third difference values ​​corresponding to multiple first moments can characterize the overall correlation between the surface temperature value and the solar radiation value of the target monitoring point in the target time period.

[0064] For example, the larger the third difference value between the first percentile and the second percentile corresponding to the same first moment, the lower the correlation between the surface temperature value and the solar radiation value at the first moment; on the contrary, the smaller the third difference value between the first percentile and the second percentile corresponding to the same first moment, the higher the correlation between the surface temperature value and the solar radiation value at the first moment.

[0065] For the target moment in the target time period, the fourth difference value between the third difference value corresponding to the target moment and the average difference value is determined. Then, based on the overall correlation between the surface temperature value and the solar radiation value in the target time period, the overall correlation between the surface temperature value and the solar radiation value at the first moment can be determined, so that the correlation coefficient better matches the actual situation at the first moment.

[0066] The fourth difference value is subjected to exponential operation using an exponential function to obtain a first exponential operation value, and the reciprocal of the first exponential operation value is used as the correlation coefficient at the target moment, which can not only realize the normalization processing of the fourth difference value, but also make the correlation coefficient negatively correlated with the fourth difference value.

[0067] The following is an example of an exemplary calculation formula to explain the process of obtaining the correlation coefficient between the surface temperature value and the solar radiation value in the embodiment of the present application: ; Where Z is the correlation coefficient between the surface temperature value and the solar radiation value at the first moment, exp is an exponential function with a natural constant as the base, and b 1 is the first percentile of the solar radiation value at the first moment in the first data set, b 2 is the second percentile of the surface temperature value at the first moment in the second data set, b 3 is the average difference value of multiple third difference values ​​corresponding to multiple first moments.

[0068] In step S103, according to the evaluation value of the solar radiation value, an abnormal solar radiation value and a non-abnormal solar radiation value are determined from a plurality of solar radiation values ​​corresponding to a plurality of first moments.

[0069] The evaluation values ​​of the solar radiation values ​​of the same target monitoring point at different times are different, and the evaluation values ​​of the solar radiation values ​​can be used to characterize the probability that the solar radiation values ​​are abnormal. Therefore, by using the evaluation values ​​of the solar radiation values, the abnormal solar radiation values ​​and the non-abnormal solar radiation values ​​can be determined from the multiple solar radiation values ​​corresponding to the multiple first moments; wherein the abnormal solar radiation value is the solar radiation value when the illumination of the target monitoring point is abnormal, and the non-abnormal solar radiation value is the solar radiation value when the illumination of the target monitoring point is not abnormal.

[0070] In one embodiment, according to the evaluation value of the solar radiation value, abnormal solar radiation values ​​and non-abnormal solar radiation values ​​are determined from multiple solar radiation values ​​corresponding to multiple first moments, including: determining an average evaluation value of the multiple evaluation values, and performing an exponential operation on the average evaluation value using an exponential function to obtain a second exponential operation value; determining the upper quartile and the lower quartile of the multiple solar radiation values, and determining the abnormal upper limit value and the abnormal lower limit value of the box plot according to the reciprocal, the upper quartile and the lower quartile of the second exponential operation value, respectively; and determining the abnormal solar radiation value and non-abnormal solar radiation value from the multiple solar radiation values ​​corresponding to the multiple first moments using the abnormal upper limit value and the abnormal lower limit value.

[0071] The box plot is also called a box-shaped plot or a box-whisker plot. It is a statistical plot used to show the range and distribution of a set of data. The box plot is based on the concept of quartiles, dividing the data set into four equal parts, each containing 25% of the data. The box plot can more easily identify the maximum, minimum, median and outliers of the data, or the box plot can show the overall distribution of the data and the degree of outliers of different data, so as to determine the abnormal data in the data set.

[0072] In the embodiment of the present application, the average evaluation value of multiple evaluation values ​​can comprehensively reflect the average level of abnormality of the solar radiation value in the target time period, which is convenient for the subsequent determination of whether there is any abnormality in the solar radiation value at the first moment in the target time period.

[0073] The lower quartile is the 25th percentile in the data set, and the lower quartile is located at the position of the smallest value of the middle 25% after the data is sorted from small to large; the upper quartile is the 75th percentile in the data set, and the upper quartile is located at the position of the largest value of the middle 50% after the data is sorted from small to large.

[0074] In this way, since the second index operation value is equal to the average evaluation value of multiple evaluation values, the average evaluation value can characterize the overall abnormality of the solar radiation values ​​in the first data set. The abnormal upper limit value and the abnormal lower limit value of the box plot are determined according to the reciprocal, upper quartile and lower quartile of the second index operation value, respectively. The abnormal upper limit value and the abnormal lower limit value can be determined in combination with the overall abnormality of the solar radiation values ​​in the first data set, which helps to more accurately determine abnormal solar radiation values ​​and non-abnormal solar radiation values ​​from the first data set.

[0075] In one embodiment, the abnormal upper limit value and the abnormal lower limit value of the box plot are determined according to the reciprocal of the second exponential operation value, the upper quartile and the lower quartile, respectively, including: determining a first difference between the upper quartile and the lower quartile; multiplying the first difference, the reciprocal of the second exponential operation value and a preset value to obtain a fifth multiplication value; adding the upper quartile and the fifth multiplication value to obtain the abnormal upper limit value; subtracting the lower quartile from the fifth multiplication value to obtain the abnormal lower limit value; wherein the preset value can be 1.5.

[0076] The following uses an exemplary calculation formula as an example to explain the process of obtaining the abnormal upper limit value and the abnormal lower limit value in the embodiment of the present application. 1 =M 2 +1.5×E×(M 2 -M 1 ), P 1 is the abnormal upper limit, M 2 is the upper quartile, M 1is the lower quartile, E is the reciprocal of the second exponential operation value; P 2 =M 1 -1.5×E×(M 2 -M 1 ), P 2 The abnormal lower limit.

[0077] According to the acquired abnormal upper limit value and abnormal lower limit value, the solar radiation value in the first data set that is between the abnormal lower limit value and the abnormal upper limit value can be taken as a non-abnormal solar radiation value, and the solar radiation value in the first data set that is greater than the abnormal upper limit value or less than the abnormal lower limit value can be taken as an abnormal solar radiation value.

[0078] The second index operation value is used to characterize the overall abnormality of the solar radiation values ​​in the first data set. The larger the second index operation value, the smaller the inverse of the second index operation value, so that the value range of the target range formed between the abnormal lower limit value and the abnormal upper limit value is smaller, the number of non-abnormal solar radiation values ​​within the target range is smaller, and the number of abnormal solar radiation values ​​outside the target range is greater.

[0079] On the contrary, the smaller the second exponential operation value is, the larger the inverse of the second exponential operation value is, so that the value range of the target range formed between the abnormal lower limit value and the abnormal upper limit value is larger, the number of non-abnormal solar radiation values ​​within the target range is greater, and the number of abnormal solar radiation values ​​outside the target range is smaller.

[0080] In this way, when the evaluation value is larger, the number of abnormal solar radiation values ​​obtained is greater; on the contrary, when the evaluation value is smaller, the number of abnormal solar radiation values ​​obtained is smaller. Compared with directly selecting abnormal solar radiation values ​​according to the size of the solar radiation value in the first data set, more accurate abnormal solar radiation values ​​can be obtained, and it is also convenient to determine more accurate non-abnormal solar radiation values, thereby obtaining more accurate prediction results of solar radiation values ​​at future moments of the target moment.

[0081] In step S104, the abnormal solar radiation value is corrected according to the non-abnormal solar radiation value adjacent to the time corresponding to the abnormal solar radiation value to obtain a first solar radiation value.

[0082] For the abnormal solar radiation values ​​and non-abnormal solar radiation values ​​in the first data set, in order to avoid the influence of the abnormal solar radiation values ​​on the predicted results of the solar radiation values ​​of the target location, the abnormal solar radiation values ​​can be corrected to obtain the first solar radiation values.

[0083] In one embodiment, the abnormal solar radiation value is corrected according to the non-abnormal solar radiation value adjacent to the moment corresponding to the abnormal solar radiation value to obtain a first solar radiation value, including: obtaining a first average change in solar radiation values ​​at adjacent moments in a first data set, and determining a second average change in surface temperature values ​​at adjacent moments in a second data set; determining a second ratio of the first average change to the second average change; for a first abnormal radiation value among multiple abnormal solar radiation values, determining a fifth difference between a surface temperature value at a third moment in a target time period and a surface temperature value at a moment adjacent to the third moment; obtaining a fourth product of the second ratio and the fifth difference, and subtracting the fourth product from the non-abnormal solar radiation value adjacent to the first abnormal radiation value at the corresponding moment to obtain the first solar radiation value.

[0084] For example, when the solar radiation value at 08:00:00 is determined to be an abnormal solar radiation value based on the evaluation value, if the solar radiation value at 08:00:01 is a non-abnormal solar radiation value, the solar radiation value at 08:00:01 can be used to correct the solar radiation value at 08:00:00; or, if the solar radiation value at 07:59:59 is also a non-abnormal solar radiation value, the solar radiation value at 08:00:01 and the solar radiation value at 07:59:59 can also be used to correct the solar radiation value at 08:00:00.

[0085] The first average change in solar radiation values ​​at adjacent moments in the first data set can reflect the average level of change in solar radiation values ​​at adjacent moments in the first data set; the second average change in surface temperature values ​​at adjacent moments in the second data set can reflect the average level of change in surface temperature values ​​at adjacent moments in the second data set, and the second ratio of the first average change to the second average change can reflect the relative size of the change in solar radiation values ​​and the change in surface temperature values ​​at adjacent moments in the second data set.

[0086] The fifth difference value between the surface temperature value at the third moment and the surface temperature value at the adjacent moment of the third moment can reflect the change in the surface temperature value at the third moment. The fourth multiplication value of the second ratio and the fifth difference value helps to correct the abnormal solar radiation value according to the average level of the change in the solar radiation values ​​at adjacent moments in the first data set, obtain the first solar radiation value, and avoid the adverse effects of the abnormal solar radiation value on the prediction process.

[0087] In step S105, a simulation prediction is performed based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at the second moment.

[0088] The second moment is a future moment of the first moment. For example, when the first moment is 08:00:00, the second moment may be 08:05:00. According to the first solar radiation value and the non-abnormal solar radiation value, the target solar radiation value of the second moment may be simulated and predicted, so as to provide a valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0089] In one embodiment, a simulation prediction is performed based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at the second moment, including: inputting the first solar radiation value and the non-abnormal solar radiation value into a pre-trained prediction model to obtain the target solar radiation value at the second moment output by the prediction model.

[0090] The prediction model is obtained by training a pre-built initial model using a training set and a test set, wherein the training set includes solar radiation values ​​in a first time period, and the test set includes solar radiation values ​​in a second time period, and the second time period is later than the first time period.

[0091] For example, the first time period may be from 09:00:00 to 09:30:00, and the second time period may be from 09:40:00 to 09:50:00. A plurality of solar radiation values ​​within the first time period may be used as a training set, and a plurality of second solar radiation values ​​within the second time period may be used as a test set to train a pre-constructed initial model. When the loss value of the loss function of the initial model meets a preset condition, it is determined that the training of the pre-constructed initial model is completed.

[0092] In one embodiment, the artificial intelligence-based sunlight radiation intensity simulation method also includes: constructing a target data set including the first solar radiation value and the non-abnormal solar radiation value; inputting the target data set into an animation generation model to obtain a target animation output by the animation generation model, wherein the images at different moments in the target animation correspond to different solar radiation values; and sending the target animation to a terminal device so that the terminal device displays the target animation.

[0093] An animation generation model refers to a model that uses artificial intelligence technology to create animation content. The animation generation model can generate animations based on text descriptions, algorithms, or other types of inputs. The animation generation model can, for example, be a generative adversarial network or a variational autoencoder, where the generative adversarial network includes a generator and a decoder.

[0094] Here, the images at different times in the target animation correspond to different solar radiation values. By sending the target animation to the terminal device, the user of the terminal device can more intuitively view the change process of the solar radiation value of the target monitoring point within the target time period through the terminal device.

[0095] By using the artificial intelligence-based sunlight radiation intensity simulation method provided in the embodiment of the present application, an evaluation value of the solar radiation value is determined, and based on the evaluation value, abnormal solar radiation values ​​and non-abnormal solar radiation values ​​are determined from multiple solar radiation values, so that the abnormal solar radiation value is corrected to obtain a first solar radiation value, so as to simulate and predict the target solar radiation value at the second moment, which can eliminate the influence of the abnormal solar radiation value on the prediction result, thereby making the predicted target solar radiation value at the second moment more accurate, and providing a valuable reference for the site selection of solar collectors or solar photovoltaic panels.

[0096] At the same time, the evaluation value of the solar radiation value is determined based on the first fluctuation value of the solar radiation value and the second fluctuation value of the surface temperature value. The evaluation value can effectively reflect the abnormal conditions of the solar radiation value and the surface temperature value of the target monitoring point. Therefore, compared with determining whether there is an abnormality in the solar radiation value at the first moment only by the size of the solar radiation value, it is possible to determine whether there is an abnormality in the solar radiation value at the first moment in combination with the surface temperature value of the target monitoring point, and can more effectively determine the abnormal solar radiation value so as to eliminate the influence of the abnormal solar radiation value on the predicted result of the solar radiation value at the second moment.

[0097] Figure 2 FIG. 1 is a schematic diagram of a sunlight radiation intensity simulation system 1000 based on artificial intelligence according to an exemplary embodiment. Figure 2 The artificial intelligence-based sunlight radiation intensity simulation system 1000 includes: a processor 1100 and a memory 1200, wherein the memory 1200 stores computer program instructions, and when the computer program instructions are executed by the processor 1100, all or part of the steps of the artificial intelligence-based sunlight radiation intensity simulation method in the present application are implemented.

[0098] It should be understood that the features of some embodiments of the various applications described herein may be combined with each other, unless specifically stated otherwise.

[0099] Although terms such as "first", "second", and "third" may be used herein to describe various components, parts, regions, layers, or sections, these components, parts, regions, layers, or sections are not limited to these terms. Instead, these terms are only used to distinguish one component, part, region, layer, or section from another component, part, region, layer, or section. Therefore, without departing from the teachings of the examples described herein, the first component, part, region, layer, or section mentioned in the examples may also be referred to as the second component, part, region, layer, or section.

[0100] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description herein, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0101] In addition, the word "exemplary" is used herein to mean serving as an example, instance, or diagram. Any aspect or design described herein as "exemplary" is not necessarily to be understood as being advantageous over other aspects or designs. Instead, the use of the word exemplary is intended to present concepts in a concrete manner. As used herein, the term "or" is intended to mean an inclusive "or" rather than an exclusive "or".

[0102] Likewise, although the present application has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. With particular regard to the various functions performed by the components (e.g., elements, resources, etc.) described above, unless otherwise noted, the terms used to describe such components are intended to correspond to any component (functionally equivalent) that performs the specific functions of the described components, even if not structurally equivalent to the disclosed structures.

[0103] Additionally, while particular features of the present application may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations as may be desired and advantageous for any given or particular application.

[0104] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary technical means in the art that are not disclosed in the present application, and the specification and embodiments are only considered as exemplary.

[0105] It will be appreciated that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A method for simulating sunlight radiation intensity based on artificial intelligence, characterized in that: include: Acquire a first data set including solar radiation values ​​at a target monitoring point within a target time period, and a second data set including ground surface temperature values ​​at the target monitoring point within the target time period; Determine a first fluctuation value of the solar radiation value in the first data set, determine a second fluctuation value of the surface temperature value in the second data set, and determine an evaluation value of the solar radiation value at a first moment in a target time period according to a first multiplication value of the first fluctuation value and the second fluctuation value; Determine, according to the evaluation value of the solar radiation value, an abnormal solar radiation value and a non-abnormal solar radiation value from a plurality of solar radiation values ​​corresponding to a plurality of first moments; Correcting the abnormal solar radiation value according to a non-abnormal solar radiation value adjacent to the time corresponding to the abnormal solar radiation value to obtain a first solar radiation value; Perform simulation prediction based on the first solar radiation value and the non-abnormal solar radiation value to obtain a target solar radiation value at a second moment, where the second moment is a future moment of the first moment; Wherein, determining the first fluctuation value of the solar radiation value in the first data set includes: determining, according to the first data set, a first mean value, information entropy, and median of multiple solar radiation values ​​of the target monitoring point within the target time period; determining a first difference value between the first mean value and the median; and using the first difference value and the second product of the information entropy as the first fluctuation value of the solar radiation value of the target monitoring point within the target time period; Determining a second fluctuation value of the ground surface temperature value in the second data set includes: determining, based on the second data set, a first change amount of the ground surface temperature value of the target monitoring point within a target time period and a target time interval; determining a second mean value of a plurality of first change amounts corresponding to a plurality of target time intervals, and using the second mean value as a second fluctuation value of the ground surface temperature value in the second data set; According to the first product of the first fluctuation value and the second fluctuation value, an evaluation value of the solar radiation value at the first moment in the target time period is determined, including: according to the first data set, a third mean of multiple solar radiation values ​​of the target monitoring point in the target time period is determined; a second difference value between the solar radiation value and the third mean value is determined, and a third product of the first product value and the second difference value is determined; the correlation coefficient between the surface temperature value and the solar radiation value is obtained, and a first ratio of the third product value to the correlation coefficient is used as the evaluation value of the solar radiation value at the first moment in the target time period.

2. The method for simulating sunlight radiation intensity based on artificial intelligence according to claim 1, characterized in that: The correlation coefficient is obtained in the following way: Determine a first percentile of the solar radiation value at the first moment in the first data set, and determine a second percentile of the land surface temperature value at the first moment in the second data set; Determine a third difference value between a first percentile and a second percentile corresponding to the same first moment, and determine an average difference value of a plurality of third difference values ​​corresponding to a plurality of first moments; For a target moment in a target time period, determining a fourth difference value between a third difference value corresponding to the target moment and the average difference value; An exponential operation is performed on the fourth difference value using an exponential function to obtain a first exponential operation value, and the reciprocal of the first exponential operation value is used as the correlation coefficient at the target time.

3. The method for simulating sunlight radiation intensity based on artificial intelligence according to claim 1, characterized in that: According to the evaluation value of the solar radiation value, determining an abnormal solar radiation value and a non-abnormal solar radiation value from a plurality of solar radiation values ​​corresponding to a plurality of first moments, including: Determine an average evaluation value of the plurality of evaluation values, and perform an exponential operation on the average evaluation value using an exponential function to obtain a second exponential operation value; Determine the upper quartile and the lower quartile of a plurality of solar radiation values, and determine the abnormal upper limit value and the abnormal lower limit value of the box plot according to the reciprocal of the second index operation value, the upper quartile and the lower quartile, respectively; By using the abnormal upper limit value and the abnormal lower limit value, an abnormal solar radiation value and a non-abnormal solar radiation value are determined from a plurality of solar radiation values ​​corresponding to a plurality of first moments.

4. The method for simulating sunlight radiation intensity based on artificial intelligence according to claim 1, characterized in that: Correcting the abnormal solar radiation value according to the non-abnormal solar radiation value adjacent to the time corresponding to the abnormal solar radiation value to obtain a first solar radiation value includes: Obtaining a first average change in solar radiation values ​​at adjacent moments in the first data set, and determining a second average change in surface temperature values ​​at adjacent moments in the second data set; Determine a second ratio of the first average change to the second average change; determine a fifth difference between a ground surface temperature value at a third moment in the target time period and a ground surface temperature value at a moment adjacent to the third moment for a first abnormal radiation value among the multiple abnormal solar radiation values; A fourth product of the second ratio and the fifth difference value is obtained, and the fourth product is subtracted from a non-abnormal solar radiation value adjacent to a moment corresponding to the first abnormal radiation value to obtain a first solar radiation value.

5. The method for simulating sunlight radiation intensity based on artificial intelligence according to claim 1, characterized in that: The target solar radiation value at the second moment is obtained by performing simulation prediction according to the first solar radiation value and the non-abnormal solar radiation value, including: Inputting the first solar radiation value and the non-abnormal solar radiation value into a pre-trained prediction model to obtain a target solar radiation value at a second moment output by the prediction model; The prediction model is obtained by training a pre-built initial model using a training set and a test set, wherein the training set includes solar radiation values ​​in a first time period, and the test set includes solar radiation values ​​in a second time period, and the second time period is later than the first time period.

6. The method for simulating sunlight radiation intensity based on artificial intelligence according to any one of claims 1 to 5, characterized in that: The method further comprises: constructing a target data set including the first solar radiation value and the non-abnormal solar radiation value; Inputting the target data set into an animation generation model to obtain a target animation output by the animation generation model, wherein images at different moments in the target animation correspond to different solar radiation values; The target animation is sent to a terminal device so that the terminal device displays the target animation.

7. A sunlight radiation intensity simulation system based on artificial intelligence, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for simulating sunlight radiation intensity based on artificial intelligence according to any one of claims 1 to 6 is implemented.

Images