Data correction method and device, medium and electronic equipment
通过在高海拔光伏电站中使用预设模型库中的订正模型,对太阳辐照度数据进行订正处理,基于评估指标确定最佳模型,解决了高海拔区域模型预测数据订正精度不高的问题, Improved prediction accuracy.
Patent Information
- Application Number
- CN202411228667.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-12-03
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing data correction method has low accuracy in model prediction data in high altitude areas and cannot meet the model prediction needs of different high altitude areas.
By obtaining actual observation data and pattern prediction data of solar irradiance of high-altitude photovoltaic power stations, the pattern prediction data is corrected using the classic statistical correction model, random forest model, XGBOOST model or CU-Net model in the preset model library to correct the pattern prediction data Based on evaluation indicators such as correlation coefficient, root mean square error and average absolute error, the performance ranking of each model is obtained, the best solar irradiance correction model is determined, and dynamic optimization is achieved.
The overall accuracy of the numerical model prediction data in high-altitude areas was improved, the best model for high-altitude photovoltaic power station correlation was obtained, and dynamic optimization of different regions at high altitude was achieved.
Smart Images

Figure CN119067316A_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of meteorological forecasting and relates to a data correction method, and in particular to a data correction method, device, medium and electronic equipment. Background Art
[0002] Photovoltaic power generation is mainly affected by solar irradiance, and its changes are characterized by randomness, intermittency and volatility, showing high sensitivity to meteorological conditions. Especially under the conditions of increasing and intensifying extreme weather and climate, the uncertainty of climate-dependent power generation such as photovoltaics has brought huge challenges to energy scheduling and power stability. Improving the timeliness and accuracy of my country's photovoltaic resource forecasts has become a prominent issue that needs to be solved urgently to achieve multi-energy complementarity and ensure power safety and stability at present and in the future.
[0003] At present, the model prediction data of solar irradiance can generally be corrected using classical statistical correction models, random forest models and other models. However, the prediction of a single model cannot meet the correction requirements of model prediction data in different high-altitude areas. Therefore, the current data correction method has the problem of low accuracy in correcting model prediction data in high-altitude areas. Summary of the invention
[0004] The purpose of the present application is to provide a data correction method, device, medium and electronic device for solving the problem that the current data correction method has low accuracy in correcting model prediction data in high altitude areas.
[0005] In a first aspect, the present application provides a data correction method, comprising: obtaining actual observation data and model prediction data of the solar irradiance of a high-altitude photovoltaic power station; correcting the model prediction data through a preset model library to obtain corrected prediction data for each solar irradiance correction model in the preset model library; obtaining a first performance ranking of each solar irradiance correction model in the preset model library based on preset evaluation indicators, the corrected prediction data and the actual observation data; and determining the best solar irradiance correction model in the preset model library based on the first performance ranking.
[0006] In the data correction method, by obtaining the performance ranking of each solar irradiance correction model in the preset model library with respect to the preset evaluation index, the optimal model associated with the high-altitude photovoltaic power station can be obtained, thereby realizing dynamic optimization of different high-altitude areas and improving the overall accuracy of the numerical model prediction data for the overall high-altitude area.
[0007] In one embodiment of the present application, the solar irradiance correction model in the preset model library is a classical statistical correction model, a random forest model, an XGBOOST model or a CU-Net model.
[0008] In one embodiment of the present application, the preset evaluation indicators include correlation coefficient, root mean square error and mean absolute error.
[0009] In one embodiment of the present application, a method for obtaining the first performance ranking of each of the solar irradiance correction models in the preset model library based on the preset evaluation index, the corrected prediction data and the actual observation data includes: based on the corrected prediction data and the actual observation data, obtaining the second performance ranking of each of the solar irradiance correction models in the preset model library under the correlation coefficient, the root mean square error and the mean absolute error respectively; based on the second performance ranking, obtaining the comprehensive score of each of the solar irradiance correction models in the preset model library; based on the comprehensive score, obtaining the first performance ranking.
[0010] In one embodiment of the present application, the comprehensive score is expressed as:
[0011] S=w1s1+w2s2+w3s3
[0012] Where S represents the comprehensive score, w1 represents the weight associated with the correlation coefficient, s1 represents the score associated with the correlation coefficient ranking, w2 represents the weight associated with the root mean square error, s2 represents the score associated with the root mean square error ranking, w3 represents the weight associated with the mean absolute error, s3 represents the score associated with the mean absolute error ranking
[0013] In one embodiment of the present application, w2>w1>w3.
[0014] In one embodiment of the present application, the data correction method also includes: updating the preset model library to obtain an updated preset model library, and during the updating process of the preset model library, a new solar irradiance correction model is introduced into the preset model library; updating the preset evaluation indicators to obtain updated preset evaluations and indicators, and during the updating process of the preset evaluation indicators, new evaluation indicators are introduced into the preset evaluation indicators.
[0015] In the second aspect, the present application provides a data correction device, which includes: a data acquisition module for acquiring actual observation data and model prediction data of solar irradiance of a high-altitude photovoltaic power station; a correction processing module for performing correction processing on the model prediction data through a preset model library to obtain corrected prediction data of each solar irradiance correction model in the preset model library; a ranking acquisition module for acquiring a first performance ranking of each solar irradiance correction model in the preset model library based on preset evaluation indicators, the corrected prediction data and the actual observation data; a model determination module for determining the best solar irradiance correction model in the preset model library based on the first performance ranking.
[0016] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data correction method described in any one of the first aspects of the present application.
[0017] In a fourth aspect, the present application provides an electronic device, comprising: a memory storing a computer program; and a processor communicatively connected to the memory, for executing any data correction method described in the first aspect of the present application when the computer program is called.
[0018] As described above, the data correction method, device, medium and electronic device of the present application have the following beneficial effects:
[0019] In the data correction method, by obtaining the performance ranking of each solar irradiance correction model in the preset model library with respect to the preset evaluation index, the optimal model associated with the high-altitude photovoltaic power station can be obtained, thereby realizing dynamic optimization of different high-altitude areas and improving the overall accuracy of the numerical model prediction data for the overall high-altitude area. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 Shown is a schematic diagram of the hardware structure for running the data correction method according to an embodiment of the present application.
[0021] Figure 2 Shown is a flow chart of the data correction method of an embodiment of the present application.
[0022] Figure 3 Shown is a flowchart of an implementation method of the present application for obtaining the first performance ranking of each solar irradiance correction model in the preset model library based on the preset evaluation index, the corrected prediction data and the actual observation data.
[0023] Figure 4 Shown is a flow chart of an implementation method of the data correction method of an embodiment of the present application.
[0024] Figure 5 Shown is a schematic diagram of the structure of the data correction device described in an embodiment of the present application.
[0025] Component number description
[0026] 10 Electronic devices
[0027] 110 Memory
[0028] 120 processors
[0029] 1210 CPU
[0030] 1220 Neural Network Processor
[0031] 12210 Neural Network Implementation Engine
[0032] 12220 Dedicated Hardware Circuit
[0033] 122210 Matrix calculation unit
[0034] 122220 Vector Computing Unit
[0035] 500 Data Correction Device
[0036] 510 Data Acquisition Module
[0037] 520 Correction Processing Module
[0038] 530 Ranking Acquisition Module
[0039] 540 Model determination module
[0040] Steps S11-S14
[0041] Steps S21-S23
[0042] S31-S32 Steps DETAILED DESCRIPTION
[0043] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.
[0044] It should be noted that the illustrations provided in the following embodiments are only used to illustrate the basic concept of the present application in a schematic manner, and therefore the illustrations only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.
[0045] The technical solutions in the embodiments of the present application are described in detail below in conjunction with the drawings in the embodiments of the present application.
[0046] The data correction method provided in the embodiment of the present application can be executed in a computer electronic device. Figure 1 For example, Figure 1The hardware structure block diagram of the electronic device of the data correction method. The electronic device 10 includes a memory 110 and a processor 120. The processor 120 can be a central processing unit 1210 or a dedicated neural network processor 1220. The neural network processor 1220 includes a neural network implementation engine 12210 and a dedicated hardware circuit 12220. The dedicated hardware circuit 12220 includes a matrix calculation unit 122210 and a vector calculation unit 122220.
[0047] Optionally, the neural network processor 1220 is a processor that uses a dedicated hardware circuit 12220 to perform neural network calculations, and the dedicated hardware circuit 12220 is an integrated circuit for performing neural network calculations and includes a matrix calculation unit 122210 and a vector calculation unit 122220 for hardware execution of vector-matrix multiplication.
[0048] Optionally, the neural network implementation engine 12210 is used to generate instructions for execution by the dedicated hardware circuit 12220. When the instructions are executed by the dedicated hardware circuit 12220, the dedicated hardware circuit 12220 performs operations specified by the neural network to generate a neural network output from the received neural network input.
[0049] like Figure 2 As shown, this embodiment provides a data correction method, which can be executed by a CPU of a computer electronic device, and the data correction method includes:
[0050] S11, obtaining actual observation data and model prediction data of solar irradiance of high-altitude photovoltaic power stations.
[0051] Optionally, the high altitude may refer to an altitude of 1500-3500 meters. The actual observation data may refer to solar irradiance data actually measured in the high sea wave photovoltaic power station, and the model prediction data may refer to output data of a numerical model.
[0052] Optionally, the high-altitude photovoltaic power station may refer to a photovoltaic power station in a high-altitude complex terrain, and the complex terrain may refer to a terrain with more landform types, greater terrain undulations and / or greater slope changes in the high-altitude area.
[0053] S12, performing correction processing on the model prediction data through a preset model library to obtain the corrected prediction data of each solar irradiance correction model in the preset model library.
[0054] Optionally, the preset model library includes a classical statistical correction model, a random forest model, an XGBOOST model and a CU-Net model. The correction process may refer to each model in the preset model library using historical meteorological information to calculate the systematic errors of the multi-year climate average numerical model for correction, and the historical meteorological information refers to the historical actual observation data and the historical model prediction data. The corrected prediction data may refer to the data after the model prediction data is corrected.
[0055] Alternatively, classical statistical correction methods improve prediction accuracy by analyzing and correcting errors and deviations in the data. Common statistical correction methods include mean correction, variance correction, and regression correction. The core idea is to use statistical models to identify and adjust systematic errors in the data, thereby improving the accuracy of the prediction results. Its advantages are: simple and intuitive, easy to understand and implement; suitable for small-scale data sets; high computational efficiency. Disadvantages are: dependent on assumptions, may not be applicable to complex data distributions; limited prediction ability, and inability to handle nonlinear relationships.
[0056] Alternatively, random forest is an ensemble learning method based on decision trees, which improves prediction performance by building multiple decision trees and combining their output results. Each tree is trained on a random subset of the sample data, and the final prediction result is the average or vote of all tree results. Its advantages are: insensitive to data noise, good resistance to overfitting, can handle high-dimensional data and identify important features, and is suitable for classification and regression problems. Disadvantages are: relatively long training and prediction time, high model complexity, and poor interpretability.
[0057] Optionally, XGBOOST is an ensemble learning method based on gradient boosting, which improves prediction accuracy by gradually building and optimizing decision trees. It uses quadratic Taylor expansion and regularization techniques in the optimization process, and has strong generalization ability and computational efficiency. Advantages: excellent performance when processing large-scale data sets, strong prediction ability, ability to handle nonlinear relationships, high flexibility, and support for multiple custom loss functions. Disadvantages: more parameters, complex parameter adjustment, and longer training time, especially on large data sets.
[0058] Alternatively, CU-Net is a deep learning model based on convolutional neural networks (CNNs), focusing on processing spatiotemporal data such as images and videos. CU-Net extracts features through convolutional layers and achieves high-resolution output through a U-shaped structure (encoder-decoder), and is widely used in fields such as image segmentation and object detection. Its advantages are: it can automatically extract multi-level features, is suitable for complex spatiotemporal data, performs well in image segmentation and object detection tasks, can process large-scale data, and has good scalability. Disadvantages are: it requires a large amount of labeled data for training, the training and inference time is long, it requires high computing resources, has high model complexity, and has poor interpretability.
[0059] S13, based on the preset evaluation index, the corrected prediction data and the actual observation data, obtaining the first performance ranking of each of the solar irradiance correction models in the preset model library.
[0060] Optionally, the preset evaluation indicators include correlation coefficient, root mean square error and mean absolute error.
[0061] Optionally, the first performance ranking may refer to a comprehensive ranking of models in the preset model library with respect to the preset evaluation index.
[0062] Optionally, the correlation coefficient is a statistical indicator used to measure the strength and direction of the linear relationship between two variables, and its value range is between -1 and 1, and is used to indicate the degree of relationship between the variables.
[0063]
[0064] Where n is the number of samples, x i and i are the i-th values of the two variables, and are the means of the two variables. R approaches 1, indicating a positive correlation, and there is a positive linear relationship between the two variables. R approaches -1, indicating a negative correlation, and there is a negative linear relationship between the two variables. R approaches 0, indicating no correlation, and there is no linear relationship between the two variables.
[0065] Optionally, the root mean square error is a standard deviation used to measure the difference between the model prediction value and the actual observation value. It reflects the degree of deviation between the predicted value and the actual value. The smaller the value, the higher the accuracy of the model prediction.
[0066]
[0067] Where n is the number of samples, y i is the actual observation data, It is the corrected forecast data. RMSE is sensitive to large errors because the errors are amplified by square. It is often used in regression analysis as a model performance evaluation indicator.
[0068] Optionally, the mean absolute error is expressed as:
[0069]
[0070] Where n is the number of samples, y i is the actual observation data, It is the corrected forecast data. The positive or negative value of MAE can indicate whether the forecast value is systematically overestimated or underestimated. MAE>0: Indicates that the model as a whole overestimates the actual value. MAE<0: Indicates that the model as a whole underestimates the actual value. It is often used to verify the accuracy of the model and calibrate the model.
[0071] S14: Determine the best solar radiation correction model in the preset model library based on the first performance ranking.
[0072] According to the above description, the data correction method described in this embodiment includes: obtaining actual observation data and model prediction data of the solar irradiance of a high-altitude photovoltaic power station; correcting the model prediction data through a preset model library to obtain corrected prediction data for each solar irradiance correction model in the preset model library; obtaining a first performance ranking of each solar irradiance correction model in the preset model library based on preset evaluation indicators, the corrected prediction data and the actual observation data; and determining the best solar irradiance correction model in the preset model library based on the first performance ranking.
[0073] In the data correction method, by obtaining the performance ranking of each solar irradiance correction model in the preset model library with respect to the preset evaluation index, the optimal model associated with the high-altitude photovoltaic power station can be obtained, thereby realizing dynamic optimization of different high-altitude areas and improving the overall accuracy of the numerical model prediction data for the overall high-altitude area.
[0074] See also Figure 3 ,like Figure 3 As shown, this embodiment provides a method for obtaining a first performance ranking of each solar irradiance correction model in the preset model library based on a preset evaluation index, the corrected prediction data and the actual observation data, including:
[0075] S21, based on the corrected prediction data and the actual observation data, obtaining the second performance ranking of each of the solar irradiance correction models in the preset model library under the correlation coefficient, the root mean square error and the mean absolute error respectively.
[0076] Optionally, the second performance ranking may refer to the performance ranking of each model in the preset model library under each preset evaluation indicator.
[0077] S22: Based on the second performance ranking, obtain the comprehensive score of each solar irradiance correction model in the preset model library.
[0078] Optionally, based on the second performance ranking, an implementation method for obtaining a comprehensive score of each model in the preset model library includes: obtaining a score associated with the second performance ranking based on the second performance ranking; and obtaining a comprehensive score of each model in the preset model library based on the score associated with the second performance ranking.
[0079] Optionally, the score associated with the second performance ranking can be flexibly set according to actual conditions, and this embodiment does not explicitly limit this. For example, under the correlation coefficient indicator, the score associated with the first place in the preset model library can be set to 100, the score associated with the second place can be set to 80, the score associated with the third place can be set to 60, and so on.
[0080] S23: Based on the comprehensive score, obtain the first performance ranking.
[0081] Optionally, the comprehensive score is expressed as:.
[0082] S=w1s1+w2s2+w3s3
[0083] Wherein S represents the comprehensive score, w1 represents the weight associated with the correlation coefficient, s1 represents the score associated with the correlation coefficient ranking, w2 represents the weight associated with the root mean square error, s2 represents the score associated with the root mean square error ranking, w3 represents the weight associated with the mean absolute error, and s3 represents the score associated with the mean absolute error ranking.
[0084] Optionally, the weight associated with each indicator in the preset evaluation indicators can be flexibly set according to actual conditions. In actual scenarios, if more emphasis is placed on a certain indicator, the weight of the indicator can be set relatively high, which will not be elaborated in this embodiment.
[0085] Optionally, w2>w1>w3, and this ranking indicates that the importance of the root mean square error indicator is greater than the importance of the correlation coefficient indicator, and the importance of the correlation coefficient indicator is greater than the importance of the mean absolute error indicator.
[0086] See also Figure 4 ,like Figure 4 As shown, this embodiment provides a data correction method, including:
[0087] S31, updating the preset model library to obtain an updated preset model library, during which a new solar irradiance correction model is introduced into the preset model library.
[0088] Optionally, by introducing a new solar irradiance correction model into the preset model library, the correction accuracy of the model prediction data in high altitude areas can be further improved. For example, the LSTM model can also be introduced into the preset model library.
[0089] S32, updating the preset evaluation index to obtain updated preset evaluation and index, during which the new evaluation index is introduced into the preset evaluation index.
[0090] The protection scope of the data correction method described in the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing or replacing steps in the prior art based on the principles of the present application are included in the protection scope of the present application.
[0091] See also Figure 5 ,like Figure 5 As shown, this embodiment provides a data correction device 500, and the data correction device 500 includes:
[0092] The data acquisition module 510 is used to acquire actual observation data and model prediction data of solar irradiance of a high-altitude photovoltaic power station.
[0093] The correction processing module 520 is used to perform correction processing on the model prediction data through the preset model library to obtain the corrected prediction data of each solar irradiance correction model in the preset model library.
[0094] The ranking acquisition module 530 is used to obtain the first performance ranking of each solar irradiance correction model in the preset model library based on the preset evaluation index, the corrected prediction data and the actual observation data.
[0095] The model determination module 540 is used to determine the best solar irradiance correction model in the preset model library based on the first performance ranking.
[0096] In the data correction device 500 provided in this embodiment, the data acquisition module 510 and Figure 2 The step S11 of the data correction method shown in FIG. 1 corresponds to the step S12 of the data correction method shown in FIG. 1 . The correction processing module 520 corresponds to the step S11 of the data correction method shown in FIG. Figure 2 The ranking acquisition module 530 corresponds to step S12 of the data correction method shown in FIG. Figure 2 The step S13 of the data correction method shown in FIG. 1 corresponds to the step S13 of the data correction method shown in FIG. 1 . The model determination module 540 corresponds to the step S13 of the data correction method shown in FIG. Figure 2There is a one-to-one correspondence with step S14 of the data correction method shown.
[0097] In the several embodiments provided in the present application, it should be understood that the disclosed device or method can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules / units is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.
[0098] The modules / units described as separate components may or may not be physically separated, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, the functional modules / units in the various embodiments of the present application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.
[0099] Those of ordinary skill in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0100] This embodiment provides an electronic device, the electronic device comprising a memory storing a computer program; a processor connected to the memory for communication, and executing the computer program when the computer program is called. Figure 2 The data correction method shown.
[0101] The present application embodiment also provides a computer-readable storage medium. A person of ordinary skill in the art can understand that all or part of the steps in the method for implementing the above embodiment can be completed by a program to instruct the processor, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state hard disk, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (DVD)), or a semiconductor medium (e.g., a solid-state hard disk (SSD)), etc.
[0102] The present application embodiment may also provide a computer program product, the computer program product including one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer or data center.
[0103] When the computer program product is executed by a computer, the computer executes the method described in the above method embodiment. The computer program product may be a software installation package, and when the above method is required, the computer program product may be downloaded and executed on a computer.
[0104] The descriptions of the processes or structures corresponding to the above-mentioned figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.
[0105] The above embodiments are merely illustrative of the principles and effects of the present application and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.
Claims
1. A data correction method, It is characterized in that include: Obtain actual observation data and model prediction data of solar irradiance for high-altitude photovoltaic power plants; Correcting the model prediction data through a preset model library to obtain corrected prediction data of each solar irradiance correction model in the preset model library; Based on the preset evaluation index, the corrected prediction data and the actual observation data, obtaining the first performance ranking of each of the solar irradiance correction models in the preset model library; An optimal solar irradiance correction model in the preset model library is determined based on the first performance ranking.
2. The data correction method according to claim 1, It is characterized in that The solar irradiance correction model in the preset model library is a classic statistical correction model, a random forest model, an XGBOOST model or a CU-Net model.
3. The data correction method according to claim 2, It is characterized in that The preset evaluation indicators include correlation coefficient, root mean square error and mean absolute error.
4. The data correction method according to claim 3, It is characterized in that The method for obtaining the first performance ranking of each of the solar irradiance correction models in the preset model library based on the preset evaluation index, the corrected prediction data and the actual observation data includes: Based on the corrected prediction data and the actual observation data, obtaining a second performance ranking of each of the solar irradiance correction models in the preset model library under the correlation coefficient, the root mean square error and the mean absolute error respectively; Based on the second performance ranking, obtaining a comprehensive score of each of the solar irradiance correction models in the preset model library; Based on the comprehensive score, the first performance ranking is obtained.
5. The data correction method according to claim 4, It is characterized in that The composite score is expressed as: S=w 1 s 1 +w 2 s 2 +w 3 s 3 Where S represents the comprehensive score, w 1 represents the weight associated with the correlation coefficient, s 1 represents the score of the correlation coefficient ranking association, w 2 represents the weight associated with the RMS error, s 2 represents the score of the root mean square error ranking association, w 3 represents the weight associated with the mean absolute error, s 3 represents the score of the mean absolute error ranking association.
6. The data correction method according to claim 5, It is characterized in that In 2 >in 1 >in 3 。 7. The data correction method according to claim 1, It is characterized in that The data correction method further comprises: The preset model library is updated to obtain an updated preset model library, wherein a new solar irradiance correction model is introduced into the preset model library during the updating process of the preset model library; The preset evaluation index is updated to obtain updated preset evaluation and index, and the new evaluation index is introduced into the preset evaluation index during the updating process of the preset evaluation index.
8. A data correction device, It is characterized in that The data correction device comprises: A data acquisition module is used to obtain actual observation data and model prediction data of solar irradiance of high-altitude photovoltaic power stations; A correction processing module, used for performing correction processing on the model prediction data through a preset model library to obtain the corrected prediction data of each solar irradiance correction model in the preset model library; A ranking acquisition module, used for acquiring a first performance ranking of each of the solar irradiance correction models in the preset model library based on a preset evaluation index, the corrected prediction data and the actual observation data; A model determination module is used to determine the best solar irradiance correction model in the preset model library based on the first performance ranking.
9. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the data correction method described in any one of claims 1 to 7 is implemented.
10. An electronic device, It is characterized in that The electronic device comprises: A memory storing a computer program; A processor is communicatively connected to the memory and executes the data correction method according to any one of claims 1 to 7 when calling the computer program.
Citation Information
Patent Citations
Solar radiation prediction correction method and system based on BP neural network
CN115545338A
Irradiance correction method and device for photovoltaic power generation prediction
CN117909432A
