Photovoltaic power generation power prediction method and device, computer equipment and storage medium

Abstract

The invention relates to a photovoltaic power generation power prediction method and device, computer equipment and a storage medium. The method comprises the following steps: constructing a photovoltaic power generation power prediction model based on an obtained first assembly parameter of the photovoltaic assembly and an obtained second assembly parameter of the inverter; acquiring a historical meteorological data set of a target area, and performing model training and parameter calibration on the photovoltaic power generation power prediction model based on the historical meteorological data set to obtain a completely trained photovoltaic power generation power prediction model; and obtaining an original meteorological data set of the current time period, and outputting and obtaining a power prediction result of the photovoltaic power generation system through the completely trained photovoltaic power generation power prediction model. By adopting the method, the construction of the photovoltaic power generation power prediction model with high mobility and high interpretability can be realized, the online learning capability is realized, and the power prediction accuracy of the photovoltaic power generation system is continuously improved.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic power generation technology, and in particular to a photovoltaic power generation power prediction method, device, computer equipment, and storage medium. Background Art

[0002] With the development of photovoltaic power generation technology, photovoltaic power generation prediction technology is becoming increasingly important in practical applications. Photovoltaic power generation prediction technology refers to the use of technical means to predict the power generation of photovoltaic power generation systems within a certain period of time in the future.

[0003] Existing photovoltaic power generation predictions are mostly implemented using big data artificial intelligence algorithms. However, due to the lack of online learning capabilities, existing photovoltaic power generation prediction systems require manual selection of prediction scenarios to train models for specific scenarios, resulting in poor interpretability and versatility. Summary of the Invention

[0004] Based on this, it is necessary to provide a photovoltaic power prediction method, device, computer equipment and storage medium that has strong interpretability and versatility, and has automatic calibration and online learning capabilities to address the above technical problems.

[0005] In a first aspect, the present application provides a photovoltaic power generation power prediction method, which is applied to a photovoltaic power generation system, wherein the photovoltaic power generation system includes a photovoltaic module and an inverter, and the method includes:

[0006] Constructing a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter;

[0007] Acquiring a historical meteorological data set of a target area, and performing model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model;

[0008] The original meteorological data set of the current time period is obtained, and the power prediction result of the photovoltaic power generation system is output through the trained photovoltaic power generation power prediction model.

[0009] In one embodiment, the performing model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological dataset includes:

[0010] Reconstructing the resolution of the historical meteorological dataset to obtain a high-precision historical meteorological dataset;

[0011] The high-precision historical meteorological data is locally revised to obtain a revised historical meteorological data set;

[0012] Training the photovoltaic power generation model according to the revised historical meteorological data set, the first component parameters, and the second component parameters, and outputting a training power prediction value of the photovoltaic power generation system;

[0013] Obtaining an actual power prediction value of the photovoltaic power generation system, and calculating a model parameter correction value of the photovoltaic power generation power prediction model based on the actual power prediction value and the training power prediction value;

[0014] The photovoltaic power generation prediction model is parameter-corrected according to the model parameter correction value to obtain the fully trained photovoltaic power generation prediction model.

[0015] In one embodiment, the historical meteorological dataset is stored in a data grid format, and reconstructing the resolution of the historical meteorological dataset to obtain a high-precision historical meteorological dataset includes:

[0016] Extracting low-frequency information of the data grid to obtain shallow features of the historical meteorological dataset;

[0017] Performing long-distance connections and convolution operations on the shallow features to obtain deep features of the historical meteorological dataset;

[0018] The shallow features and the deep features are fused and upsampled to obtain the high-precision historical meteorological dataset.

[0019] In one embodiment, the locally revising the high-precision historical meteorological dataset to obtain a revised historical meteorological dataset includes:

[0020] Obtaining a deviation correction value of the meteorological data at the same time on the previous day, an actual value of the meteorological data at the same time on the previous day, and a forecast value of the meteorological data at the same time on the previous day, and calculating a deviation correction value for the same time on the current day in the high-precision historical meteorological data;

[0021] The deviation correction value at the same time on the same day is superimposed with the high-precision historical meteorological data at the same time on the same day to obtain the corrected historical meteorological data.

[0022] In one embodiment, the training power prediction value includes a DC power training prediction value and an AC power training prediction value, and the photovoltaic power generation power model is trained based on the revised historical meteorological data, the first component parameters, and the second component parameters, and the training power prediction value of the photovoltaic power generation system obtained by outputting includes:

[0023] Calculating a DC power training prediction value of the photovoltaic module based on the revised historical meteorological data and the first module parameter;

[0024] The AC power training prediction value of the inverter is calculated based on the revised historical meteorological data, the second component parameters, and the DC power training prediction value of the photovoltaic component.

[0025] In one embodiment, the model parameter correction value includes a DC parameter correction value and an AC parameter correction value, and obtaining the actual power prediction value of the photovoltaic power generation system and calculating the model parameter correction value of the photovoltaic power generation power prediction model based on the actual power prediction value and the training power prediction value includes:

[0026] Obtaining an actual value of the DC power of the photovoltaic module and an actual value of the AC power of the inverter;

[0027] Determining a DC parameter correction value of the photovoltaic power generation prediction model according to the DC power actual value and the DC power training prediction value;

[0028] An AC parameter correction value of the photovoltaic power generation power prediction model is determined according to the AC power actual value and the AC power training prediction value.

[0029] In a second aspect, the present application further provides a photovoltaic power generation prediction device, characterized in that the device comprises:

[0030] A model building module, configured to build a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the acquired second component parameters of the inverter;

[0031] A model training module is used to obtain a historical meteorological data set of a target area, and perform model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model;

[0032] The power prediction module is used to obtain the original meteorological data set of the current time period, and output the power prediction result of the photovoltaic power generation system through the trained photovoltaic power generation power prediction model.

[0033] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the content of the first aspect described above when executing the computer program.

[0034] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which implements the content of the first aspect described above when the computer program is executed by a processor.

[0035] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements the content of the first aspect mentioned above when executed by a processor.

[0036] The above-mentioned photovoltaic power generation power prediction method, device, computer equipment and storage medium construct a photovoltaic power generation power prediction model based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter; obtain a historical meteorological data set of the target area, and perform model training and parameter calibration on the photovoltaic power generation power prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation power prediction model; obtain the original meteorological data set of the current time period, and output the power prediction result of the photovoltaic power generation system through the trained photovoltaic power generation power prediction model, thereby realizing the construction of a photovoltaic power generation power prediction model with strong portability and strong interpretability, and having online learning capabilities, thereby continuously improving the power prediction accuracy of the photovoltaic power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0038] Figure 1 A diagram showing an application environment of a photovoltaic power generation power prediction method in one embodiment;

[0039] Figure 2 1 is a flow chart of a photovoltaic power generation power prediction method according to an embodiment;

[0040] Figure 3 202 is a flow chart of step 202 in one embodiment;

[0041] Figure 4 301 is a flow chart of step 301 in one embodiment;

[0042] Figure 5 302 is a flowchart of an embodiment;

[0043] Figure 6 203 is a flow chart of step 203 in one embodiment;

[0044] Figure 7 is a structural block diagram of a photovoltaic power generation prediction device in one embodiment;

[0045] Figure 8 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0047] Unless otherwise defined, technical or scientific terms used herein shall have the ordinary meaning as understood by persons of ordinary skill in the art to which this application belongs. The terms "a," "an," "an," "the," and similar expressions used herein do not denote quantitative limitations and may refer to either the singular or the plural. The terms "comprise," "include," "have," and any variations thereof, used herein, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or modules (units) is not limited to the listed steps or units but may also include steps or units not listed, or may include other steps or units inherent to the process, method, product, or apparatus. The terms "connected," "connected," "coupled," and similar expressions used herein are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. As used herein, "plurality" means two or more. "And / or" describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" may mean: A exists alone; A and B exist simultaneously; or B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0048] The photovoltaic power generation power prediction method provided in the embodiment of the present application can be applied to Figure 1 In the application scenario shown, the photovoltaic power generation system includes photovoltaic modules, an inverter, and a controller. The controller can be a terminal 102, which communicates with a server 104 via a network. The data storage system can store data that the server 104 needs to process. The data storage system can be integrated with the server 104 or placed in the cloud or other network servers.

[0049] On server 104, a photovoltaic power generation prediction model is constructed based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter; a historical meteorological data set of the target area is obtained, and model training and parameter calibration of the photovoltaic power generation prediction model are performed based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model; an original meteorological data set of the current time period is obtained, and the power prediction result of the photovoltaic power generation system is output through the trained photovoltaic power generation prediction model.

[0050] The terminal 102 may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, etc. The server 104 may be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0051] In an exemplary embodiment, Figure 2 As shown, a photovoltaic power generation power prediction method is provided, which is applied to Figure 1 The server side in the example is used to illustrate, including the following steps 201 to 203. Among them:

[0052] Step 201: construct a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the acquired second component parameters of the inverter.

[0053] The data storage system on the server includes a photovoltaic module database and an inverter database. Based on the photovoltaic power generation system in the target area, first component parameters corresponding to the photovoltaic modules and second component parameters corresponding to the inverter are obtained from the data storage system. A first simulation model of the photovoltaic modules is constructed based on the first component parameters, and a second simulation model of the inverter is constructed based on the second component parameters. The photovoltaic power generation prediction model includes the first simulation model and the second simulation model.

[0054] Step 202 : Acquire a historical meteorological dataset of the target area, perform model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological dataset, and obtain a fully trained photovoltaic power generation prediction model.

[0055] Specifically, using numerical models to forecast the Global Forecast System (GFS) allows for the acquisition of global meteorological data. This data encompasses multiple meteorological elements, including surface radiation, temperature, and wind speed. These elements significantly impact photovoltaic power generation, so it is necessary to obtain a historical meteorological dataset for the target region. This dataset is then used to train and calibrate the first and second simulation models in the photovoltaic power generation prediction model, resulting in a fully trained photovoltaic power generation prediction model.

[0056] Preferably, the photovoltaic power generation prediction model includes an error compensation module, which can perform local correction on meteorological data and calibrate model parameters of the first simulation model and the second simulation model during the model training process.

[0057] Step 203 , obtaining the original meteorological data set of the current time period, and outputting the power prediction result of the photovoltaic power generation system through the trained photovoltaic power generation power prediction model.

[0058] Specifically, when it is necessary to predict the power of the photovoltaic power generation system in the current time period, the original meteorological data set of the current time period is obtained, and a complete photovoltaic power generation power prediction model is trained to output a high-precision power prediction result of the photovoltaic power generation system.

[0059] Furthermore, after obtaining the power prediction result, the actual power value of the current time period is obtained. According to the actual power value, the error compensation module is used to calibrate the parameters of the trained photovoltaic power prediction model based on the actual power value to realize real-time online learning of the model.

[0060] In the above-mentioned photovoltaic power generation prediction method, a photovoltaic power generation prediction model is constructed based on the obtained first component parameters of the photovoltaic component and the second component parameters of the inverter; a historical meteorological data set of the target area is obtained, and the photovoltaic power generation power prediction model is trained and parameter calibrated based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model; the original meteorological data set of the current time period is obtained, and the power prediction result of the photovoltaic power generation system is output through the trained photovoltaic power generation power prediction model, thereby realizing the construction of a photovoltaic power generation power prediction model with strong portability and strong interpretability, and having online learning capabilities, and continuously improving the power prediction accuracy of the photovoltaic power generation system.

[0061] In an exemplary embodiment, Figure 3 As shown, step 202 performs model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set, specifically including steps 301 to 305. Among them:

[0062] Step 301 : reconstruct the resolution of the historical meteorological dataset to obtain a high-precision historical meteorological dataset.

[0063] Specifically, a SwinIR model was constructed and trained to reconstruct the resolution of the historical meteorological dataset to obtain high-precision historical meteorological data. The resolution of the data in the historical meteorological dataset was 0.5°, and the resolution of the data in the high-precision historical meteorological dataset was 0.1°.

[0064] Step 302: Localize and revise the high-precision historical meteorological dataset to obtain a revised historical meteorological dataset.

[0065] Specifically, the error compensation module utilizes an adaptive Kalman filtering method to perform localized revision on the high-precision historical meteorological dataset to obtain the revised historical meteorological dataset.

[0066] Step 303: Training the photovoltaic power generation model based on the revised historical meteorological data set, the first component parameters, and the second component parameters, and outputting a training power prediction value of the photovoltaic power generation system.

[0067] Specifically, the revised historical meteorological dataset is input into the photovoltaic power generation model. Based on the revised historical meteorological dataset and first component parameters, the first simulation model is trained by simulating photovoltaic component performance, and a first trained prediction value for the photovoltaic component is output. Based on the revised historical meteorological dataset and second component parameters, the second simulation model is trained by simulating inverter performance, and a second trained prediction value for the inverter is output. The trained power prediction value for the photovoltaic power generation system includes the first trained prediction value and the second trained prediction value.

[0068] Step 304 : obtaining the actual power prediction value of the photovoltaic power generation system, and calculating the model parameter correction value of the photovoltaic power generation power prediction model according to the actual power prediction value and the training power prediction value.

[0069] Specifically, the error compensation module is used to obtain the actual power prediction value of the photovoltaic power generation system, and the model parameter correction value of the photovoltaic power generation power prediction model is calculated based on the actual power prediction value and the training power prediction value.

[0070] Step 305 : calibrate the parameters of the photovoltaic power generation prediction model according to the model parameter correction values ​​to obtain the fully trained photovoltaic power generation prediction model.

[0071] In this embodiment, the algorithm, based on the Global Weather Forecast (GFS) data model and resolution reconstruction technology, can be rapidly deployed to any region of the world for forecasting. A photovoltaic power prediction model is constructed based on the power generation principles and processes of photovoltaic modules and inverters. By correcting meteorological data and training and revising model parameters, high-precision training of the photovoltaic power prediction model is achieved. The model is highly portable and efficiently deployed.

[0072] In one embodiment, the historical meteorological data set is stored in the form of a data grid, such as Figure 4As shown, step 301 reconstructs the resolution of the historical meteorological data set to obtain high-precision historical meteorological data, which specifically includes the following steps 401 to 403.

[0073] Step 401: extract low-frequency information of the data grid to obtain shallow features of the historical meteorological dataset.

[0074] Step 402: Perform long-distance connection and convolution operations on the shallow features to obtain deep features of the historical meteorological dataset.

[0075] Step 403: Up-sampling the shallow features and the deep features to obtain the high-precision historical meteorological dataset.

[0076] Specifically, the main structure of the SwinIR model includes three parts, namely the shallow feature extraction layer, the deep feature extraction layer and the high-quality grid reconstruction layer. The historical meteorological dataset in the form of a data grid is input into the SwinIR model, and a 3*3 convolution kernel of the shallow feature extraction layer is used to extract the low-frequency information of each area in the data grid, thereby obtaining the shallow features of the historical meteorological dataset. The low-frequency information includes the edges and gradients of different meteorological elements. Then, the shallow features are connected over long distances through the residual network of the deep feature extraction layer, and the data results are subjected to a 3*3 convolution operation to obtain the deep features of the historical meteorological dataset. Finally, the shallow features and the deep features are passed through the high-quality grid reconstruction layer, and the sub-pixel convolution layer is used to fuse and upsample the two features to obtain the high-precision historical meteorological dataset.

[0077] In this embodiment, high-precision meteorological data is obtained by reconstructing the resolution, extracting features, and fusing and upsampling the meteorological data, thereby improving the accuracy of the photovoltaic power prediction model in power prediction based on the meteorological data.

[0078] In one embodiment, Figure 5 As shown, step 302 performs localized revision on the high-precision historical meteorological dataset to obtain a revised historical meteorological dataset, which specifically includes the following steps 501 to 502.

[0079] Step 501, obtain the deviation correction value of the meteorological data at the same time of the previous day, the actual value of the meteorological data at the same time of the previous day, and the forecast value of the meteorological data at the same time of the previous day, and calculate the deviation correction value of the high-precision historical meteorological data at the same time of the day.

[0080] Step 502: Superimpose the deviation correction value at the same time on the same day with the high-precision historical meteorological data at the same time on the same day to obtain the corrected historical meteorological data.

[0081] Specifically, the calculation formula for the deviation correction value of the revised meteorological data is as follows:

[0082]

[0083] In the formula Indicates the deviation correction value at a certain time of the day. Indicates the deviation correction value at the same time of the previous day. represents the weight coefficient, Indicates the actual value at that time the previous day. Indicates correspondence The predicted value of .

[0084] The implementation method is as follows: Step 1: When When implementing a "cold start", that is, Step 2: Given appropriate weights, calculate the lagged deviation correction value according to the formula Repeat step 2. After a period of iterative accumulation of historical meteorological data (typically one to two months), the resulting error stabilizes and, to a certain extent, characterizes the system error. The core algorithm is plug-and-play, and through cold-start data accumulation and online learning, it continuously improves forecast accuracy, enhancing the precision of meteorological data forecasts for the target area.

[0085] In an exemplary embodiment, the training power prediction value includes a DC power training prediction value and an AC power training prediction value, such as Figure 6 As shown, step 203 trains the photovoltaic power generation model based on the corrected historical meteorological data, the first component parameters, and the second component parameters, and outputs the training power prediction value of the photovoltaic power generation system, which specifically includes the following steps 601 to 603.

[0086] Step 601: Acquire the actual value of the DC power of the photovoltaic module and the actual value of the AC power of the inverter.

[0087] Step 602: Determine a DC parameter correction value of the photovoltaic power generation prediction model according to the actual DC power value and the trained DC power prediction value.

[0088] Specifically, the DC power training prediction value at time t The calculation formula is as follows:

[0089]

[0090] in, Indicates time The characteristic vector of the input meteorological data, such as temperature, surface radiation value, etc. Indicates the diode factor, resistance, bias residual and other parameters of the photovoltaic module.

[0091] The DC power actual value at time t is subtracted from the DC power training prediction value to obtain the DC power prediction error of the photovoltaic power generation power prediction model. The root mean square error of the first simulation model is calculated based on the DC power prediction error, and the root mean square error is used as the loss function of the first simulation model. According to the loss function of the first simulation model , use the following formula to calculate the DC parameter correction value of the first simulation model at time t+1 :

[0092] ;

[0093] in, 1 represents the learning rate of the first simulation model; is the DC power loss function Gradients of the first simulation model parameters, Represents the DC parameter correction value at time t. The gradient calculation formula is: ,in, , Indicates time The DC power prediction value, It is at the moment The actual value of DC power.

[0094] Step 603 : determining an AC parameter correction value of the photovoltaic power generation prediction model according to the AC power actual value and the AC power training prediction value.

[0095] Specifically, the AC power training prediction value at time t is The calculation formula is as follows:

[0096] ;

[0097] in, Indicates parameters such as the inverter's conversion efficiency.

[0098] The AC power actual value at time t is subtracted from the AC power training prediction value to obtain the AC power prediction error of the photovoltaic power generation power prediction model. The root mean square error of the second simulation model is calculated based on the AC power prediction error, and the root mean square error is used as the loss function of the second simulation model. According to the loss function of the first simulation model , use the following formula to calculate the AC parameter correction value of the first simulation model at time t+1 :

[0099] ;

[0100] in, 2 represents the learning rate of the first simulation model; is the AC power loss function Gradients of the first simulation model parameters, Indicates the AC parameter correction value at time t.

[0101] The specific calculation formula of the gradient is: ,in, , Indicates time The predicted value of AC power, Indicates at time The actual value of AC power.

[0102] In this embodiment, the error compensation module is used to correct the model parameters of the first simulation model and the second simulation model according to the actual power value, thereby improving the power prediction accuracy of the photovoltaic power generation prediction model. The system power generation power will be affected by many factors, such as individual differences in equipment, component aging, regional meteorological data forecast deviations, etc. The error compensation module can adjust the model parameters through real-time online learning to maintain the accuracy of the model and improve the adaptability of the model.

[0103] In a preferred embodiment, a photovoltaic power generation power prediction method is provided for a photovoltaic power generation system. The photovoltaic power generation power prediction model includes a first simulation model of a photovoltaic module, namely a SAPM model, a second simulation model of an inverter, namely a Sandia Inverter Model model, and an error compensation module. The error compensation module includes a first compensation model for localizing meteorological data and a second compensation model for correcting model parameters, specifically including the following:

[0104] Step 1: Use the numerical model forecast GFS to obtain the meteorological dataset of the photovoltaic power generation system in the target area within the set historical time period.

[0105] Step 2: Use the SwinIR model to reconstruct the resolution of the meteorological dataset to obtain a high-precision meteorological dataset. Use the first compensation model to perform adaptive Kalman filtering to perform localized revisions on the high-precision meteorological dataset to obtain a revised meteorological dataset.

[0106] Step 3: Input the revised meteorological data set into the SAPM model of the photovoltaic power generation prediction model. By sequentially calculating the photocurrent, saturation current, diode factor, parallel resistance, etc., the predicted output current of the photovoltaic module is obtained.

[0107] Photocurrent calculation formula: ;

[0108] Saturation current calculation formula: ;

[0109] Diode factor calculation formula: ;

[0110] Series resistance calculation formula: ;

[0111] Parallel resistance calculation formula: ;

[0112] Current-voltage relationship: ;

[0113] In the above formula, is the output current, is the photocurrent, is the reference photogenerated current; is the irradiance (unit: W / m²); is the reference irradiance (unit: W / m²); is the reverse saturation current; is the output voltage; is the thermal voltage, calculated as , where is the Boltzmann constant, is the electron charge; is the series resistance; is a parallel resistor; is the diode factor; is the battery temperature (unit: Celsius); is the reference temperature in degrees Celsius.

[0114] Based on the predicted output current, the DC power prediction value of the photovoltaic module is calculated.

[0115] Step 4: Input the DC power prediction value into the Sandia Inverter Model to evaluate the inverter output power under different working conditions, i.e., the AC power prediction value. Calculation formula:

[0116]

[0117] is the AC power prediction value of the inverter (unit: W); is the DC to AC conversion efficiency; is the input DC power of the inverter (unit: W).

[0118] Input DC power Calculation formula: ; is the input DC voltage of the inverter (unit: V); is the input DC current of the inverter (unit: A).

[0119] conversion efficiency Calculation formula: ; is a constant term; is the linear term coefficient; is the coefficient of the quadratic term; is the rated power of the inverter (unit: W).

[0120] Step 5: The second compensation module obtains the actual power values ​​corresponding to the power prediction values ​​of the photovoltaic components and the inverter in the historical time period, automatically calculates the correction values ​​of each model parameter of the power conversion model, and completes the corresponding model parameter calibration.

[0121] Step 6: Obtain the meteorological data set at the current time t and input the well-trained photovoltaic power prediction, and output the power prediction value of the photovoltaic inverter system at the current time t.

[0122] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0123] Based on the same inventive concept, the present application also provides a photovoltaic power prediction device for implementing the photovoltaic power prediction method described above. The solution provided by this device is similar to the solution described in the method described above. Therefore, the specific limitations of one or more photovoltaic power prediction device embodiments provided below can be found in the limitations of the photovoltaic power prediction method described above and will not be repeated here.

[0124] In an exemplary embodiment, Figure 7 As shown, a photovoltaic power generation power prediction device is provided, including: a model construction module 701, a model training module 702 and a power prediction module 703, wherein:

[0125] The model building module 701 is used to build a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the acquired second component parameters of the inverter.

[0126] The model training module 702 is used to obtain a historical meteorological data set of a target area, and perform model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model.

[0127] The power prediction module 703 is used to obtain the original meteorological data set of the current time period, and output the power prediction result of the photovoltaic power generation system through the trained photovoltaic power generation power prediction model.

[0128] In one embodiment, the model training module 702 is also used to: reconstruct the resolution of the historical meteorological data set to obtain a high-precision historical meteorological data set; locally revise the high-precision historical meteorological data to obtain a revised historical meteorological data set; train the photovoltaic power generation model based on the revised historical meteorological data set, the first component parameters, and the second component parameters, and output the training power prediction value of the photovoltaic power generation system; obtain the actual power prediction value of the photovoltaic power generation system, and calculate the model parameter correction value of the photovoltaic power generation power prediction model based on the actual power prediction value and the training power prediction value; and perform parameter correction on the photovoltaic power generation power prediction model based on the model parameter correction value to obtain the fully trained photovoltaic power generation power prediction model.

[0129] In one embodiment, the model training module 702 is also used to: extract low-frequency information of the data grid to obtain shallow features of the historical meteorological data set; perform long-distance connection and convolution operations on the shallow features to obtain deep features of the historical meteorological data set; and fuse and upsample the shallow features and the deep features to obtain the high-precision historical meteorological data set.

[0130] In one embodiment, the model training module 702 is also used to: obtain the deviation correction value of the meteorological data at the same time of the previous day, the actual value of the meteorological data at the same time of the previous day, and the forecast value of the meteorological data at the same time of the previous day, and calculate the deviation correction value of the high-precision historical meteorological data at the same time of the day; superimpose the deviation correction value at the same time of the day with the high-precision historical meteorological data at the same time of the day to obtain the corrected historical meteorological data.

[0131] In one embodiment, the model training module 702 is also used to: calculate the DC power training prediction value of the photovoltaic component based on the revised historical meteorological data and the first component parameters; calculate the AC power training prediction value of the inverter based on the revised historical meteorological data, the second component parameters, and the DC power training prediction value of the photovoltaic component.

[0132] In one embodiment, the model training module 702 is also used to: obtain the actual DC power value of the photovoltaic component and the actual AC power value of the inverter; determine the DC parameter correction value of the photovoltaic power generation prediction model based on the actual DC power value and the DC power training prediction value; determine the AC parameter correction value of the photovoltaic power generation prediction model based on the actual AC power value and the AC power training prediction value.

[0133] Each module in the photovoltaic power prediction device described above can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in hardware form, or can be stored in a computer device memory in software form, so that the processor can call and execute the corresponding operations of each module.

[0134] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, memory and input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store photovoltaic module data and inverter data. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a photovoltaic power generation power prediction method is implemented.

[0135] Those skilled in the art will understand that Figure 8 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0136] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0137] Constructing a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter;

[0138] Acquiring a historical meteorological data set of a target area, and performing model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model;

[0139] The original meteorological data set of the current time period is obtained, and the power prediction result of the photovoltaic power generation system is output through the trained photovoltaic power generation power prediction model.

[0140] In one embodiment, when the processor executes the computer program, it also implements the following steps: reconstructing the resolution of the historical meteorological data set to obtain a high-precision historical meteorological data set; locally revising the high-precision historical meteorological data to obtain a revised historical meteorological data set; training the photovoltaic power generation model based on the revised historical meteorological data set, the first component parameters, and the second component parameters, and outputting a training power prediction value of the photovoltaic power generation system; obtaining the actual power prediction value of the photovoltaic power generation system, and calculating the model parameter correction value of the photovoltaic power generation power prediction model based on the actual power prediction value and the training power prediction value; and performing parameter correction on the photovoltaic power generation power prediction model based on the model parameter correction value to obtain the fully trained photovoltaic power generation power prediction model.

[0141] In one embodiment, when the processor executes the computer program, it also implements the following steps: extracting low-frequency information of the data grid to obtain shallow features of the historical meteorological dataset; performing long-distance connection and convolution operations on the shallow features to obtain deep features of the historical meteorological dataset; and fusing and upsampling the shallow features and the deep features to obtain the high-precision historical meteorological dataset.

[0142] In one embodiment, when the processor executes the computer program, it also implements the following steps: obtaining the deviation correction value of the meteorological data at the same time of the previous day, the actual value of the meteorological data at the same time of the previous day, and the forecast value of the meteorological data at the same time of the previous day, and calculating the deviation correction value of the high-precision historical meteorological data at the same time of the day; superimposing the deviation correction value at the same time of the day with the high-precision historical meteorological data at the same time of the day to obtain the corrected historical meteorological data.

[0143] In one embodiment, when the processor executes the computer program, it also implements the following steps: based on the revised historical meteorological data and the first component parameters, the DC power training prediction value of the photovoltaic component is calculated; based on the revised historical meteorological data, the second component parameters, and the DC power training prediction value of the photovoltaic component, the AC power training prediction value of the inverter is calculated.

[0144] In one embodiment, when the processor executes the computer program, the following steps are also implemented: obtaining the actual DC power value of the photovoltaic component and the actual AC power value of the inverter; determining the DC parameter correction value of the photovoltaic power generation prediction model based on the actual DC power value and the DC power training prediction value; determining the AC parameter correction value of the photovoltaic power generation prediction model based on the actual AC power value and the AC power training prediction value.

[0145] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the photovoltaic power generation power prediction method described in the above embodiments is implemented.

[0146] In one embodiment, a computer program product is provided, comprising a computer program, which implements the photovoltaic power generation power prediction method when executed by a processor.

[0147] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), quantum computing-based data processing logic devices, artificial intelligence (AI) processors, and the like.

[0148] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0149] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A photovoltaic power generation power prediction method, applied to a photovoltaic power generation system, wherein the photovoltaic power generation system includes a photovoltaic module and an inverter, characterized in that: The method comprises: Constructing a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter; Acquiring a historical meteorological data set of a target area, and performing model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model; The original meteorological data set of the current time period is obtained, and the power prediction result of the photovoltaic power generation system is output through the trained photovoltaic power generation power prediction model.

2. The method according to claim 1, characterized in that The performing model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set includes: Reconstructing the resolution of the historical meteorological dataset to obtain a high-precision historical meteorological dataset; The high-precision historical meteorological data is locally revised to obtain a revised historical meteorological data set; Training the photovoltaic power generation model according to the revised historical meteorological data set, the first component parameters, and the second component parameters, and outputting a training power prediction value of the photovoltaic power generation system; Obtaining an actual power prediction value of the photovoltaic power generation system, and calculating a model parameter correction value of the photovoltaic power generation power prediction model based on the actual power prediction value and the training power prediction value; The photovoltaic power generation prediction model is parameter-corrected according to the model parameter correction value to obtain the fully trained photovoltaic power generation prediction model.

3. The method according to claim 2, characterized in that The historical meteorological dataset is stored in a data grid format, and reconstructing the resolution of the historical meteorological dataset to obtain a high-precision historical meteorological dataset includes: Extracting low-frequency information of the data grid to obtain shallow features of the historical meteorological dataset; Performing long-distance connections and convolution operations on the shallow features to obtain deep features of the historical meteorological dataset; The shallow features and the deep features are fused and upsampled to obtain the high-precision historical meteorological dataset.

4. The method according to claim 3, characterized in that The localized revision of the high-precision historical meteorological dataset to obtain a revised historical meteorological dataset includes: Obtaining the deviation correction value of the meteorological data at the same time on the previous day, the actual value of the meteorological data at the same time on the previous day, and the forecast value of the meteorological data at the same time on the previous day, and calculating the deviation correction value of the high-precision historical meteorological data at the same time on the current day; The deviation correction value at the same time on the same day is superimposed with the high-precision historical meteorological data at the same time on the same day to obtain the corrected historical meteorological data.

5. The method according to claim 2, characterized in that The training power prediction value includes a DC power training prediction value and an AC power training prediction value. The photovoltaic power generation power model is trained based on the corrected historical meteorological data, the first component parameters, and the second component parameters, and the training power prediction value of the photovoltaic power generation system obtained by outputting includes: Calculating a DC power training prediction value of the photovoltaic module based on the revised historical meteorological data and the first module parameter; The AC power training prediction value of the inverter is calculated based on the revised historical meteorological data, the second component parameters, and the DC power training prediction value of the photovoltaic component.

6. The method according to claim 5, characterized in that The model parameter correction values ​​include DC parameter correction values ​​and AC parameter correction values. The obtaining of the actual power prediction value of the photovoltaic power generation system and calculating the model parameter correction values ​​of the photovoltaic power generation power prediction model according to the actual power prediction value and the training power prediction value include: Obtaining an actual value of the DC power of the photovoltaic module and an actual value of the AC power of the inverter; Determining a DC parameter correction value of the photovoltaic power generation prediction model according to the DC power actual value and the DC power training prediction value; An AC parameter correction value of the photovoltaic power generation power prediction model is determined according to the AC power actual value and the AC power training prediction value.

7. A photovoltaic power generation power prediction device, characterized in that: The device comprises: A model building module is used to build a photovoltaic power generation prediction model based on the acquired first component parameters of the photovoltaic component and the second component parameters of the inverter; A model training module is used to obtain a historical meteorological data set of a target area, and perform model training and parameter calibration on the photovoltaic power generation prediction model based on the historical meteorological data set to obtain a fully trained photovoltaic power generation prediction model; The power prediction module is used to obtain the original meteorological data set of the current time period, and output the power prediction result of the photovoltaic power generation system through the trained photovoltaic power generation power prediction model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images