Multi-source new energy output prediction method and device

Abstract

The embodiment of the application provides a kind of multi-source new energy output prediction method and device, its method includes: through error time sequence model, confidence interval evaluation and exception correction are carried out to multi-source meteorological data, utilize AI algorithm in D-1 day initiative selection weather source;Adopt color scale feature recognition technology to extract meteorological data spatial distribution characteristics, realize numerical prediction and geographical coordinate multi-modal accurate matching in combination with physical significance mapping model;Coupling data driving and physical mechanism is constructed Hybrid prediction model, generates new energy power generation power prediction curve;Through the dynamic feedback mechanism of D+1 day prediction result and actual output, iteration optimization weather source evaluation parameter and model weight.The application breaks through the technical bottleneck of multi-source data standardization fusion, improves the utilization efficiency of meteorological data and the explainability of prediction model, provides high reliability decision support for new energy output prediction in power trading scenario.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system new energy power prediction, in particular to a multi-source new energy output prediction method and device. BACKGROUND

[0002] In the modern power system, new energy output prediction is the core technical support of power transaction, which directly affects the transaction strategy formulation and power grid dispatching safety. With the rapid increase of wind power and photovoltaic installed capacity, the strong volatility of their output leads to a sharp increase in transaction deviation penalty risk. The current prediction system highly depends on multi-source meteorological data (such as ECMWF, GFS, regional numerical model), but the characteristics of meteorological sources are significantly different: the spatial and temporal resolution is not unified (hourly to minute level), the prediction reliability fluctuates greatly, and the data format is heterogeneous. In the context of power spot transaction, the existing prediction methods cannot meet the demand for high accuracy, interpretability and dynamic adaptability, resulting in rising losses and balancing costs of abandoned wind and light, and there is an urgent need to break through the technical bottlenecks of multi-source data fusion and physical mechanism embedding.

[0003] The current new energy output prediction field generally adopts data-driven models, and mostly uses LSTM, XGBoost and other pure data-driven algorithms to train the historical meteorological-output mapping relationship. And it relies on artificial experience or historical average error to statically select meteorological sources. This results in the following shortcomings: 1. Low efficiency of meteorological utilization, and the time-varying reliability of meteorological sources is not quantified. Fixed weight or subjective optimization leads to error amplification, and the prediction failure rate increases in extreme weather periods. 2. Poor model interpretability, pure black box model output prediction results cannot be associated with geographical and physical boundaries, and traders cannot assess the risk. SUMMARY

[0004] Therefore, the embodiments of the present application provide a multi-source new energy output prediction method. One or more embodiments of the present application also relate to a multi-source new energy output prediction device, a computing device, a computer readable storage medium and a computer program, to solve the technical defects of the prior art, such as the significant differences in characteristics of multi-source meteorological data, the non-uniformity of meteorological data granularity, the insufficient prediction reliability, and the poor interpretability of machine learning models in the context of power transaction.

[0005] According to a first aspect of the embodiments of the present application, a multi-source new energy output prediction method is provided, comprising:

[0006] Collecting historical meteorological data of multiple meteorological sources in the same region at the same time period, and using the historical meteorological data of each meteorological source to obtain dynamic error weight parameters of each meteorological source at different time periods;

[0007] An AI reinforcement learning multi-objective optimization model of multiple meteorological sources is constructed by using the dynamic error weight parameters of each meteorological source in different time periods, and the optimal meteorological source combination in each time period is obtained by solving the AI reinforcement learning multi-objective optimization model of multiple meteorological sources.

[0008] Meteorological pictures are obtained from the corresponding meteorological source in each time period of the optimal meteorological source combination, and meteorological element information, meteorological element image coordinates and meteorological element color scale parameters are extracted from each meteorological picture by using color scale feature recognition technology.

[0009] A multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination is obtained by spatial matching the meteorological element image coordinates in each time period of the optimal meteorological source combination with target geographical region coordinates and new energy installation information.

[0010] A mixed prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination is determined respectively, and the output prediction result of the multi-source new energy is obtained by inputting the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination into the corresponding mixed prediction model.

[0011] Preferably, the dynamic error weight parameters of each meteorological source in different time periods are obtained by using the historical meteorological data of each meteorological source, including:

[0012] A dynamic confidence interval of each meteorological source in different time periods is constructed by using the forecast value and measured value of each meteorological element data in different time periods contained in the historical meteorological data of each meteorological source.

[0013] The error of each meteorological source is quantitatively evaluated by using the dynamic confidence interval of each meteorological source in different time periods, and the dynamic error weight parameters of each meteorological source in different time periods are obtained.

[0014] Preferably, the dynamic confidence interval of each meteorological source in different time periods is constructed by using the forecast value and measured value of each meteorological element data in different time periods contained in the historical meteorological data of each meteorological source, including:

[0015] The historical forecast error sequence of each meteorological source in different time periods is calculated according to the forecast value and measured value of each meteorological element data in different time periods contained in the historical meteorological data of each meteorological source.

[0016] The distribution characteristics of the historical forecast error sequence of each meteorological source in different time periods are modeled by using quantile regression method, and the upper and lower quantiles of the historical forecast error of each meteorological source in different time periods at a confidence level are calculated respectively.

[0017] According to the upper and lower quantiles of each meteorological source in different time periods, a dynamic confidence interval of each meteorological source in different time periods is constructed.

[0018] Preferably, the utilization of the dynamic confidence interval of each meteorological source in different time periods for quantitative evaluation of the error of each meteorological source obtains a dynamic error weight parameter of each meteorological source in different time periods, which includes:

[0019] According to the dynamic confidence interval of each meteorological source in different time periods, the number of times that the historical prediction error of each meteorological source in different time periods falls into the dynamic confidence interval of the corresponding time period is counted.

[0020] According to the number of times that the historical prediction error of each meteorological source in different time periods falls into the dynamic confidence interval of the corresponding time period, the actual coverage rate of each meteorological source is calculated, and based on the actual coverage rate of each meteorological source, a dynamic error weight parameter of each meteorological source in different time periods is obtained.

[0021] Preferably, the AI reinforcement learning multi-objective optimization model of the multi-meteorological source includes:

[0022]

[0023] In the formula, indicates the activation state of the jth historical meteorological data in the tth time period and the jth meteorological source; , , is a weight coefficient and ; indicates the dynamic error weight parameter of the jth historical meteorological data in the tth time period and the jth meteorological source; indicates the actual coverage rate of the jth historical meteorological data in the tth time period and the jth meteorological source; respectively indicate the mean and variance of the jth historical meteorological data in the tth time period and the jth meteorological source. Preferably, the utilization of the color scale feature recognition technology extracts meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture, which includes:

[0024] The color scale interval for representing the RGB or HSV color value range of each color scale is divided, and based on the color scale interval, a mapping function between the color scale and the meteorological element is established.

[0025]

[0026] ​​​​extracting meteorological element information and meteorological element color scale parameters from each meteorological picture according to the mapping function between the color scale and the meteorological element;

[0027] extracting coordinates in each color scale block from each meteorological picture by using color scale feature recognition technology, and generating coordinate data sets of each pixel point in each color scale block by defining original image coordinate reference points of each meteorological picture, and taking the coordinate data sets of each pixel point in each color scale block as meteorological element image coordinates.

[0028] Preferably, the spatial matching of the meteorological element image coordinates of the optimal meteorological source combination in each time period with target geographical region coordinates and new energy installed capacity information to obtain a multi-modal fusion meteorological data set of the optimal meteorological source combination in each time period comprises:

[0029] defining a conversion relationship between a meteorological element image coordinate system of each meteorological picture and a target geographical region coordinate system;

[0030] spatially matching and positioning the meteorological element image coordinates of each meteorological picture with corresponding target geographical region coordinates thereof by the conversion relationship of each meteorological picture to obtain target geographical region coordinates of each meteorological picture;

[0031] obtaining a multi-modal fusion meteorological data set of the optimal meteorological source combination in each time period by fusing the meteorological element information, meteorological element color scale parameters, meteorological element image coordinates and target geographical region coordinates of each meteorological picture.

[0032] According to the second aspect of the embodiments of the present application, a multi-source new energy output prediction device is provided, comprising:

[0033] The dynamic error weight parameter acquisition module is configured to collect historical meteorological data of multiple meteorological sources in the same time period in the same region, and obtain dynamic error weight parameters of each meteorological source in different time periods by using the historical meteorological data of each meteorological source;

[0034] The optimal meteorological source combination acquisition module is configured to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources by using the dynamic error weight parameters of each meteorological source in different time periods, and obtain an optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of multiple meteorological sources.

[0035] The extraction module is configured to obtain meteorological pictures from corresponding meteorological sources in each time period of the optimal meteorological source combination, and extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture by using color scale feature recognition technology;

[0036] a spatial matching module configured to obtain a multi-modal fusion meteorological data set of each time period of the optimal meteorological source combination by spatially matching the meteorological element image coordinates of each time period of the optimal meteorological source combination with target geographical region coordinates and new energy installation information;

[0037] an output prediction module configured to determine a hybrid prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set of each time period of the optimal meteorological source combination, respectively, and obtain an output prediction result of the multi-source new energy by inputting the multi-modal fusion meteorological data set of each time period of the optimal meteorological source combination into the corresponding hybrid prediction model.

[0038] According to a third aspect of the embodiments of the present application, a computing device is provided, comprising:

[0039] a memory and a processor;

[0040] The memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions to implement the steps of any one of the multi-source new energy output prediction methods.

[0041] According to a fourth aspect of the embodiments of the present application, a computer readable storage medium is provided, which stores computer executable instructions, and the instructions are executed by a processor to implement the steps of any one of the multi-source new energy output prediction methods.

[0042] According to a fifth aspect of the embodiments of the present application, a computer program is provided, wherein when the computer program is executed in a computer, the computer is caused to execute the steps of the multi-source new energy output prediction method.

[0043] The multi-source new energy output prediction scheme provided by the embodiments of the present application performs confidence interval evaluation and abnormal correction on multi-source meteorological data through an error time series model, actively selects an optimal meteorological source on D-1 day using an AI algorithm, extracts spatial distribution characteristics of meteorological data using a color scale feature recognition technology, and realizes multi-modal accurate matching of numerical prediction and geographical coordinates in combination with a physical significance mapping model. A hybrid prediction model coupled with data driving and physical mechanism is constructed to generate a new energy power generation prediction curve. Through a dynamic feedback mechanism of the prediction result on D+1 day and the actual output, meteorological source evaluation parameters and model weights are iteratively optimized. The present application breaks through the technical bottleneck of multi-source data standardization fusion, improves the utilization efficiency of meteorological data and the explainability of the prediction model, and provides high-reliability decision support for new energy output prediction in the power trading scenario. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 is a flowchart of a multi-source new energy output prediction method provided by an embodiment of the present application;

[0045] Figure 2 is a schematic diagram of a multi-source new energy output prediction device provided by an embodiment of the present application;

[0046] Figure 3 is a whole flowchart of a multi-source new energy output prediction method provided by an embodiment of the present application;

[0047] Figure 4 is a structural block diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0048] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application. However, the present application can be practiced without the specific details, and it is understood that the present application is not limited to the embodiments described herein. In other instances, well-known methods, procedures, components, and circuits have not been described in detail as not to unnecessarily obscure aspects of the present application.

[0049] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of one or more embodiments of the present application. As used in one or more embodiments of the present application and the accompanying claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0050] Figure 1 A flowchart of a multi-source new energy output prediction method provided by an embodiment of the present application is shown, which specifically includes the following steps.

[0051] Step S101: Collect historical meteorological data of the same time period in the same region from multiple meteorological sources, and obtain dynamic error weight parameters of each meteorological source in different time periods by using historical meteorological data of each meteorological source.

[0052] In an embodiment of the present application, the step of obtaining dynamic error weight parameters of each meteorological source in different time periods by using historical meteorological data of each meteorological source includes: constructing a dynamic confidence interval of each meteorological source in different time periods by using forecast values and measured values of each meteorological element data in different time periods contained in the historical meteorological data of each meteorological source; and quantitatively evaluating the error of each meteorological source by using the dynamic confidence interval of each meteorological source in different time periods to obtain the dynamic error weight parameters of each meteorological source in different time periods.

[0053] In one embodiment of this application, the step of constructing a dynamic confidence interval for each meteorological source in different time periods using the forecast and measured values ​​of each meteorological element data contained in the historical meteorological data of each meteorological source at different time periods includes: calculating the historical forecast error sequence of each meteorological source in different time periods based on the forecast and measured values ​​of each meteorological element data contained in the historical meteorological data of each meteorological source at different time periods; modeling the distribution characteristics of the historical forecast error sequence of each meteorological source in different time periods using the quantile regression method, and calculating the upper and lower quantiles of the historical forecast error of each meteorological source at different time periods at the confidence level; and constructing a dynamic confidence interval for each meteorological source in different time periods based on the upper and lower quantiles of each meteorological source at different time periods.

[0054] In one embodiment of this application, the step of using the dynamic confidence intervals of each meteorological source under different time periods to quantify and evaluate the error of each meteorological source and obtain the dynamic error weight parameters of each meteorological source under different time periods includes: according to the dynamic confidence intervals of each meteorological source under different time periods, counting the number of times the historical forecast error of each meteorological source falls into the dynamic confidence interval of its corresponding time period under different time periods; calculating the actual coverage rate of each meteorological source based on the number of times the historical forecast error of each meteorological source falls into the dynamic confidence interval of its corresponding time period, and obtaining the dynamic error weight parameters of each meteorological source under different time periods based on the actual coverage rate of each meteorological source.

[0055] Step S102: Using the dynamic error weight parameters of each meteorological source under different time periods, construct an AI reinforcement learning multi-objective optimization model for multiple meteorological sources, and obtain the optimal combination of meteorological sources under each time period by solving the AI ​​reinforcement learning multi-objective optimization model for multiple meteorological sources.

[0056] In one embodiment of this application, the AI ​​reinforcement learning multi-objective optimization model for multiple meteorological sources includes:

[0057]

[0058] In the formula, Indicates the t-th time period. The activation status of the j-th historical meteorological data in a meteorological source; , , The weighting coefficients and ; Indicates the t-th time period. Dynamic error weighting parameters for the j-th historical meteorological data from a meteorological source; Indicates the t-th time period. The actual coverage rate of the j-th historical meteorological data in a meteorological source; They represent the t-th time period and the th time period respectively. The mean and variance of the j-th historical meteorological data from _ _ meteorological sources_.

[0059] Step S103: Obtain meteorological images from the meteorological sources corresponding to each time period of the optimal meteorological source combination, and use color scale feature recognition technology to extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological image;

[0060] In one embodiment of this application, the step of extracting meteorological element information, meteorological element image coordinates, and meteorological element color level parameters from each meteorological image using color level feature recognition technology includes: dividing color level intervals into RGB or HSV color value ranges to represent each color level, and establishing a mapping function between color levels and meteorological elements based on the color level intervals; extracting meteorological element information and meteorological element color level parameters from each meteorological image according to the mapping function between color levels and meteorological elements; extracting coordinates within each color level block from each meteorological image using color level feature recognition technology, and generating a coordinate dataset of each pixel in each color level block by defining the original image coordinate reference point of each meteorological image, and using the coordinate dataset of each pixel in each color level block as the meteorological element image coordinates.

[0061] Step S104: By spatially matching the image coordinates of the meteorological elements in each time period of the optimal meteorological source combination with the coordinates of the target geographic area and the information on the installed capacity of new energy, a multimodal fusion meteorological dataset for each time period of the optimal meteorological source combination is obtained.

[0062] In one embodiment of this application, the step of obtaining a multimodal fused meteorological dataset for each time period of the optimal meteorological source combination by spatially matching the meteorological element image coordinates of each meteorological source combination with the target geographic region coordinates and new energy installed capacity information includes: defining the transformation relationship between the meteorological element image coordinate system and the target geographic region coordinate system of each meteorological image; spatially matching and locating the meteorological element image coordinates of each meteorological image with its corresponding target geographic region coordinates through the transformation relationship of each meteorological image to obtain the target geographic region coordinates of each meteorological image; and fusing the meteorological element information, meteorological element color level parameters, meteorological element image coordinates, and target geographic region coordinates of each meteorological image to obtain the multimodal fused meteorological dataset for each time period of the optimal meteorological source combination.

[0063] Step S105: Determine the hybrid prediction model corresponding to the meteorological element information in the multimodal fusion meteorological dataset for each time period of the optimal meteorological source combination, and obtain the power output prediction result of the multi-source new energy by inputting the multimodal fusion meteorological dataset for each time period of the optimal meteorological source combination into the corresponding hybrid prediction model.

[0064] The following is in conjunction with the appendix Figure 2 Taking the application of the multi-source renewable energy output prediction method provided in this application in the field of renewable energy power prediction in power systems as an example, the multi-source renewable energy output prediction method will be further explained. Among them, Figure 2 The flowchart of a multi-source renewable energy output prediction method according to an embodiment of this application is shown, which specifically includes the following steps.

[0065] S1: Collect historical meteorological data from each meteorological source, process outliers and missing values, and construct an error time series model. Evaluate and analyze the errors of each meteorological source within a certain confidence interval; specifically, this includes:

[0066] S11: Collect historical meteorological data from multiple meteorological sources for the same region or region at the same time, including key meteorological elements such as wind speed, temperature, and irradiance; use the sliding window statistical method to detect local outliers; and use interpolation to modify or fill in the data to generate standardized historical meteorological time-series data.

[0067] S12: For each meteorological source, its historical forecast error sequence is defined as follows:

[0068]

[0069] In the formula, For the first Time period The first of the meteorological sources A historical weather forecast value, For the first Time period The first of the meteorological sources Historical meteorological measurements; For the first Time period The first of the meteorological sources Forecasting errors based on historical meteorological data.

[0070] Quantile regression method is used to analyze the error series Model the distribution characteristics and calculate the errors respectively. At confidence level Lower and lower quantiles , .

[0071]

[0072]

[0073] Set different confidence levels according to different error distributions in different time periods Corresponding to generate different quantile intervals , Different from the fixed interval method, dynamically adapt to the time-varying characteristics of errors.

[0074] For example, the confidence level is 95%, that is , is 1-5% / 2=97.5%, and 5% / 2=2.5%.

[0075] S13: For each weather source, use the point-by-point backtracking method to verify whether its historical error value falls into the dynamic confidence interval at the corresponding time , Count the number of times , Calculate the actual coverage rate:

[0076]

[0077] In the formula, is the total number of times that the error value of the jth historical weather of the nth weather source in the tth period falls into its corresponding confidence interval, is the total number of data of the nth historical weather of the nth weather source under the weather data in the tth period, is the actual coverage rate of the jth historical weather in the nth weather source in the tth period.

[0078] Construct a dynamic weight function:

[0079]

[0080] In the formula, , is the mean and variance of the error sequence in the tth period; is the smoothing coefficient to avoid zero denominator. Normalize the weight to get the normalized dynamic weight parameter

[0081] :

[0082] ​​​​​​​​​​

[0083] S14: The normalized dynamic weight parameter is associated and stored with the weather source and weather data, forming a preferred parameter library. The higher the weight, the more optimal the comprehensive reliability.

[0084]

[0085] S2: On D-1 day, according to the D-day weather original information and evaluation analysis result, the AI is used for active selection of weather source;

[0086] On D-1 day, according to the D-day weather original forecast data and the weather data error weight parameter of each weather source, the AI is used for dynamic screening of adaptive weather source, and the optimal weather source combination is output.

[0087] Further, the AI active selection of weather source includes:

[0088] S21: An AI reinforcement learning multi-objective optimization model of multiple weather sources is constructed for the required weather data:

[0089]

[0090] In the formula, represents the enabled state of the weather source in the time period , , is the weight coefficient; and .

[0091] S22: At least weather sources are selected, and the Q-Learning algorithm is used to iteratively solve the optimal weather source combination of the required weather data in 24 time periods of a day

[0092] Among them, is the optimal weather source selected in the time period under the historical weather .

[0093] S3: The weather data is preprocessed and stored;

[0094] The historical weather forecast value in the weather data is standardized, the abnormal value is removed, and the storage management is performed, and the standardized weather data set is generated;

[0095] S4: The original weather data and the corresponding image coordinates are extracted by using the color scale feature recognition method;

[0096] The color step feature recognition technology is used to analyze and process the spatial distribution characteristics of the meteorological elements in the optimal meteorological source combination, and the original image coordinates and corresponding color step parameters are extracted. The color step parameters are various colors corresponding to the meteorological data such as wind speed, temperature and solar radiation intensity. For example, the analyzed color step parameter is wind speed, 10 m / s corresponds to blue, and 20 m / s corresponds to red. That is, the colors in the image are extracted into specific data of meteorological data and coordinate values of the color. Further, each image corresponds to one kind of meteorological data.

[0097] Further, the spatial distribution characteristics of the meteorological data are analyzed and processed, including:

[0098] S41: Dividing color step intervals , wherein represents the RGB or HSV color value range of the i-th color step.

[0099] Based on the optimal meteorological source combination prediction data, a color step-element mapping function is established:

[0100]

[0101] In the formula, is the color step corresponding to the meteorological element value interval.

[0102] S42: Using image recognition technology to extract the coordinates in each color step block, defining the original image coordinate reference point, and generating the i-th block coordinate data set ;

[0103] In the formula, is the horizontal coordinate of the i-th point in the i-th block, is the vertical coordinate of the i-th point in the i-th block. The boundary coordinates of the preset target province and the i-th county are preset.

[0104] In the formula, is the horizontal coordinate of the i-th point in the i-th block, is the vertical coordinate of the i-th point in the i-th block.

[0105] S43: Defining the conversion relationship between the image coordinate system and the geographic coordinate system

[0106] S43: Defining the conversion relationship between the image coordinate system and the geographic coordinate system

[0107] ​​​​​​​​​​​Let the original meteorological image resolution be Pixels

[0108]

[0109]

[0110] In the formula, , Respectively, , Corresponding horizontal and vertical coordinates in the geographic coordinate system. , The maximum and minimum values of the horizontal coordinate in the geographic coordinate system; , The maximum and minimum values of the vertical coordinate in the geographic coordinate system.

[0111] Through the conversion relationship, the color scale coordinate point is matched and positioned with the geographical boundary of the target county.

[0112] S5: Build an image coordinate-provincial and county area coordinate physical meaning mapping model to complete the fusion of multi-modal fusion meteorological data;

[0113] Build a physical meaning mapping model of original image coordinates and geographic area coordinates, and perform spatial matching of meteorological data numerical prediction results and geographic information and new energy installation information of the target province and county area to generate a multi-modal fusion meteorological data set;

[0114] S6: Based on the physical correlation analysis of meteorological elements and historical output data, build a new energy power generation power prediction model (i.e. hybrid prediction model) driven by fusion data and physical mechanism, and realize the prediction of the generation curve;

[0115] Analyze the physical correlation of meteorological elements and historical new energy output data, build a hybrid prediction model, fuse data-driven features and physical mechanism constraints, and output a new energy power generation power prediction curve;

[0116] S7: Compare and analyze the prediction results of day D and the actual output data on day D+1, and update and improve the evaluation of each meteorological source.

[0117] Compare and analyze the prediction results of day D and the actual output data on day D+1, update the meteorological source evaluation parameters and model weights based on error feedback, and complete the closed-loop optimization of the prediction system.

[0118] Among them, the output prediction coupled with data-driven and physical mechanism includes:

[0119] S51: Calculate the theoretical output power value using the corresponding provincial installed capacity and conversion coefficient;

[0120]

[0121] In the formula, is the region At the time The conversion coefficient is related to the new energy output start lower limit value; is the region The installed capacity.

[0122]

[0123] In the formula, is The new energy predicted output value at the moment.

[0124] Figure 3 The structure diagram of a multi-source new energy output prediction device provided by an embodiment of the application is shown. As Figure 3 shown, the device includes:

[0125] The dynamic error weight parameter acquisition module is configured to collect historical meteorological data of the same time period in the same region by multiple meteorological sources, and obtain dynamic error weight parameters of each meteorological source in different time periods by using the historical meteorological data of each meteorological source;

[0126] The optimal meteorological source combination module is configured to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources by using the dynamic error weight parameters of each meteorological source in different time periods, and obtain an optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of multiple meteorological sources;

[0127] The extraction module is configured to obtain meteorological pictures from the corresponding meteorological sources in each time period of the optimal meteorological source combination, and extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture by using a color scale feature recognition technology;

[0128] The spatial matching module is configured to obtain a multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination by spatially matching the meteorological element image coordinates in each time period of the optimal meteorological source combination with target geographical region coordinates and new energy installed information;

[0129] The output prediction module is configured to determine a mixed prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination, respectively, and obtain the output prediction result of the multi-source new energy by inputting the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination into the corresponding mixed prediction model.

[0130] The above is a schematic scheme of the multi-source new energy output prediction device of the embodiment. It should be noted that the technical scheme of the multi-source new energy output prediction device and the technical scheme of the multi-source new energy output prediction method described above belong to the same concept, and the details of the technical scheme of the multi-source new energy output prediction device that are not described in detail can be referred to the description of the technical scheme of the multi-source new energy output prediction method.

[0131] Figure 4 A structural block diagram of a computing device 400 according to an embodiment of the present application is shown. The components of the computing device 400 include, but are not limited to, a memory 410 and a processor 420. The processor 420 is connected to the memory 410 through a bus 430, and a database 450 is used to save data.

[0132] The computing device 400 also includes an access device 440 that enables the computing device 400 to communicate via one or more networks 460. Examples of these networks include the public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 440 can include one or more of any type of network interface (e.g., network interface cards (NICs)) such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like, either wired or wireless.

[0133] In an embodiment of the present application, the above-mentioned components of the computing device 400 and other components not shown in the above-mentioned components can be connected to each other, for example, through a bus. It should be understood that Figure 4 the computing device structure block diagram shown is only for the purpose of example, and is not a limitation on the scope of the present application. Those skilled in the art can add or replace other components as needed. Figure 4

[0134] The computing device 400 can be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smartphone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other type of mobile device, or a stationary computing device such as a desktop computer or a PC. The computing device 400 can also be a mobile or stationary server.

[0135] The processor 420 is configured to execute computer-executable instructions that, when executed by the processor, implement the steps of the multi-source new energy output prediction method described above.​

[0136] The above is a schematic scheme of the computing device of the embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the multi-source new energy output prediction method described above belong to the same concept, and the details of the technical scheme of the computing device that are not described in detail can be referred to the description of the technical scheme of the multi-source new energy output prediction method.

[0137] An embodiment of the present application further provides a computer readable storage medium, which stores computer executable instructions, and the computer executable instructions are executed by a processor to implement the steps of the multi-source new energy output prediction method.

[0138] The above is a schematic scheme of the computer readable storage medium of the embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the multi-source new energy output prediction method described above belong to the same concept, and the details of the technical scheme of the storage medium that are not described in detail can be referred to the description of the technical scheme of the multi-source new energy output prediction method.

[0139] An embodiment of the present application further provides a computer program, and when the computer program is executed in a computer, the computer program causes the computer to execute the steps of the multi-source new energy output prediction method.

[0140] The above is a schematic scheme of the computer program of the embodiment. It should be noted that the technical scheme of the computer program and the technical scheme of the multi-source new energy output prediction method described above belong to the same concept, and the details of the technical scheme of the computer program that are not described in detail can be referred to the description of the technical scheme of the multi-source new energy output prediction method.

[0141] The specific embodiments of the present application are described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order and still achieve desirable results. Additionally, the process depicted in the figures does not necessarily require the particular order shown, or sequential order to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0142] The computer instructions include computer program code, which may be in the form of source code, object code, executable file, or certain intermediate forms. The computer-readable medium may include any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium may be appropriately added to or subtracted according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media may not include electrical carrier signals and telecommunication signals.

[0143] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the embodiments of this application.

[0144] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0145] The preferred embodiments disclosed above are merely illustrative of this application. The optional embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the embodiments of this application. These embodiments are selected and specifically described in this application to better explain the principles and practical applications of the embodiments of this application, thereby enabling those skilled in the art to better understand and utilize this application. This application is limited only by the claims and their full scope and equivalents.

Claims

1. A method for multi-source new energy output prediction, characterized in that, The application relates to a method for constructing a multi-source meteorological data set for a new energy power plant. The method comprises the following steps: collecting historical meteorological data of the same region and the same time period from multiple meteorological sources, and using the historical meteorological data of each meteorological source to obtain a dynamic error weight parameter of each meteorological source in different time periods, which comprises the following steps: using the predicted values and the measured values of each meteorological element data in the historical meteorological data of each meteorological source in different time periods to construct a dynamic confidence interval of each meteorological source in different time periods; using the dynamic confidence interval of each meteorological source in different time periods to quantitatively evaluate the error of each meteorological source and obtain a dynamic error weight parameter of each meteorological source in different time periods; using the dynamic error weight parameter of each meteorological source in different time periods to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources, and obtaining an optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources; the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources comprises the following steps: obtaining meteorological pictures from the corresponding meteorological source in each time period of the optimal meteorological source combination, and using a color scale feature recognition technology to extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture; performing spatial matching on the meteorological element image coordinates in each time period of the optimal meteorological source combination, target geographical region coordinates and new energy installation information to obtain a multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination; determining a hybrid prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination, and inputting the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination into the corresponding hybrid prediction model to obtain an output prediction result of a multi-source new energy. The method comprises the following steps: using the predicted values and the measured values of each meteorological element data in the historical meteorological data of each meteorological source in different time periods to construct a dynamic confidence interval of each meteorological source in different time periods; using the dynamic confidence interval of each meteorological source in different time periods to quantitatively evaluate the error of each meteorological source and obtain a dynamic error weight parameter of each meteorological source in different time periods; using the dynamic error weight parameter of each meteorological source in different time periods to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources, and obtaining an optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources; the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources comprises the following steps: obtaining meteorological pictures from the corresponding meteorological source in each time period of the optimal meteorological source combination, and using a color scale feature recognition technology to extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture; performing spatial matching on the meteorological element image coordinates in each time period of the optimal meteorological source combination, target geographical region coordinates and new energy installation information to obtain a multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination; determining a hybrid prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination, and inputting the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination into the corresponding hybrid prediction model to obtain an output prediction result of a multi-source new energy. ； In the formula, Indicates the t-th time period. The activation status of the j-th historical meteorological data in a meteorological source; , , The weighting coefficients and ; Indicates the t-th time period. Dynamic error weighting parameters for the j-th historical meteorological data from a meteorological source; Indicates the t-th time period. The actual coverage rate of the j-th historical meteorological data in a meteorological source; They represent the t-th time period and the th time period respectively. The mean and variance of the j-th historical meteorological data from _ _ meteorological sources; The method comprises the following steps: using the predicted values and the measured values of each meteorological element data in the historical meteorological data of each meteorological source in different time periods to construct a dynamic confidence interval of each meteorological source in different time periods; using the dynamic confidence interval of each meteorological source in different time periods to quantitatively evaluate the error of each meteorological source and obtain a dynamic error weight parameter of each meteorological source in different time periods; using the dynamic error weight parameter of each meteorological source in different time periods to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources, and obtaining an optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources; the AI reinforcement learning multi-objective optimization model of the multiple meteorological sources comprises the following steps: obtaining meteorological pictures from the corresponding meteorological source in each time period of the optimal meteorological source combination, and using a color scale feature recognition technology to extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture; performing spatial matching on the meteorological element image coordinates in each time period of the optimal meteorological source combination, target geographical region coordinates and new energy installation information to obtain a multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination; determining a hybrid prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination, and inputting the multi-modal fusion meteorological data set in each time period of the optimal meteorological source combination into the corresponding hybrid prediction model to obtain an output prediction result of a multi-source new energy. ​ ​ 2. The method of claim 1, wherein, ​ ​ ​ ​ 3. The method of claim 2, wherein, ​ ​ According to the number of times that the historical prediction error of each meteorological source in different time periods falls within the dynamic confidence interval of the corresponding time period, the actual coverage rate of each meteorological source is calculated, and based on the actual coverage rate of each meteorological source, a dynamic error weight parameter of each meteorological source in different time periods is obtained.

4. The method of claim 3, wherein, The color step feature recognition technology is used to extract meteorological element information, meteorological element image coordinates and meteorological element color step parameters from each meteorological picture, which includes: Divide the color step interval for representing the RGB or HSV color value range of each color step, and establish a mapping function between the color step and the meteorological element based on the color step interval; According to the mapping function between the color step and the meteorological element, meteorological element information and meteorological element color step parameters are extracted from each meteorological picture; The color step feature recognition technology is used to extract the coordinates within each color step block from each meteorological picture, and by defining the original image coordinate reference point of each meteorological picture, a coordinate data set of each pixel point in each color step block is generated, and the coordinate data set of each pixel point in each color step block is taken as the meteorological element image coordinates.

5. The method of claim 4, wherein, The spatial matching of the meteorological element image coordinates of the optimal meteorological source combination in each time period with the target geographical area coordinates and new energy installed capacity information is performed to obtain a multi-modal fusion meteorological data set of the optimal meteorological source combination in each time period, which includes: Define the conversion relationship between the meteorological element image coordinate system of each meteorological picture and the target geographical area coordinate system; Through the conversion relationship of each meteorological picture, the meteorological element image coordinates of each meteorological picture are spatially matched and positioned with the corresponding target geographical area coordinates to obtain the target geographical area coordinates of each meteorological picture; Through the fusion processing of the meteorological element information, meteorological element color step parameters, meteorological element image coordinates and target geographical area coordinates of each meteorological picture, a multi-modal fusion meteorological data set of the optimal meteorological source combination in each time period is obtained.

6. A multi-source new energy output prediction device, characterized in that, It includes: The dynamic error weight parameter acquisition module is configured to collect historical meteorological data of multiple meteorological sources in the same time period in the same area, and obtain dynamic error weight parameters of each meteorological source in different time periods based on the historical meteorological data of each meteorological source, which includes: using the predicted values and measured values of each meteorological element data in different time periods contained in the historical meteorological data of each meteorological source, constructing dynamic confidence intervals of each meteorological source in different time periods; using the dynamic confidence intervals of each meteorological source in different time periods, quantitatively evaluating the error of each meteorological source to obtain dynamic error weight parameters of each meteorological source in different time periods; The optimal meteorological source combination acquisition module is configured to construct an AI reinforcement learning multi-objective optimization model of multiple meteorological sources by using the dynamic error weight parameters of each meteorological source in different time periods, and obtain the optimal meteorological source combination in each time period by solving the AI reinforcement learning multi-objective optimization model of multiple meteorological sources; the AI reinforcement learning multi-objective optimization model of multiple meteorological sources includes: ； In the formula, Indicates the t-th time period. The activation status of the j-th historical meteorological data in a meteorological source; , , The weighting coefficients and ; Indicates the t-th time period. Dynamic error weighting parameters for the j-th historical meteorological data from a meteorological source; Indicates the t-th time period. The actual coverage rate of the j-th historical meteorological data in a meteorological source; They represent the t-th time period and the th time period respectively. The mean and variance of the j-th historical meteorological data from _ _ meteorological sources; The extraction module is configured to obtain meteorological pictures from the corresponding meteorological source under each time period of the optimal meteorological source combination, and extract meteorological element information, meteorological element image coordinates and meteorological element color scale parameters from each meteorological picture by using a color scale feature recognition technology. The space matching module is configured to obtain a multi-modal fusion meteorological data set under each time period of the optimal meteorological source combination by performing space matching between the meteorological element image coordinates under each time period of the optimal meteorological source combination and target geographical region coordinates and new energy installation information. The output prediction module is configured to determine a hybrid prediction model corresponding to the meteorological element information in the multi-modal fusion meteorological data set under each time period of the optimal meteorological source combination, respectively, and obtain a new energy output prediction result by inputting the multi-modal fusion meteorological data set under each time period of the optimal meteorological source combination into the corresponding hybrid prediction model.

7. A computing device comprising: a memory and a processor; the memory is configured to store computer executable instructions, and the processor is configured to execute the computer executable instructions, and the computer executable instructions, when executed by the processor, implement the steps of the multi-source new energy output prediction method of any one of claims 1 to 5.

8. A computer readable storage medium storing computer executable instructions, and the computer executable instructions, when executed by a processor, implement the steps of the multi-source new energy output prediction method of any one of claims 1 to 5.

Images