Photovoltaic power generation control method, system, equipment, medium and product

Abstract

The invention discloses a photovoltaic power generation control method, system and device, a medium and a product, and belongs to the technical field of photovoltaic power generation control, and the method comprises the steps: coding sunrise time, sunset time and a weather shade and sunny prediction value, and obtaining sunrise and sunset coding information and shade and sunny coding information; predicting weather according to the weather information, the sunrise and sunset coding information, the sunshine coding information and the weather intervention influence information to obtain a weather prediction value; according to the weather prediction value, the historical generating capacity and the historical load, predicting the load and the generating capacity of the next time period to obtain a load prediction value and a generating capacity prediction value; and controlling photovoltaic power generation according to the load predicted value and the generating capacity predicted value. According to the invention, accurate control of photovoltaic power generation is realized, and the technical problem that the photovoltaic power generation cannot be accurately controlled due to the fact that the generating capacity cannot be accurately predicted at present is solved.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power generation control technology, and in particular to a photovoltaic power generation control method, system, equipment, medium and product. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] As a new energy source, photovoltaic power generation is becoming increasingly important and has a wide range of application prospects. However, due to the influence of weather, seasons, climate and other factors, the power generation of photovoltaic power generation has certain fluctuations and instability. In order to achieve accurate control of photovoltaic power generation, it is necessary to accurately predict the power generation of photovoltaic power generation.

[0004] Currently, the power generation of photovoltaic power generation is mainly predicted through weather information and historical load at the photovoltaic power generation site. However, due to the different sunrise and sunset times in different locations and the impact of meteorological intervention measures on the weather, when the power generation of photovoltaic power generation is predicted only by weather information and historical load, the accuracy of the power generation prediction value obtained is not high, and thus precise control of photovoltaic power generation cannot be achieved. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a photovoltaic power generation control method, system, equipment, medium and product, which can accurately predict the power generation of photovoltaic power generation based on accurate weather prediction, and then realize precise control of photovoltaic power generation.

[0006] To achieve the above object, the present invention adopts the following technical solution: In the first aspect, a photovoltaic power generation control method is proposed, comprising: Obtain weather information, sunrise time, sunset time, light intensity forecast, weather forecast, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location; Encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information; Predict the weather based on weather information, sunrise and sunset coding information, cloudy and sunny coding information and meteorological intervention impact information to obtain weather forecast values; The load and power generation in the next period are predicted through weather forecast values, historical power generation and historical load, so as to obtain load forecast values ​​and power generation forecast values; Photovoltaic power generation is controlled according to the load forecast value and the power generation forecast value.

[0007] Furthermore, the process of obtaining weather information at the photovoltaic power generation location includes: Obtain weather information for the nearest time period at the photovoltaic power generation location; The weather information of the nearest time period is completed to obtain the weather information at the photovoltaic power generation location.

[0008] Further, determining the time of missing data in the weather information of the nearest neighboring time period; For missing data in continuous time, the weather information prediction model is used to process the weather information of the nearest time period to obtain the missing data prediction results; For the missing data in discrete time, the matrix decomposition mechanism is used to decompose the weather information of the nearest time period into a low-rank representation to obtain the missing data prediction results; According to the time of missing data, the missing data prediction results are added to the weather information of the nearest time period to obtain the weather information at the photovoltaic power generation location.

[0009] Furthermore, information data related to weather information, historical impact information of light intensity, the tilt angle of the earth and the longitude of each photovoltaic power generation location are obtained at the photovoltaic power generation location; According to the historical impact information of light intensity, the light intensity is predicted to obtain the predicted value of light intensity; Determine the sunrise time at the photovoltaic power generation location based on the earth's tilt angle and the longitude of each photovoltaic power generation location; According to the information data related to the weather information at the photovoltaic power generation location, the weather forecast value at the photovoltaic power generation location is obtained.

[0010] Further, obtaining meteorological intervention text related to the photovoltaic power generation location; Extract intervention purpose, intervention content and intervention impact from meteorological intervention texts; According to the intervention impact, obtain the impact of the intervention on the weather; According to the impact of intervention on weather and weather information, the impact information of meteorological intervention is obtained.

[0011] Furthermore, the sunrise time and the sunset time are encoded using a sine function and a cosine function to obtain sunrise encoding information and sunset encoding information.

[0012] Further, the power generation forecast value is allocated to different power grid areas according to the load forecast value and the power generation forecast value; Each grid area controls photovoltaic power generation according to the allocated power generation.

[0013] Furthermore, with the goal of minimizing the sum of squared errors between the load forecast value and the power generation forecast value, and with the constraint that the sum of the power generation allocated to all power grid areas is equal to the power generation forecast value, a power generation allocation model is constructed: the power generation allocation model is solved to obtain the power generation allocated to each power grid area.

[0014] Furthermore, the power generation forecast value is also visualized.

[0015] Furthermore, a time series prediction model is used to process weather information, sunrise and sunset coding information, cloudy and sunny coding information and meteorological intervention impact information to obtain weather prediction values.

[0016] In the second aspect, a photovoltaic power generation control system is proposed, comprising: An acquisition unit is used to acquire weather information, sunrise time, sunset time, light intensity forecast value, weather forecast value, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location; The processing unit is used to encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information; predict the weather according to the weather information, sunrise and sunset coding information, weather forecast information and meteorological intervention impact information to obtain weather forecast value; predict the load and power generation of the next period by weather forecast value, historical power generation and historical load to obtain load forecast value and power generation forecast value; control photovoltaic power generation according to the load forecast value and power generation forecast value.

[0017] In a third aspect, a computer device is provided, the device comprising: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a photovoltaic power generation control method proposed in the first aspect is implemented.

[0018] In a fourth aspect, a computer-readable storage medium is proposed, wherein the computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a photovoltaic power generation control method proposed in the first aspect.

[0019] In a fifth aspect, a computer program product is proposed, wherein the computer program product includes a computer program, and when the computer program is executed by a processor, the photovoltaic power generation control method proposed in the first aspect is implemented.

[0020] Compared with the prior art, the present invention has the following beneficial effects: 1. The method disclosed in the present invention fully considers factors such as weather information, sunrise time, sunset time, light intensity, weather conditions and the influence of meteorological intervention when making weather forecasts, thereby achieving accurate weather forecasts. When the accurate weather forecast value is used to predict photovoltaic power generation, the accuracy of the power generation forecast value is improved.

[0021] 2. The method disclosed in the present invention can better analyze and control weather factors by completing weather information, encoding sunrise time, sunset time, and weather information, further improving the accuracy of weather forecasts, and achieving precise control of photovoltaic power generation.

[0022] 3. The present invention aims to minimize the sum of squares of errors between load prediction values ​​and power generation prediction values, and distributes power generation prediction values ​​so that the error between the power generation allocated to each power generation area and the load is minimized, thereby further ensuring the accuracy of control over photovoltaic power generation.

[0023] 4. When predicting power generation, the present invention fully considers weather information, historical power generation and historical load, thereby ensuring the accuracy of power generation prediction.

[0024] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A flow chart of a photovoltaic power generation control method is disclosed as an embodiment. DETAILED DESCRIPTION

[0027] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0028] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0029] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0030] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0031] Example 1 In order to achieve precise control of photovoltaic power generation, in this embodiment, a photovoltaic power generation control method is disclosed, which accurately predicts the weather by acquiring weather information, sunrise time, sunset time, light intensity prediction value, weather forecast value and meteorological intervention impact information to obtain accurate weather forecast values, and further predicts the power generation by combining the weather forecast values ​​with historical power generation and historical load to obtain power generation forecast values, and precisely controls photovoltaic power generation based on the power generation forecast values.

[0032] This embodiment discloses a photovoltaic power generation control method, such as Figure 1 As shown, including: S1: Obtain weather information, sunrise time, sunset time, light intensity forecast value, weather forecast value, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location.

[0033] Among them, the information data related to the weather information at the photovoltaic power generation location, the historical impact information of the light intensity, the earth's tilt angle and the longitude of each photovoltaic power generation location are obtained; According to the historical impact information of light intensity, the light intensity is predicted to obtain the predicted value of light intensity; Determine the sunrise time at the photovoltaic power generation location based on the earth's tilt angle and the longitude of each photovoltaic power generation location; According to the information data related to the weather information at the photovoltaic power generation location, the weather forecast value at the photovoltaic power generation location is obtained.

[0034] The weather information obtained at the photovoltaic power generation location is the weather information of the nearest time period, including temperature, humidity, wind speed, wind direction, rainfall, cloud cover and other information.

[0035] The process of obtaining weather information at the photovoltaic power generation location includes: S101: Obtain weather information of the nearest time period at the photovoltaic power generation location.

[0036] The weather information of the nearest time period includes the current weather information and the historical weather information of the previous time period. This embodiment crawls the historical weather information from data sources such as past weather patterns, seasonal changes and other meteorological condition records, and obtains the current weather information through real-time detection.

[0037] This embodiment adopts an incremental crawling method to crawl historical weather information from a data source, and only obtains updated or newly added weather information to avoid repeated crawling of weather information. The process of crawling historical weather information from a data source using an incremental crawling method includes: S1011: For the weather information item to be crawled, its status information is recorded by means of a timestamp.

[0038] S1012: Perform a check operation regularly to determine weather information items that need to be incrementally crawled.

[0039] S1013: Determine newly added and updated weather information by comparing the recorded weather information status with the latest weather information status in the data source.

[0040] S1014: According to the newly added and updated weather information determined in S1013, only the newly added and updated weather information is crawled.

[0041] S1015: After successfully crawling the weather information, update the recorded weather information status to reflect the latest status.

[0042] In addition, in order to avoid excessive load on the crawled website, this embodiment uses an exponential backoff algorithm to adjust the crawl request waiting time. If the response time of a crawl request is long, the waiting time of the next crawl request is increased exponentially to adjust the waiting time. The initial waiting time is set to , the multiplication factor is , the maximum waiting time is , and the other waiting times are , the calculation formula is: ; ; ; When the maximum waiting time is reached If there is still no success, the failed request, error type and error frequency are recorded, and the backup data source is switched. When all backup data sources are polled, the crawler is restarted and a notification is sent.

[0043] S102: Completing the weather information of the nearest time period to obtain the weather information at the photovoltaic power generation location, the process includes: S1021: Determine the time when data is missing in the weather information of the nearest time period.

[0044] Since the current weather information is acquired through real-time detection, there is usually no missing problem. Therefore, this embodiment mainly supplements the missing data of the historical weather information in the weather information of the nearest time period. Assume that the time series data of the historical weather information in the weather information of the nearest time period is , the time set of missing data is M.

[0045] S1022: For missing data in continuous time, the weather information prediction model is used to process the weather information of the nearest time period to obtain the missing data prediction result; for missing data in discrete time, the matrix decomposition mechanism is used to decompose the weather information of the nearest time period into a low-rank representation to obtain the missing data prediction result.

[0046] Among them, the weather information prediction model is based on the time series data of historical known weather information and As input, the predicted results with missing data For output, use The network model is constructed and obtained through training with relevant training data, specifically: .

[0047] For missing data at discrete time, assume that the missing data is , using the matrix decomposition mechanism, decompose the weather dataset into a low-rank representation: .

[0048] Then, based on the calculated low-rank representation, the missing data is predicted: X ’ j = U.V j X ≈ U.V j .in, represents the complete weather dataset matrix, and They represent the decomposed low-rank matrices, Representation and missing data Corresponding V , X ’ j Represents the predicted value of the missing data point.

[0049] S1023: According to the time of the missing data, the missing data prediction result is added to the weather information of the nearest neighboring time period to complete the weather information of the nearest neighboring time period and obtain the weather information at the photovoltaic power generation location.

[0050] After all missing data are obtained by S1022, the missing data prediction results are added to the weather information of the nearest time period in chronological order according to the time of the missing data to obtain the complete weather information. .

[0051] This embodiment obtains historical impact information of light intensity; predicts the light intensity based on the historical impact information of light intensity to obtain a predicted value of light intensity.

[0052] Among them, the historical impact information of light intensity includes cloud cover , rainfall and temperature Etc. This embodiment processes the historical impact information of light intensity through the light intensity prediction model to achieve the prediction of light intensity and obtain the light intensity prediction value. The process includes: The acquired historical impact information of light intensity is cleaned and filled in to process missing values ​​and abnormal values. After completing the historical impact information, the completed historical impact information of light intensity is standardized. The standardized historical impact information of light intensity is processed through the light intensity prediction model to obtain the light intensity prediction value.

[0053] Among them, the light intensity prediction model takes the historical impact information of light intensity as input and the predicted value of light intensity as output. It is constructed through the SVR model and obtained through relevant data training.

[0054] The specific expression of the light intensity prediction model is: . represents the predicted value of light intensity, and is the Lagrange multiplier corresponding to the support vector (part of the training sample), is the kernel function used to calculate the training samples and prediction samples The kernel function can be a linear kernel, a polynomial kernel, a radial basis function (RBF), etc., which are used to process nonlinear relationships. is the intercept term of the model.

[0055] This embodiment uses cross-validation technology to adjust the hyperparameters of SVR to improve model performance, and detects the effect on the divided test set to calculate the root mean square error (RMSE), R2 score or other appropriate regression performance indicators.

[0056] The process of obtaining the sunrise time and sunset time at the photovoltaic power generation location in this embodiment includes: Get the Earth's tilt angle and the longitude of each photovoltaic power generation location ; According to the tilt of the earth and the longitude of each photovoltaic power generation location , determine the sunrise time at the photovoltaic power generation location for: .

[0057] Sunset time for: .

[0058] The process of obtaining the weather forecast value in this embodiment is as follows: Obtain information data related to sunny and cloudy weather information at the photovoltaic power generation location. Information data related to negative weather information includes temperature, historical average temperature of the day, humidity and wind speed, etc.; The weather forecast model is used to process the information data related to the weather at the photovoltaic power generation location to obtain the weather forecast value at the photovoltaic power generation location.

[0059] Among them, the weather prediction model takes information data related to the weather in the next period as input, and the weather forecast value for the next period as output. It is constructed using the BERT model and obtained through training with relevant data.

[0060] In the specific implementation, the historical weather information, temperature, humidity, wind speed and other information data related to the weather information in the next period are preprocessed. The preprocessing process includes normalization, standardization and data enhancement. Assume that the temperature is The historical average temperature for the day was , humidity is , wind speed is , the weather status of the next period is , collect relevant data , perform data preprocessing : ,in, The preprocessed data set.

[0061] The dataset Arrange it into a prompt template, and mask the cloudy and sunny information. For example: the temperature is 18°C, the historical average temperature of the day is 24°C, the humidity is 92%, the wind speed is level 3, and the weather condition is masked; the temperature is 22°C, the historical average temperature of the day is 24°C, the humidity is 80%, the wind speed is level 2, and the weather condition is masked. The specific formula is as follows: In the formula, For the dataset The dataset obtained after labeling.

[0062] Build a weather prediction model to associate various factors such as temperature, historical daily average temperature, humidity, wind speed, etc. with weather information. Use the natural language model BERT with self-attention mechanism to train and capture the relationship between features, and finally output the weather forecast value. Use probability distribution and label encoding to represent different weather conditions.

[0063] Assume that the weather forecast value is , is the BERT model, whose parameters are : .

[0064] The acquired data on the temperature at the photovoltaic power generation location, the historical average temperature of the day, humidity, wind speed and other information related to the weather are input into the weather prediction model to obtain the weather prediction value.

[0065] The process of obtaining the meteorological intervention impact information in this embodiment is as follows: Get the meteorological intervention text related to the photovoltaic power generation location; Extract intervention purpose, intervention content and intervention impact from meteorological intervention texts; According to the intervention impact, obtain the impact of the intervention on the weather; According to the impact of intervention on weather and weather information, the impact information of meteorological intervention is obtained.

[0066] Assume that the set of meteorological intervention texts related to the photovoltaic power generation location is ,in Indicates Weather intervention texts; Using large models For text collection Interpret and obtain the set of intervention purposes of artificial meteorological intervention , intervention content collection and intervention impact set : .

[0067] According to the intervention impact , determine the impact of intervention on weather: .in, is the weather impact function of intervention, The impact of intervention on weather is collected.

[0068] Function Incorporating weather forecast models , improve prediction accuracy: .in, For weather information, It is the new forecast result, namely the meteorological intervention impact information.

[0069] The meteorological forecast model takes meteorological information and the impact of intervention on the weather as input, and takes information on the impact of meteorological intervention as output, and is constructed through machine learning.

[0070] S2: Encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information.

[0071] This embodiment uses sine function and cosine function to encode the sunrise time and sunset time to obtain sunrise and sunset encoding information: In the formula, is the encoding function, Indicates the sequence number of the current sunrise or sunset time in the year. The total number of days in a year.

[0072] This encoding method allows the model to capture the periodic laws of time information. Each encoding vector is unique, ensuring that the encoded value is bounded and can be linearly transformed, thereby enhancing the learning efficiency of the model.

[0073] This embodiment encodes the information data related to the weather information and the weather forecast value to obtain the weather coding information. : The weather forecast value determined based on factors such as temperature, historical average temperature of the day, humidity and wind speed is encoded, which improves the accuracy and interpretability of the weather coding.

[0074] S3: Forecast the weather according to the weather information, sunrise and sunset coding information, cloudy and sunny coding information and meteorological intervention impact information to obtain a weather forecast value.

[0075] This embodiment uses a time series prediction model to process weather information, sunrise and sunset coding information, cloudy and sunny coding information, and meteorological intervention impact information to obtain weather prediction values. Specifically: S31: Collect 9km*9km grid weather information and other relevant data ,in: Including sunrise and sunset coded information 、Cloud and sky coding information and weather intervention impact information .

[0076] S32: Yes and Perform preprocessing, including cleaning the data ,format and standardization , get the preprocessed data set : .

[0077] S33: Based on the dataset Training a time series forecasting model , the model can predict future weather and obtain weather forecast values : ; .

[0078] S34: Weather forecast value Evaluate and obtain the calculation accuracy and error : ; .

[0079] S35: Model based on evaluation results Perform iterative optimization : .

[0080] The weather information and meteorological intervention information obtained by S1, as well as the sunrise and sunset coding information and cloudy and sunny coding information obtained by S2, are input into the trained time series prediction model. , obtain the weather forecast value, which Including temperature ,humidity , wind speed , precipitation and cloud cover .

[0081] Among them, the time series prediction model can be constructed using a statistical model or a machine learning model.

[0082] S4: forecast the load and power generation in the next period through the weather forecast value, historical power generation and historical load, and obtain the load forecast value and power generation forecast value.

[0083] This embodiment processes the weather forecast value, historical power generation and historical load through the power generation forecast model to predict the load and power generation in the next period and obtain the load forecast value and power generation forecast value.

[0084] The power generation forecasting model takes the time series data such as historical power generation, historical load and weather forecast as input, and uses the power generation forecast value within a certain period of time in the future as the and load forecast value As the output, a recurrent neural network model such as LSTM or GRU is used to build: ,in, f represents the neural network model, and represents the predicted power generation and load values, Indicates the current time. and Respectively represent The actual value of power generation and load at the moment, Represents the weather forecast value.

[0085] S5: Control photovoltaic power generation according to the load forecast value and the power generation forecast value.

[0086] According to the load forecast value and power generation forecast value, the power generation forecast value Distribution to different grid areas; Each grid area controls photovoltaic power generation according to the allocated power generation.

[0087] In this embodiment, the load prediction value and power generation forecast The goal is to minimize the sum of square errors between the two, with the sum of the power generation allocated to all grid areas P total The power generation forecast value is taken as a constraint, and the power generation allocation model is constructed: ; ,in, Representing the power grid area The predicted value of power generation, Represents the total amount of electricity generated by all grid regions.

[0088] The multi-objective optimization algorithm such as greedy algorithm or dynamic programming is used to solve the power generation allocation model and obtain the power generation allocated to each power grid area.

[0089] After obtaining the power generation forecast value, this embodiment also performs a visual display of the power generation forecast value.

[0090] The time series database technology is used to visualize the power generation forecast value, including: real-time display and storage of power generation forecast values, and viewing of stored historical power generation forecast values.

[0091] The process of real-time display and storage of power generation forecast values ​​includes: Query the database to obtain the most recent power generation forecast ,right Perform exponentially weighted smoothing: ; yes The predicted power generation value at the time, yes The predicted value of power generation after smoothing at each moment, is the smoothing weight.

[0092] The design interface dynamically displays the smoothed power generation forecast value , providing time window selection function.

[0093] Using time series database to predict power generation , The time series data is stored and a SQL interface query function is provided for the time series database. In addition, this embodiment also compresses the time series database regularly to reduce costs.

[0094] The process of viewing the stored historical power generation forecast values ​​is as follows: Select any historical time period and load the power generation forecast value for that time period ; Calculation accuracy index , : ; .

[0095] The power generation forecast value to be loaded Displayed in the form of statistical charts.

[0096] By analyzing the relationship between consumption, usage and power generation, intelligent power generation control is achieved to optimize energy utilization and improve supply and demand matching efficiency. The process includes: Read the newly stored consumption, usage and power generation from the database in real time. Suppose the consumption is , the usage is , the power generation is , comparing each group relationship, if If the sum of the consumption and the usage is less than the power generation, an alarm message will be generated and displayed in the front-end alarm bar, and the alarm time and alarm information will be saved in the database in real time.

[0097] A photovoltaic power generation control method disclosed in this embodiment fully considers factors such as weather information, sunrise time, sunset time, light intensity, cloudy or sunny weather, and the influence of meteorological intervention when making weather forecasts, thereby achieving accurate weather forecasts. When the accurate weather forecast value is used to predict photovoltaic power generation, the accuracy of the power generation forecast value is improved.

[0098] This embodiment can better analyze and control weather factors by completing weather information, encoding sunrise time, sunset time, and weather information, further improving the accuracy of weather forecasts, and achieving precise control of photovoltaic power generation.

[0099] The method disclosed in this embodiment aims to minimize the sum of squares of errors between load prediction values ​​and power generation prediction values, and distributes power generation prediction values ​​so that the error between the power generation allocated to each power generation area and the load is minimized, thereby further ensuring the accuracy of control over photovoltaic power generation.

[0100] The method disclosed in this embodiment fully considers weather information, historical power generation and historical load when predicting power generation, thereby ensuring the accuracy of power generation prediction.

[0101] It should be noted that all data is obtained in compliance with laws and regulations and user consent, and the data is used legally.

[0102] Example 2 In this embodiment, a photovoltaic power generation control system is disclosed, including: An acquisition unit is used to acquire weather information, sunrise time, sunset time, light intensity forecast value, weather forecast value, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location; The processing unit is used to encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information; predict the weather according to the weather information, sunrise and sunset coding information, weather forecast information and meteorological intervention impact information to obtain weather forecast value; predict the load and power generation of the next period by weather forecast value, historical power generation and historical load to obtain load forecast value and power generation forecast value; control photovoltaic power generation according to the load forecast value and power generation forecast value.

[0103] Preferably, the process of the acquisition unit for acquiring weather information at the photovoltaic power generation location includes: Obtain weather information for the nearest time period at the photovoltaic power generation location; The weather information of the nearest time period is completed to obtain the weather information at the photovoltaic power generation location.

[0104] Preferably, the acquisition unit is used to determine the time of missing data in the weather information of the nearest neighboring time period; For missing data in continuous time, the weather information prediction model is used to process the weather information of the nearest time period to obtain the missing data prediction results; for missing data in discrete time, the matrix decomposition mechanism is used to decompose the weather information of the nearest time period into a low-rank representation to obtain the missing data prediction results; According to the time of missing data, the missing data prediction results are added to the weather information of the nearest time period to obtain the weather information at the photovoltaic power generation location.

[0105] Preferably, the acquisition unit is used to acquire information data related to weather information, historical impact information of light intensity, the earth's tilt angle and the longitude of each photovoltaic power generation location at the photovoltaic power generation location; According to the historical impact information of light intensity, the light intensity is predicted to obtain the predicted value of light intensity; Determine the sunrise time at the photovoltaic power generation location based on the earth's tilt angle and the longitude of each photovoltaic power generation location; According to the information data related to the weather information at the photovoltaic power generation location, the weather forecast value at the photovoltaic power generation location is obtained.

[0106] Preferably, the acquisition unit is used to acquire the meteorological intervention text related to the photovoltaic power generation location; Extract intervention purpose, intervention content and intervention impact from meteorological intervention texts; According to the intervention impact, obtain the impact of the intervention on the weather; According to the impact of intervention on weather and weather information, the impact information of meteorological intervention is obtained.

[0107] Preferably, the processing unit is further used to distribute the power generation forecast value to different grid areas according to the load forecast value and the power generation forecast value; each grid area controls photovoltaic power generation according to the distributed power generation.

[0108] It should be noted that the photovoltaic power generation control system provided in the above embodiment is only illustrated by the division of the above functions. In actual applications, the above functions can be assigned to different functional modules as needed to complete all or part of the functions described above. In addition, the photovoltaic power generation control system provided in the above embodiment and the photovoltaic power generation control method embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0109] The present invention also discloses a computer device, which includes: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a photovoltaic power generation control method disclosed in Example 1 is implemented.

[0110] The present invention also discloses a computer-readable storage medium, which stores a computer program. The computer program is suitable for being loaded by a processor and executing a photovoltaic power generation control method disclosed in Example 1.

[0111] The present invention also discloses a computer program product, which includes a computer program. When the computer program is executed by a processor, a photovoltaic power generation control method disclosed in Example 1 is implemented.

[0112] The method disclosed in Example 1 can be directly embodied as a hardware processor, or can be executed by a combination of hardware and software modules in the processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0113] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with this embodiment can be implemented in electronic hardware or a combination of computer software and electronic hardware. 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.

[0114] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A photovoltaic power generation control method, characterized in that: include: Obtain weather information, sunrise time, sunset time, light intensity forecast, weather forecast, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location; Encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information; Predict the weather based on weather information, sunrise and sunset coding information, cloudy and sunny coding information and meteorological intervention impact information to obtain weather forecast values; The load and power generation in the next period are predicted through weather forecast values, historical power generation and historical load, so as to obtain load forecast values ​​and power generation forecast values; Photovoltaic power generation is controlled according to the load forecast value and the power generation forecast value.

2. A photovoltaic power generation control method as claimed in claim 1, characterized in that: The process of obtaining weather information at the photovoltaic power generation location includes: Obtain weather information for the nearest time period at the photovoltaic power generation location; The weather information of the nearest time period is completed to obtain the weather information at the photovoltaic power generation location.

3. A photovoltaic power generation control method as claimed in claim 2, characterized in that: Determine the time when the data is missing in the weather information of the nearest neighboring time period; For missing data in continuous time, the weather information prediction model is used to process the weather information of the nearest time period to obtain the missing data prediction results; For the missing data in discrete time, the matrix decomposition mechanism is used to decompose the weather information of the nearest time period into a low-rank representation to obtain the missing data prediction results; According to the time of missing data, the missing data prediction results are added to the weather information of the nearest time period to obtain the weather information at the photovoltaic power generation location.

4. A photovoltaic power generation control method as claimed in claim 1, characterized in that: Obtain information data related to weather conditions at photovoltaic power generation locations, historical impact information on light intensity, the earth's tilt angle, and the longitude of each photovoltaic power generation location; According to the historical impact information of light intensity, the light intensity is predicted to obtain the predicted value of light intensity; Determine the sunrise time at the photovoltaic power generation location based on the earth's tilt angle and the longitude of each photovoltaic power generation location; According to the information data related to the weather information at the photovoltaic power generation location, the weather forecast value at the photovoltaic power generation location is obtained.

5. A photovoltaic power generation control method as claimed in claim 1, characterized in that: Get the meteorological intervention text related to the photovoltaic power generation location; Extract intervention purpose, intervention content and intervention impact from meteorological intervention texts; According to the intervention impact, obtain the impact of the intervention on the weather; According to the impact of intervention on weather and weather information, the impact information of meteorological intervention is obtained.

6. A photovoltaic power generation control method as claimed in claim 1, characterized in that: According to the load forecast value and the power generation forecast value, the power generation forecast value is allocated to different power grid areas; Each grid area controls photovoltaic power generation according to the allocated power generation.

7. A photovoltaic power generation control system, characterized in that: include: An acquisition unit is used to acquire weather information, sunrise time, sunset time, light intensity forecast value, weather forecast value, meteorological intervention impact information, historical power generation and historical load at the photovoltaic power generation location; The processing unit is used to encode the sunrise time, sunset time and weather forecast value respectively to obtain sunrise and sunset coding information and weather forecast information; predict the weather according to the weather information, sunrise and sunset coding information, weather forecast information and meteorological intervention impact information to obtain weather forecast value; predict the load and power generation of the next period by weather forecast value, historical power generation and historical load to obtain load forecast value and power generation forecast value; control photovoltaic power generation according to the load forecast value and power generation forecast value.

8. An electronic device, characterized in that: The device comprises: a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, a photovoltaic power generation control method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing a photovoltaic power generation control method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the photovoltaic power generation control method according to any one of claims 1 to 6 is implemented.

Images