Medium and long term power prediction method based on era5 reanalysis data and related equipment

Abstract

This invention belongs to the field of medium- and long-term power generation in new energy, and discloses a medium- and long-term power forecasting method and related equipment based on ERA5 reanalysis data. This method uses multi-year ERA5 reanalysis data to perform climate model analysis on the area where the power plant is located, obtains the climatological average irradiance, and constructs an error correction model by combining it with historical measured data from the power plant. This allows for medium- and long-term power forecasting using historical data from the same period of the power plant's climatological output. By using multi-year ERA5 reanalysis data to perform climate model analysis on the area where the power plant is located, this method can accurately capture the climatological average irradiance of the area, effectively solving the forecasting problem caused by the lack of historical data for newly built or short-term operating photovoltaic power plants. Using this method is of crucial practical significance for power plants to reasonably declare medium- and long-term trading volumes in electricity transactions and reduce revenue losses caused by positive and negative volume deviations, significantly enhancing the competitiveness of power plants in the electricity market.

Description

Technical Field

[0001] This invention belongs to the field of medium- and long-term power generation prediction technology for new energy sources, and particularly relates to a medium- and long-term power generation prediction method and related equipment based on ERA5 reanalysis data. Background Technology

[0002] Against the backdrop of increasingly market-oriented and liberalized electricity trading, the operating mechanisms of the electricity market are becoming more complex and volatile. Electricity trading is no longer limited to traditional direct transactions between supply and demand parties or government regulation models, but is gradually shifting towards a diversified and highly competitive trading platform. On this platform, power generation companies, electricity retailers, large users, and various electricity traders can participate in electricity buying and selling through various methods such as bilateral negotiation, centralized bidding, and listing transactions. In this market environment, power generation plants, as a crucial link in electricity production, are closely related to the accuracy of their operational strategies and power generation forecasts. Medium- and long-term power generation forecasts not only concern the power plants' power generation plans and resource allocation, but also directly affect their competitiveness and profitability in the electricity trading market.

[0003] Due to frequent price fluctuations and supply-demand changes in the electricity market, power plants need to formulate reasonable medium- and long-term electricity trading declaration plans based on accurate electricity forecasts to ensure a favorable position in electricity transactions. However, electricity forecasting faces many challenges: on the one hand, electricity load is affected by various factors such as weather, holidays, and economic activities, exhibiting high uncertainty and complexity; on the other hand, for newly built or recently commissioned photovoltaic power plants and other new power generation facilities, traditional statistical forecasting methods are often ineffective due to a lack of sufficient historical data. To address these issues, before obtaining historical meteorological data for each power plant, it is generally necessary to first analyze the spatial geographical location characteristics of each plant. This involves using the latitude and longitude location information of the plant as spatial sampling points to locate the meteorological grid points where each plant is located, thereby obtaining the most relevant meteorological data. However, due to the large number of plants involved, their wide distribution, the large latitude and longitude span, and the inconsistent historical data obtained, there is a certain degree of missing measured data, which limits the accuracy of the forecast results.

[0004] It is evident that existing medium- and long-term power generation forecasting methods suffer from a certain degree of missing measured data due to the large number and wide distribution of power stations involved, the large latitude and longitude span, and the inconsistent historical data obtained. This results in limited accuracy of the forecasting results. Summary of the Invention

[0005] The purpose of this invention is to provide a medium- and long-term power generation forecasting method and related equipment based on ERA5 reanalysis data, in order to solve the technical problem of limited accuracy in existing medium- and long-term power generation forecasting methods due to the large number of power stations involved, their wide distribution, large latitude and longitude span, and the inconsistent historical data obtained, which leads to a certain degree of missing measured data.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: A medium- to long-term power generation forecasting method based on ERA5 reanalysis data includes: Climate model analysis of the area where the station is located was performed using multi-year ERA5 reanalysis data to obtain the climatological mean irradiance of the area where the station is located. An error correction model for climatological and measured data is constructed based on climatological average irradiance and historical measured data from the site. The historical data of the acquired climatological output of the power station are input into the error correction model between the climatological and measured data, and the medium- and long-term power generation prediction results are output.

[0007] Furthermore, the specific steps for using multi-year ERA5 reanalysis data to perform climate model analysis on the area where the site is located, and obtaining the corresponding climatological mean irradiance for the area where the site is located, are as follows: Daily or hourly meteorological data from multiple years are obtained from the ERA5 database as ERA5 reanalysis data, including irradiance, temperature, humidity and wind speed; Based on the geographical location of the site, the relevant areas in the ERA5 reanalysis data are divided; The ERA5 reanalysis data within the divided regions were integrated to calculate the multi-year climatological average irradiance.

[0008] Furthermore, the specific steps for constructing the error correction model between climatological and measured data based on climatological average irradiance and historical measured data from the station are as follows: The historical measured data from the site were matched with the irradiation data in the ERA5 reanalysis data to ensure consistency in time scale and data points; Calculate the error between climatological mean irradiance and measured data; Based on the error analysis results, a linear regression model was selected to construct an error correction model between climatological and measured data.

[0009] Furthermore, the specific steps for inputting the historical data of the acquired station climatological output into the error correction model between the climatological and measured data, and outputting the medium- and long-term power prediction results are as follows: Collect historical contemporaneous data of the station's climatological output, including climatological mean irradiance data and measured irradiance data; Historical data collected from the same period are input into the error correction model between climatological and measured data to correct the climatological average irradiance data and obtain medium- and long-term electricity forecast results.

[0010] Furthermore, after collecting historical data from the same period, the historical data from the same period is normalized.

[0011] Furthermore, it also includes: correcting the medium- and long-term electricity forecast results by introducing climate factors.

[0012] Furthermore, the specific steps for correcting the medium- and long-term electricity forecast results by introducing climate factors are as follows: Obtain medium- and long-term electricity forecast results and corresponding climate factors; The climate factor correction model is pre-constructed using the medium- and long-term electricity forecast results and the corresponding climate factor input values, and outputs the medium- and long-term electricity correction results; wherein, the climate factor correction model is constructed based on historical climate factors and short-term climate forecast datasets.

[0013] A medium- to long-term power forecasting system based on ERA5 reanalysis data includes: The climatological mean irradiance acquisition module is used to perform climate model analysis on the area where the station is located using multi-year ERA5 reanalysis data to obtain the climatological mean irradiance corresponding to the area where the station is located. The error correction model construction module is used to construct an error correction model for climatological and measured data based on climatological average irradiance and historical measured data from the station. The medium- and long-term power generation forecast module is used to input the historical data of the same period from the acquired climatological output of the power station into the error correction model between the climatological and measured data, and output the medium- and long-term power generation forecast results.

[0014] An apparatus comprising: Memory, used to store computer programs; A processor is used to implement the steps of the above-described medium- and long-term power prediction method based on ERA5 reanalysis data when executing the computer program.

[0015] A computer-readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of the above-described medium- and long-term power prediction method based on ERA5 reanalysis data.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a medium- to long-term power generation forecasting method based on ERA5 reanalysis data. This method uses multi-year ERA5 reanalysis data to perform climate model analysis on the region where the power plant is located, obtaining the climatological average irradiance. It then combines this with historical measured data from the power plant to construct an error correction model, enabling medium- to long-term power generation forecasting using historical data from the same period of the climatological output. By using multi-year ERA5 reanalysis data to perform climate model analysis on the region where the power plant is located, this method can accurately capture the climatological average irradiance of the region, effectively solving the forecasting problem caused by the lack of historical data for newly built or short-term operating photovoltaic power plants. Using this method is of crucial practical significance for power plants to reasonably declare medium- to long-term trading volumes in electricity transactions and reduce revenue losses caused by positive and negative volume deviations, significantly enhancing the competitiveness of power plants in the electricity market.

[0017] Preferably, in this invention, daily or hourly meteorological data over many years are obtained from the ERA5 database as ERA5 reanalysis data. Based on the geographical location of the station, the relevant areas in the ERA5 reanalysis data are divided, and the ERA5 reanalysis data within the divided areas are integrated to obtain the multi-year climatological average irradiance. The above steps ensure the comprehensiveness and accuracy of the data, providing a reliable foundation for the subsequent construction of error correction models.

[0018] Preferably, in this invention, by matching historical measured data with irradiance data in ERA5 reanalysis data and calculating the error between climatological average irradiance and measured data, an error correction model for climatological and measured data is constructed. This can correct the difference between climatological average irradiance and actual irradiance, and improve the accuracy of prediction.

[0019] Preferably, in this invention, historical data from the same period are input into the error correction model to output medium- and long-term power generation prediction results. This allows for accurate prediction of medium- and long-term power generation using existing climate and measured data, providing strong support for the operation and maintenance of photovoltaic power plants.

[0020] Preferably, in this invention, normalization processing after collecting historical data from the same period can eliminate dimensional differences between data points, improving the model's generalization ability and prediction accuracy. This step is crucial for ensuring the stability and reliability of the prediction results.

[0021] Preferably, in this invention, the accuracy of the prediction can be further improved by introducing climate factors to correct the medium- and long-term electricity forecast results; climate factors reflect the trends and patterns of climate change and play an important role in adjusting and optimizing the prediction results.

[0022] More preferably, in this invention, the introduction of climate factors can comprehensively consider the impact of climate change on the power generation of photovoltaic power plants, thereby obtaining more accurate and reliable prediction results. In addition, the climate factor correction model adopts a machine learning model, and its construction also relies on historical climate factors and short-term climate prediction datasets, which further enhances the scientificity and practicality of the prediction method. Attached Figure Description

[0023] Figure 1 A flowchart of a medium- to long-term power prediction method based on ERA5 reanalysis data provided by the present invention; Figure 2 A schematic diagram of the structure of a medium- and long-term power prediction system based on ERA5 reanalysis data provided by the present invention; Figure 3 A flowchart of a medium- to long-term power prediction method based on ERA5 reanalysis data is provided for an embodiment of the present invention. Detailed Implementation

[0024] Example 1 This embodiment provides a medium- to long-term power forecasting method based on ERA5 reanalysis data, including the following steps: S1: Use multi-year ERA5 reanalysis data to perform climate model analysis on the area where the station is located to obtain the climatological mean irradiance corresponding to the area where the station is located. The specific steps are as follows: Daily or hourly meteorological data from multiple years are obtained from the ERA5 database as ERA5 reanalysis data, including irradiance, temperature, humidity and wind speed; The above ERA5 reanalysis data were preprocessed, including normalization and outlier removal. Based on the geographical location of the site, the relevant areas in the ERA5 reanalysis data are divided; The ERA5 reanalysis data within the divided regions were integrated to calculate the multi-year climatological average irradiance.

[0025] S2: Construct an error correction model for climatological and measured data based on climatological average irradiance and historical measured data from the station; The specific steps are as follows: The historical measured data from the site were matched with the irradiation data in the ERA5 reanalysis data to ensure consistency in time scale and data points; Calculate the error between climatological mean irradiance and measured data; Based on the error analysis results, a linear regression model was selected to construct an error correction model between climatological and measured data.

[0026] S3: Input the historical data of the same period of the acquired station climate-state output into the error correction model between climate-state and measured data, and output the medium- and long-term power generation prediction results.

[0027] The specific steps for outputting medium- and long-term electricity forecast results are as follows: Historical data from the same period of the climatological output of the station were collected. The historical data included climatological mean irradiance data and measured irradiance data. After collecting the historical data, the historical data were normalized.

[0028] Historical data collected from the same period are input into the error correction model between climatological and measured data to correct the climatological average irradiance data and obtain medium- and long-term electricity forecast results.

[0029] This also includes: correcting medium- and long-term electricity forecasts by incorporating climate factors; the specific steps are as follows: Obtain medium- and long-term electricity forecast results and corresponding climate factors; The climate factor correction model is pre-constructed using the medium- and long-term electricity forecast results and the corresponding climate factor input values, and outputs the medium- and long-term electricity correction results; wherein, the climate factor correction model is constructed based on historical climate factors and short-term climate forecast datasets.

[0030] like Figure 2 As shown in the figure, this embodiment also provides a medium- and long-term power forecasting system based on ERA5 reanalysis data, including: a climatological average irradiance acquisition module, used to perform climate model analysis on the area where the power station is located using multi-year ERA5 reanalysis data to obtain the climatological average irradiance corresponding to the area where the power station is located; an error correction model construction module, used to construct an error correction model between the climatological average irradiance and the historical measured data of the power station; and a medium- and long-term power forecasting module, used to input the historical data of the same period of the acquired climatological output of the power station into the error correction model between the climatological and measured data, and output the medium- and long-term power forecasting results.

[0031] The present invention also provides an apparatus comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the medium- and long-term power prediction method based on ERA5 reanalysis data.

[0032] When the processor executes the computer program, it implements the steps of medium- and long-term power prediction based on ERA5 reanalysis data, such as: using multi-year ERA5 reanalysis data to perform climate model analysis on the area where the power station is located to obtain the climatological average irradiance corresponding to the area where the power station is located; constructing an error correction model between the climatological average irradiance and the historical measured data of the power station; inputting the historical data of the same period of the obtained climatological output of the power station into the error correction model between the climatological and measured data, and outputting the medium- and long-term power prediction results.

[0033] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above system. For example: a climatological average irradiance acquisition module is used to perform climate model analysis on the area where the station is located using multi-year ERA5 reanalysis data to obtain the climatological average irradiance corresponding to the area where the station is located; an error correction model construction module is used to construct an error correction model between the climatological average irradiance and the historical measured data of the station; and a medium- and long-term power prediction module is used to input the historical data of the same period of the acquired climatological output of the station into the error correction model between the climatological and measured data, and output the medium- and long-term power prediction results.

[0034] For example, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing preset functions, wherein the instruction segments describe the execution process of the computer program in the medium- and long-term power prediction device based on ERA5 reanalysis data. For example, the computer program can be divided into a climatological average irradiance acquisition module, an error correction model construction module, and a medium- and long-term power prediction module; the specific functions of each module are as follows: the climatological average irradiance acquisition module is used to perform climate model analysis on the area where the station is located using multi-year ERA5 reanalysis data to obtain the climatological average irradiance corresponding to the area where the station is located; the error correction model construction module is used to construct an error correction model between the climatological average irradiance and the historical measured data of the station; the medium- and long-term power prediction module is used to input the historical contemporaneous data of the acquired climatological output of the station into the error correction model between the climatological average irradiance and the measured data, and output the medium- and long-term power prediction results.

[0035] The medium- to long-term power forecasting device based on ERA5 reanalysis data can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This device may include, but is not limited to, processors and memory. Those skilled in the art will understand that the above examples of medium- to long-term power forecasting devices based on ERA5 reanalysis data do not constitute a limitation on such devices. They may include more components than described above, or combine certain components, or use different components. For example, the medium- to long-term power forecasting device based on ERA5 reanalysis data may also include input / output devices, network access devices, buses, etc.

[0036] The processor referred to can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or any conventional processor. This processor is the control center for the medium- and long-term power forecasting based on ERA5 reanalysis data, connecting various parts of the entire medium- and long-term power forecasting device based on ERA5 reanalysis data through various interfaces and lines.

[0037] The memory can be used to store the computer program and / or modules. The processor implements various functions of the medium- and long-term power forecasting device based on ERA5 reanalysis data by running or executing the computer program and / or modules stored in the memory and calling the data stored in the memory.

[0038] The memory may primarily include a program storage area and a data storage area. The program storage area may store the operating system and at least one application program required for a function (such as sound playback, image playback, etc.). The data storage area may store data created based on the use of the mobile phone (such as audio data, phonebook, etc.). Furthermore, the memory may include high-speed random access memory and non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital cards (SD cards), flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0039] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the medium- to long-term power prediction method based on ERA5 reanalysis data.

[0040] If the modules / units integrated in the medium- and long-term power forecasting system based on ERA5 reanalysis data are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0041] Based on this understanding, the present invention can implement all or part of the processes in the aforementioned medium- and long-term power forecasting method based on ERA5 reanalysis data. This can also be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of the aforementioned medium- and long-term power forecasting method based on ERA5 reanalysis data. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or a preset intermediate form, etc.

[0042] The computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0043] It should be noted that the content contained in the computer-readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0044] The present invention will be further described below with reference to embodiments and accompanying drawings: Example 2 As described in the background section, power load forecasting faces numerous challenges: on the one hand, power load is affected by various factors such as weather, holidays, and economic activities, exhibiting high uncertainty and complexity; on the other hand, for newly built or recently commissioned photovoltaic power plants and other new power generation facilities, traditional statistical forecasting methods often fail to be effective due to a lack of sufficient historical data. To address these issues, before obtaining historical meteorological data for each station, it is generally necessary to first analyze the spatial geographical location characteristics of each station. This involves using the latitude and longitude location information of the stations as spatial sampling points to locate the meteorological grid points where each station is situated, thereby obtaining the most relevant meteorological data. However, due to the large number and wide distribution of the stations involved, the significant latitude and longitude span, and the varying historical data obtained, there is a certain degree of missing measured data, which limits the accuracy of the forecast results.

[0045] To achieve the above objectives, this invention provides a medium- to long-term power generation forecasting method based on ERA5 reanalysis data. This method uses multi-year ERA5 reanalysis data to perform climate model analysis on the area where the power station is located, which can accurately capture the climatological average irradiance of the area. It also cleverly combines the historical measured data of the power station to construct an error correction model between the climatological and measured data, further improving the accuracy of the forecast.

[0046] The medium- to long-term power generation forecasting method based on ERA5 reanalysis data provided in this embodiment is based on the idea that, before obtaining historical meteorological data for each station involved, a spatial geographical location characteristic analysis of each station is first performed. This is done by using the latitude and longitude location information of the stations as spatial sampling points to locate the meteorological grid points where each station is situated, thereby obtaining the most relevant meteorological data. However, due to the large number of stations involved, their wide distribution, the large latitude and longitude span, and the inconsistencies in the obtained historical data, there is a certain degree of missing measured data. Therefore, after dividing the overall area of ​​all station distribution into grids, for stations lacking historical data, spatial sampling points are added at the center point of their respective unit grid. ERA5 historical reanalysis meteorological data is then used as an approximation of the historical measured data to proceed with the next step of analysis and judgment.

[0047] This embodiment provides a medium- to long-term power generation forecasting method based on ERA5 reanalysis data. This method deeply integrates advanced climate data analysis technology with actual power production needs, providing power market participants with a more accurate and reliable forecasting tool. The core of this method lies in utilizing years of accumulated ERA5 reanalysis data, which covers daily or hourly meteorological information worldwide, including but not limited to key parameters such as irradiance, temperature, humidity, and wind speed, providing a rich data source for power generation forecasting. The specific steps include: (1) Analysis of station spatial characteristics First, through in-depth analysis of multi-year meteorological data in the ERA5 database, detailed climate model analysis was conducted for the specific regions where each station is located. This step not only requires accurate identification of the station's geographical location but also precise subdivision of the relevant areas within the ERA5 data to ensure the data's relevance and accuracy. Subsequently, the ERA5 data from these regions were integrated, and through a complex calculation process, the climatological mean irradiance of the station's region was derived, which is one of the fundamental inputs for the prediction model.

[0048] (2) Establish historical average climate state like Figure 3 As shown, the meteorological forecast mainly starts from the global climate system that affects each station, based on ERA5 reanalysis data, and combines the station's measured data and the main common short-term climate characteristics of each season to predict whether the annual radiation is higher or lower for all the stations involved.

[0049] like Figure 3 As shown, the first step is to analyze and calculate the multi-year climatological mean irradiance of the area where the site is located based on the climate model of multi-year ERA5 reanalysis data.

[0050] The second step is to establish an error correction model between the climate state and the measured data for the power stations that provide historical measured data. Based on this model, statistical regression correction is performed on the historical data of the same period output by the power station's climate state to obtain the preliminary medium- and long-term power generation forecast results. For power stations that do not have historical measured data, the average results of the historical climate state are used as the basis for medium- and long-term power generation forecasts.

[0051] In other words, to overcome the limitations of traditional forecasting methods for newly built or short-term operational sites, this step innovatively constructs an error correction model between climatological and measured data. This model first ensures the consistency of historical measured data and ERA5 irradiance data in terms of time scale and data points. Through precise matching, it calculates the error between the climatological mean irradiance and the measured data. Based on these error analyses, a linear regression model was chosen as the core algorithm for error correction because it can effectively capture the linear relationship between data and is easy to implement and calculate.

[0052] During the prediction phase, historical data on the climatological output of the field stations were collected. This data included both climatological mean irradiance and measured irradiance, providing rich material for the model input. To improve prediction accuracy, these historical data were normalized to eliminate the influence of different units and numerical ranges on model performance.

[0053] (3) Correction of electricity prediction results based on climate factor analysis Step 1: Analysis of the impact of multiple climatic factors on radiation Climate change assessments play a crucial role in analyzing the evolution and impacts of regional climate resources and meteorological disasters. As the butterfly effect theory suggests, "a butterfly flapping its wings in the Amazon rainforest of South America can, two weeks later, cause a tornado in Texas." In the complex models of meteorological systems, errors can grow exponentially, meaning that even a small error can have enormous consequences over time. Therefore, the longer the forecast period, the more important it is to consider the impacts of climate processes.

[0054] Therefore, when considering the impact of changes in meteorological elements in other regions of the world on their own region, the concept of "climate factors" is used to distinguish different types of climate change processes.

[0055] The second step is to revise the prediction model based on climate factors. In this embodiment, a multi-model prediction scheme is established based on short-term climate prediction (C3S) and historical climate factor datasets, using machine learning models to model the main climate factors affecting the affected area and regional meteorological anomalies. Climate states primarily represent historical average patterns, but changes in climate factors occurring during the same historical period often take effect after a certain time lag. Introducing research on the impact of climate factors effectively optimizes the capture of future meteorological anomaly trends.

[0056] In other words, in this embodiment, considering the significant impact of climate factors on power generation forecasting, a climate factor correction step is further introduced. This step first collects medium- and long-term power generation forecasts and their corresponding climate factors, and then inputs this information into a pre-built climate factor correction model. This correction model is trained based on historical climate factors and short-term climate forecast datasets, and can dynamically adjust the forecast results to reflect the potential impacts of future climate change.

[0057] In addition, to obtain more accurate medium- and long-term electricity forecast results, this embodiment also includes the following steps: (4) Time series-based prediction of power curtailment at power plants Besides incorporating future weather forecasts to optimize medium- and long-term power generation prediction methods, factors such as power rationing also affect the final actual power generation. Therefore, for areas with severe power rationing, it is crucial to focus on methods for identifying abnormal power rationing data. Semi-supervised learning is a machine learning technique suitable for situations where some data is labeled and some is unlabeled. It not only significantly reduces manual labeling work but also makes full use of the data, resulting in higher accuracy. Furthermore, by introducing active learning and abnormal data identification and reconstruction techniques, the most informative or representative samples are selected for priority labeling. For unlabeled data, active learning techniques are used to label key and difficult-to-identify abnormal power rationing data. The labeling task for less important and easier-to-identify abnormal power rationing data is addressed using data filtering techniques.

[0058] Using the aforementioned labeled data, further machine learning algorithms are used to model the historical power rationing periods of the power station. Based on the project's own operational characteristics and the absorption characteristics of surrounding nodes, combined with multiple internationally leading meteorological forecast sources and self-developed meteorological forecasts, the power rationing periods and degrees of the power station are predicted, enabling medium- and long-term power rationing predictions for the power station and contributing to more accurate medium- and long-term power generation forecasts.

[0059] In summary, this invention provides a medium- to long-term power generation forecasting method based on ERA5 reanalysis data, which has the following advantages compared to existing medium- to long-term power generation forecasting methods: This method integrates and utilizes multi-year ERA5 reanalysis data to conduct precise climate model analysis of the power station's location, thereby deriving the corresponding climatological mean irradiance for the region. Furthermore, by combining historical measured data from the power station, an error correction model between the climatological and measured data is constructed, significantly improving the accuracy of the forecast. During the forecasting process, historical data from the same period output by the power station's climatological data are input into this error correction model, enabling the output of high-precision medium- and long-term power generation forecasts. In addition, this method includes normalization processing of historical data from the same period and climate factor correction of the medium- and long-term power generation forecasts, further enhancing the reliability and adaptability of the forecast results. This method not only improves the accuracy of power generation forecasts but also provides power market participants with a more scientific basis for decision-making, helping them to gain a favorable position in power trading and reduce economic losses caused by inaccurate power generation forecasts.

[0060] The above embodiments are merely one of the implementation methods for achieving the technical solution of the present invention. The scope of protection claimed by the present invention is not limited to this embodiment, but also includes any variations, substitutions and other implementation methods that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention.

Claims

1. A medium- to long-term power generation forecasting method based on ERA5 reanalysis data, characterized in that, include: Climate model analysis of the area where the station is located was performed using multi-year ERA5 reanalysis data to obtain the climatological mean irradiance of the area where the station is located. An error correction model for climatological and measured data is constructed based on climatological average irradiance and historical measured data from the site. The historical data of the acquired climatological output of the power station are input into the error correction model between the climatological and measured data, and the medium- and long-term power generation prediction results are output.

2. The medium- and long-term power forecasting method based on ERA5 reanalysis data according to claim 1, characterized in that, The specific steps for using multi-year ERA5 reanalysis data to perform climate model analysis on the area where the site is located, and to obtain the corresponding climatological mean irradiance for the area where the site is located, are as follows: Daily or hourly meteorological data from multiple years are obtained from the ERA5 database as ERA5 reanalysis data, including irradiance, temperature, humidity and wind speed; Based on the geographical location of the site, the relevant areas in the ERA5 reanalysis data are divided; The ERA5 reanalysis data within the divided regions were integrated to calculate the multi-year climatological average irradiance.

3. The medium- and long-term power forecasting method based on ERA5 reanalysis data according to claim 1, characterized in that, The specific steps for constructing the error correction model between climatological and measured data based on climatological average irradiance and historical measured data from the station are as follows: The historical measured data from the site were matched with the irradiation data in the ERA5 reanalysis data to ensure consistency in time scale and data points; Calculate the error between climatological mean irradiance and measured data; Based on the error analysis results, a linear regression model was selected to construct an error correction model between climatological and measured data.

4. The medium- and long-term power forecasting method based on ERA5 reanalysis data according to claim 1, characterized in that, The specific steps for inputting the historical data of the acquired station climatological output into the error correction model between climatological and measured data, and outputting the medium- and long-term power prediction results are as follows: Collect historical contemporaneous data of the station's climatological output, including climatological mean irradiance data and measured irradiance data; Historical data collected from the same period are input into the error correction model between climatological and measured data to correct the climatological average irradiance data and obtain medium- and long-term electricity forecast results.

5. The medium- and long-term power generation forecasting method based on ERA5 reanalysis data according to claim 4, characterized in that, After collecting historical data from the same period, the historical data was normalized.

6. The medium- and long-term power generation forecasting method based on ERA5 reanalysis data according to claim 1, characterized in that, Also includes: The medium- and long-term electricity forecast results were corrected by introducing climate factors.

7. The medium- and long-term power generation forecasting method based on ERA5 reanalysis data according to claim 1, characterized in that, The specific steps for correcting medium- and long-term electricity forecasts by introducing climate factors are as follows: Obtain medium- and long-term electricity forecast results and corresponding climate factors; The climate factor correction model is pre-constructed using the medium- and long-term electricity forecast results and the corresponding climate factor input values, and outputs the medium- and long-term electricity correction results; wherein, the climate factor correction model is constructed based on historical climate factors and short-term climate forecast datasets.

8. A medium- to long-term power forecasting system based on ERA5 reanalysis data, characterized in that, include: The climatological mean irradiance acquisition module is used to perform climate model analysis on the area where the station is located using multi-year ERA5 reanalysis data to obtain the climatological mean irradiance corresponding to the area where the station is located. The error correction model construction module is used to construct an error correction model for climatological and measured data based on climatological average irradiance and historical measured data from the station. The medium- and long-term power generation forecast module is used to input the historical data of the same period from the acquired climatological output of the power station into the error correction model between the climatological and measured data, and output the medium- and long-term power generation forecast results.

9. A device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the medium- to long-term power forecasting method based on ERA5 reanalysis data as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it is used to implement the steps of the medium- and long-term power prediction method based on ERA5 reanalysis data as described in any one of claims 1-7.

Images