Power system generation power prediction method and system based on multi-source meteorological data
By combining multi-source meteorological data and neural network models, the accuracy problem of power generation prediction in traditional power systems has been solved, achieving high-precision power generation prediction under complex weather conditions and supporting the stable operation and maintenance of power systems.
Patent Information
- Application Number
- CN202511050121.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-29
- Publication Date
- 2025-10-31
AI Technical Summary
Traditional power generation forecasting methods rely on single meteorological data, which makes it difficult to accurately reflect the impact of complex weather on the power generation of new energy power systems, leading to problems such as sudden load changes and sharp drops in power generation in grid dispatching.
Using multi-source meteorological data and trained neural network models, especially LSTM and Transformer models, combined with an adaptive weighted fusion algorithm, the power generation capacity of the power system is predicted.
It has improved the accuracy of power generation forecasting, provided effective reference information for power system operation and maintenance, and enhanced the ability to forecast power generation under complex weather conditions.
Smart Images
Figure CN120879561A_ABST
Abstract
Description
Technical Field
[0001] This application generally relates to the fields of power system operation and maintenance and artificial intelligence technology. More specifically, this application relates to a method, system, medium, and equipment for predicting power generation capacity of a power system based on multi-source meteorological data. Background Technology
[0002] With the large-scale grid connection of new energy (wind and solar) power systems, the power generation of the entire power system is significantly affected by meteorological conditions, exhibiting strong fluctuations and uncertainties. Traditional power generation forecasting methods mostly rely on single meteorological data sources, making it difficult to accurately reflect the impact of complex weather (such as extreme rainfall, strong winds, and sandstorms) on the power generation of new energy power systems. Simultaneously, grid dispatching faces problems such as sudden load changes and sharp drops in power generation from new energy power systems caused by meteorological factors. Therefore, a high-precision power generation forecasting solution for the power system is urgently needed. Summary of the Invention
[0003] In order to at least solve one or more of the technical problems mentioned above, this application proposes a method, system, medium and equipment for predicting power generation of power systems based on multi-source meteorological data in several aspects.
[0004] In the first aspect, the power generation prediction method for power systems based on multi-source meteorological data provided in this application includes the following steps:
[0005] Based on meteorological data generated from multiple meteorological forecasting models, meteorological data for each moment of the day is obtained to generate the first multi-source meteorological data.
[0006] First spatiotemporal feature data directly related to power generation are extracted from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0007] Input the first spatiotemporal feature data into the trained neural network model;
[0008] Based on the first spatiotemporal feature data and the preset time period, the trained neural network model continuously outputs the power generation power of the power system within the future time period.
[0009] In some examples, the training process of the trained neural network model includes:
[0010] Based on historical meteorological data collected from multiple weather forecasting systems, meteorological data for each moment within a preset historical period is obtained to generate a second multi-source meteorological data.
[0011] Extract the second spatiotemporal feature data that is directly related to power generation at each time point from the second multi-source meteorological data. The second spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0012] Under the second spatiotemporal feature data, the power generation power of the power system at each moment within a preset historical period is obtained respectively;
[0013] The second spatiotemporal feature data and the power generation power of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
[0014] In some examples, the training process of the trained neural network model further includes:
[0015] Under the second spatiotemporal feature data, the equipment status of the power system at each moment within a preset historical period is obtained respectively;
[0016] The second spatiotemporal feature data and the corresponding equipment status of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
[0017] In some examples, the trained neural network model includes both LSTM and Transformer models.
[0018] In some examples, based on the first spatiotemporal feature data and a preset time period, the trained neural network model outputs the power generation capacity of the power system within the future time period, including:
[0019] If the time period is less than a preset first threshold, the trained neural network model selects the LSTM model to calculate and continuously output the power generation of the power system in the future time period.
[0020] In some examples, based on the first spatiotemporal feature data and a preset time period, the trained neural network model outputs the power generation capacity of the power system within the future time period, which further includes:
[0021] If the time period is longer than the preset first threshold, the trained neural network model selects the Transformer model to calculate and continuously output the power generation of the power system in the future time period.
[0022] In some examples, after obtaining the meteorological data at the current moment based on meteorological data generated from multiple meteorological forecasting models, the method further includes:
[0023] An adaptive weighted fusion algorithm is used to fuse various types of meteorological data in the first multi-source meteorological data.
[0024] In the second aspect, the power generation prediction system for power systems based on multi-source meteorological data provided in this application includes:
[0025] The acquisition module is configured to acquire meteorological data at various times of the day based on meteorological data generated by multiple meteorological forecast models, and generate the first multi-source meteorological data.
[0026] The extraction module is configured to extract first spatiotemporal feature data directly related to power generation from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0027] The input module is configured to input the first spatiotemporal feature data into the trained neural network model;
[0028] The output module is configured to continuously output the power generation power of the power system within the future time period based on the first spatiotemporal feature data and a preset time period after the neural network model has been trained.
[0029] In a third aspect, the computer-readable storage medium provided in this application includes program instructions that, when executed by a processor, cause the method disclosed in the first aspect to be implemented.
[0030] In the fourth aspect, the electronic device provided in this application includes:
[0031] A processor and a memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method disclosed in the first aspect.
[0032] The method, system, medium, and equipment for predicting power generation capacity of power systems based on multi-source meteorological data provided in this application improve the accuracy of prediction by using multi-source meteorological data and a trained neural network model to predict the power generation capacity of power systems, and provide effective reference information for power system operation and maintenance. Attached Figure Description
[0033] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein:
[0034] Figure 1 This application shows an exemplary flowchart of a power system generation power prediction method based on multi-source meteorological data, representing some embodiments of the present application.
[0035] Figure 2 An exemplary structural block diagram of a power system generation power prediction system based on multi-source meteorological data according to some embodiments of this application is shown;
[0036] Figure 3 An exemplary structural block diagram of an electronic device according to some embodiments of this application is shown. Detailed Implementation
[0037] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0038] It should be understood that the terms "comprising" and "including" used in the specification and claims of this application indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.
[0039] It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application. As used in this specification and claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this specification and claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes such combinations.
[0040] As used in this specification and claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if [described condition or event] is detected" may be interpreted, depending on the context, as "once determined," "in response to determination," "once [described condition or event] is detected," or "in response to detection of [described condition or event]."
[0041] The specific embodiments of this application will now be described in detail with reference to the accompanying drawings.
[0042] Example 1
[0043] like Figure 1 As shown in the embodiments of this application, the power generation prediction method for power systems based on multi-source meteorological data includes the following steps:
[0044] S101: Based on meteorological data generated by multiple meteorological forecast models, acquire meteorological data for each moment of the day to generate the first multi-source meteorological data.
[0045] Specifically, the accuracy of the collected meteorological data was ensured by collecting meteorological data generated from multiple meteorological forecasting models.
[0046] In some examples, prior to step S101, the method further includes:
[0047] An adaptive weighted fusion algorithm is used to fuse various types of meteorological data in the first multi-source meteorological data.
[0048] Specifically, before using the adaptive weighted fusion algorithm to fuse various types of meteorological data in the first multi-source meteorological data, the following operations are performed on the first multi-source meteorological data:
[0049] (1) Data preprocessing (missing value imputation, outlier removal, normalization);
[0050] (2) Feature alignment (spatial / temporal scale unification);
[0051] (3) Dimensionality reduction and embedding of high-dimensional features.
[0052] Specifically, multiple weather forecasting models include the ECMWF weather forecasting model, the GFS weather forecasting model, and satellite remote sensing weather forecasting models.
[0053] Specifically, the basic principle of the adaptive weighted fusion algorithm is to achieve more accurate results by weighting data collected from different sensors. Specifically, the adaptive weighted fusion algorithm weights the data collected from different sensors based on the reliability and accuracy of the sensors. Reliability refers to the stability and accuracy of the sensor during operation, while accuracy refers to the degree of consistency between the sensor-collected data and the actual data. During the weighting process, the adaptive weighted fusion algorithm assigns a weight coefficient to each sensor based on its reliability and accuracy. This weight coefficient can be calculated using methods such as least squares, information entropy, and fuzzy logic. Different calculation methods will have different effects on the calculation of the weight coefficient, so it is necessary to choose an appropriate method according to the specific situation. Weighting methods include weighted average and Bayesian inference, etc. Different weighting methods have different effects on the accuracy of data fusion, so different weighting methods need to be selected according to the situation.
[0054] S102, extract first spatiotemporal feature data directly related to power generation from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0055] Specifically, power generation capacity measures a power plant's ability to supply electrical energy to the grid per unit of time, typically measured in megawatts (MW). Higher power generation capacity indicates higher power plant efficiency, meaning it can supply more electrical energy to the grid. The following discussion examines the impact of wind speed, wind direction, and solar radiation on power plants.
[0056] (1) The wind speed and wind direction directly affect the power generation of the power system (wind power station). Generally speaking, the greater the wind speed and the more the wind direction matches the location of the wind power station blades, the greater the power generation of the wind power station.
[0057] (2) The amount of solar radiation directly affects the power generation capacity of the power system (solar power station). The greater the amount of solar radiation, the greater the power generation capacity of the solar power station.
[0058] S103, input the first spatiotemporal feature data into the trained neural network model.
[0059] In some examples, the training process of the trained neural network model includes:
[0060] Based on historical meteorological data collected from multiple weather forecasting systems, meteorological data for each moment within a preset historical period is obtained to generate a second multi-source meteorological data.
[0061] Extract the second spatiotemporal feature data that is directly related to power generation at each time point from the second multi-source meteorological data. The second spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0062] Under the second spatiotemporal feature data, the power generation power of the power system at each moment within a preset historical period is obtained respectively;
[0063] The second spatiotemporal feature data and the power generation power of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
[0064] Specifically, the preset historical periods are one month, two months, half a year, one year, two years, etc., with each moment in hours, or in minutes or seconds.
[0065] In some examples, the training process of the trained neural network model further includes:
[0066] Under the second spatiotemporal feature data, the equipment status of the power system at each moment within a preset historical period is obtained respectively;
[0067] The second spatiotemporal feature data and the corresponding equipment status of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
[0068] The trained neural network model outputs the power generation capacity of the power plant and the equipment status. Specifically, when an anomaly is determined in the power plant based on the output power capacity of the trained neural network model, the equipment status output by the neural network model can be used to determine the problematic equipment in the power plant, facilitating subsequent maintenance.
[0069] In some examples, the trained neural network model includes both LSTM and Transformer models.
[0070] Specifically:
[0071] The LSTM model has the following advantages:
[0072] (1) It has a simple structure and few parameters, making it suitable for small and medium-sized datasets and resource-limited scenarios;
[0073] (2) It is more efficient at capturing the time dependencies of short sequences (such as n<100);
[0074] (3) It has low computational complexity and is suitable for real-time applications.
[0075] The Transformer model has the following advantages:
[0076] (1) It has strong parallelism and is suitable for large-scale data training and long sequence processing;
[0077] (2) Self-attention mechanism improves long-distance dependency modeling ability and transfer learning effect is good (such as pre-trained model);
[0078] (3) It can be extended to cross-modal tasks (such as text generation and speech recognition).
[0079] S104, Based on the first spatiotemporal feature data and the preset time period, the trained neural network model continuously outputs the power generation power of the power system in the future time period.
[0080] In some examples, step S104 specifically includes:
[0081] If the time period is less than a preset first threshold, the trained neural network model selects the LTSM model to calculate and continuously output the power generation of the power system in the future time period.
[0082] Specifically, the first threshold is 4 hours. When it is necessary to output the power generation of the power system in the next 4 hours, the trained neural network model selects the LTSM model to calculate and continuously output the power generation of the power system in the future time period.
[0083] In some examples, step S104 specifically includes:
[0084] If the time period exceeds a preset first threshold, the trained neural network model selects the Transformer model to calculate and output the power generation of the power system in the future time period.
[0085] Specifically, the trained neural network model can predict the power generation capacity of the power system within a preset future time period based on current meteorological data. When it is necessary to output the power generation capacity of the power system for the next 72 hours, the trained neural network model selects the Transformer model to calculate and continuously output the power generation capacity of the power system within the specified future time period.
[0086] Specifically, the trained neural network model works as follows:
[0087] (1) Construct a causal relationship graph between variables based on causal graph modeling (e.g., PC algorithm, Causal-CNN algorithm);
[0088] (2) Use Bayesian networks or structural equation models to infer risk sources;
[0089] (3) Identify the causal effects of weather factors (such as typhoons and blizzards) on the power generation capacity and line failure probability of wind power generation systems.
[0090] Example 2
[0091] like Figure 2 As shown in the embodiment of this application, the power generation prediction system for power systems based on multi-source meteorological data includes:
[0092] The acquisition module is configured to acquire meteorological data at various times of the day based on meteorological data generated by multiple meteorological forecast models, and generate the first multi-source meteorological data.
[0093] The extraction module is configured to extract first spatiotemporal feature data directly related to power generation from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation.
[0094] The input module is configured to input the first spatiotemporal feature data into the trained neural network model;
[0095] The output module is configured to continuously output the power generation power of the power system within the future time period based on the first spatiotemporal feature data and a preset time period after the neural network model has been trained.
[0096] The method, system, medium, and equipment for predicting power generation capacity of power systems based on multi-source meteorological data provided in this application improve the accuracy of prediction by using multi-source meteorological data and a trained neural network model to predict the power generation capacity of power systems, and provide effective reference information for power system operation and maintenance.
[0097] Example 3
[0098] On the other hand, embodiments of this application also provide an electronic device, see [link to relevant documentation]. Figure 3 , Figure 3 This is an exemplary structural block diagram of an electronic device according to an embodiment of this application, such as... Figure 3 As shown, the electronic device includes a processor and a memory, the memory storing computer instructions, and the processor executing the computer instructions to perform the method provided in this application.
[0099] Specifically, processor 601 may include a central processing unit (CPU) or a graphics processing unit (GPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application. Memory 602 may include memory for data or instructions. For example, memory 602 may be at least one of the following: a hard disk drive (HDD), read-only memory (ROM), random access memory (RAM), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, Universal Serial Bus (USB) drive, or other physical / tangible memory storage device. Alternatively, memory 602 may include removable or non-removable (or fixed) media. Furthermore, memory 602 may be internal or external to the integrated gateway disaster recovery device. Memory 602 may be non-volatile solid-state memory. In other words, memory 602 typically includes a tangible (non-transitory) computer-readable storage medium (such as a memory device) encoded with executable instructions, wherein the stored executable instructions, when executed by processor 601 (such as by one or more processors), can implement the methods in the embodiments of this application.
[0100] In one example Figure 3The illustrated electronic device may also include a communication interface 603 and a bus 610. The processor 601, memory 602, and communication interface 603 are connected via bus 610 and communicate with each other. Communication interface 603 is primarily used to enable communication between modules, devices, units, and / or equipment within the electronic device. Bus 610, including hardware, software, or both, couples components of the online data flow metering device together. For example, the bus may include at least one of the following: Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), HyperTransport (HT) interconnect, Industry Standard Architecture (ISA) bus, Infinite Bandwidth Interconnect, Low Pin Count (LPC) bus, memory bus, Microchannel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus, or other suitable buses. Bus 610 may include one or more buses. Although specific buses are described or illustrated in the embodiments of this application, any suitable bus or interconnection method may be considered in the embodiments of this application.
[0101] In another aspect, embodiments of this application also provide a computer-readable storage medium storing computer program instructions that, when executed by a processor, implement the aforementioned method. The computer-readable storage medium may be, for example, a classic computer-readable storage medium, such as a read-only memory (ROM), random access memory (RAM), disk storage medium, optical storage medium, flash memory, or other electrical, optical, or other physical / tangible memory storage device.
[0102] In another aspect, embodiments of this application also provide a computer program product, which includes computer program instructions that, when executed by a processor, implement the method provided in embodiments of this application. This computer program product may be, for example, a software installation package, a plug-in compatible with a related software system, etc.
[0103] The flowcharts and / or block diagrams of the methods and systems of embodiments of this application have been described above by way of example, and related aspects have been described. It should be understood that each block or combination thereof in the flowcharts and / or block diagrams can be implemented by computer program instructions, by dedicated hardware performing a specified function or action, or by a combination of dedicated hardware and computer instructions. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc.; when implemented in software, it is a program or code segment used to perform the required task. The program or code segment can be stored in memory or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.
[0104] While this application has shown and described numerous embodiments, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will arise for those skilled in the art without departing from the spirit and intent of this application. It should be understood that various alternatives to the embodiments of this application described herein may be employed in the practice of this application. The appended claims are intended to define the scope of protection of this application and therefore cover equivalents or alternatives within the scope of these claims.
Claims
1. A method for predicting power generation capacity of a power system based on multi-source meteorological data, characterized in that, include: Based on meteorological data generated from multiple meteorological forecasting models, meteorological data for each moment of the day is obtained to generate the first multi-source meteorological data. First spatiotemporal feature data directly related to power generation are extracted from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation. Input the first spatiotemporal feature data into the trained neural network model; Based on the first spatiotemporal feature data and the preset time period, the trained neural network model continuously outputs the power generation power of the power system within the future time period.
2. The power generation prediction method for a power system according to claim 1, characterized in that, The training process of the trained neural network model includes: Based on historical meteorological data collected from multiple weather forecasting systems, meteorological data for each moment within a preset historical period is obtained to generate a second multi-source meteorological data. Extract the second spatiotemporal feature data that is directly related to power generation at each time point from the second multi-source meteorological data. The second spatiotemporal feature data includes wind speed, wind direction and solar radiation. Under the second spatiotemporal feature data, the power generation power of the power system at each moment within a preset historical period is obtained respectively; The second spatiotemporal feature data and the power generation power of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
3. The power generation prediction method for a power system according to claim 2, characterized in that, The training process of the trained neural network model also includes: Under the second spatiotemporal feature data, the equipment status of the power system at each moment within a preset historical period is obtained respectively; The second spatiotemporal feature data and the corresponding equipment status of the power system at each moment within a preset historical period are input into the neural network model to obtain the trained neural network model.
4. The power generation prediction method for a power system according to claim 3, characterized in that: The trained neural network model includes an LSTM model and a Transformer model.
5. The power generation prediction method for a power system according to claim 4, characterized in that, Based on the first spatiotemporal feature data and the preset time period, the trained neural network model outputs the power generation capacity of the power system within the future time period, including: If the time period is less than a preset first threshold, the trained neural network model selects the LSTM model to calculate and continuously output the power generation of the power system in the future time period.
6. The power generation prediction method for a power system according to claim 4, characterized in that, Based on the first spatiotemporal feature data and the preset time period, the trained neural network model outputs the power generation capacity of the power system within the future time period, which further includes: If the time period is longer than the preset first threshold, the trained neural network model selects the Transformer model to calculate and continuously output the power generation of the power system in the future time period.
7. The power generation prediction method for a power system according to claim 1, characterized in that, After obtaining the meteorological data for the current moment based on meteorological data generated from multiple meteorological forecasting models, the method further includes: An adaptive weighted fusion algorithm is used to fuse various types of meteorological data in the first multi-source meteorological data.
8. A power generation prediction system for a power system based on multi-source meteorological data, characterized in that, include: The acquisition module is configured to acquire meteorological data at various times of the day based on meteorological data generated by multiple meteorological forecast models, and generate the first multi-source meteorological data. The extraction module is configured to extract first spatiotemporal feature data directly related to power generation from the first multi-source meteorological data, wherein the first spatiotemporal feature data includes wind speed, wind direction and solar radiation. The input module is configured to input the first spatiotemporal feature data into the trained neural network model; The output module is configured to continuously output the power generation power of the power system within the future time period based on the first spatiotemporal feature data and a preset time period after the neural network model has been trained.
9. A computer-readable storage medium, characterized in that, It includes program instructions that, when executed by a processor, cause the method according to any one of claims 1-7 to be implemented.
10. An electronic device, characterized in that, include: processor; as well as A memory storing computer instructions that, when executed by the processor, cause the electronic device to perform the method according to any one of claims 1-7.