WO2023009070A2 - Method and apparatus for forecasting optical power, computer device and storage medium - Google Patents

Method and apparatus for forecasting optical power, computer device and storage medium Download PDF

Info

Publication number
WO2023009070A2
WO2023009070A2 PCT/SG2022/050533 SG2022050533W WO2023009070A2 WO 2023009070 A2 WO2023009070 A2 WO 2023009070A2 SG 2022050533 W SG2022050533 W SG 2022050533W WO 2023009070 A2 WO2023009070 A2 WO 2023009070A2
Authority
WO
WIPO (PCT)
Prior art keywords
forecast
optical power
forecasting
layer
time period
Prior art date
Application number
PCT/SG2022/050533
Other languages
French (fr)
Other versions
WO2023009070A3 (en
Inventor
Zibo DONG
Renyu YUAN
Hui Yang
Qingsheng ZHAO
Original Assignee
Envision Digital International Pte. Ltd.
Shanghai Envision Digital Co., Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Envision Digital International Pte. Ltd., Shanghai Envision Digital Co., Ltd. filed Critical Envision Digital International Pte. Ltd.
Publication of WO2023009070A2 publication Critical patent/WO2023009070A2/en
Publication of WO2023009070A3 publication Critical patent/WO2023009070A3/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G06N20/20—Ensemble learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24H—FLUID HEATERS, e.g. WATER OR AIR HEATERS, HAVING HEAT-GENERATING MEANS, e.g. HEAT PUMPS, IN GENERAL
    • F24H15/00—Control of fluid heaters
    • F24H15/10—Control of fluid heaters characterised by the purpose of the control
    • F24H15/144—Measuring or calculating energy consumption
    • F24H15/152—Forecasting future energy consumption
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24H—FLUID HEATERS, e.g. WATER OR AIR HEATERS, HAVING HEAT-GENERATING MEANS, e.g. HEAT PUMPS, IN GENERAL
    • F24H15/00—Control of fluid heaters
    • F24H15/20—Control of fluid heaters characterised by control inputs
    • F24H15/262—Weather information or forecast
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F24—HEATING; RANGES; VENTILATING
    • F24H—FLUID HEATERS, e.g. WATER OR AIR HEATERS, HAVING HEAT-GENERATING MEANS, e.g. HEAT PUMPS, IN GENERAL
    • F24H15/00—Control of fluid heaters
    • F24H15/40—Control of fluid heaters characterised by the type of controllers
    • F24H15/414—Control of fluid heaters characterised by the type of controllers using electronic processing, e.g. computer-based
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/01—Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
    • H02J2300/20—The dispersed energy generation being of renewable origin
    • H02J2300/22—The renewable source being solar energy

Definitions

  • the present disclosure relates to the technical field of photovoltaic (PV) power generation, and in particular, relates to a method and apparatus for forecasting optical power, and a computer device and a storage medium thereof.
  • PV photovoltaic
  • PV power generation is a power generation method in which output optical power is closely related to meteorological conditions.
  • PV power generation enterprises generally input forecast data from meteorological sources into a machine learning model based on this single machine learning model to output forecasted optical power, which is not sufficiently accurate in forecasting.
  • Embodiments of the present disclosure provide a method and apparatus for forecasting optical power, and a computer device and a storage medium thereof.
  • the forecasting level of each meteorological source may be maximized, such that the accuracy in forecasting optical power is improved.
  • a method for forecasting optical power includes:
  • n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1 ;
  • the forecasting model including an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer including n machine learning processing modules;
  • an apparatus for forecasting optical power includes: a forecast data acquiring module, a model calling module, an optical power forecasting module, and an optical power correcting module; wherein
  • the forecast data acquiring module is configured to acquire n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1;
  • the model calling module is configured to call a forecasting model, the forecasting model including an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer including n machine learning processing modules;
  • the optical power forecasting module is configured to acquire i th reference forecast optical power for the target time period by inputting i th forecast data for the target time period into an i th machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n;
  • the optical power correcting module is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the optical power correcting layer.
  • a computer device includes: a processor and a memory storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the processor, when loading and executing the at least one instruction, the at least one program, the code set, or the instruction set, is caused to perform the method for forecasting optical power according to the above aspect.
  • a computer-readable storage medium stores at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above aspect.
  • a computer program product or a computer program includes one or more computer instructions stored in a computer-readable storage medium, wherein the one or more computer instructions, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above aspect.
  • the technical solutions according to the embodiments of the present disclosure at least may achieve the following beneficial effects.
  • the n reference forecast optical power corresponding to the n meteorological sources is acquired by acquiring the n forecast data from the n meteorological sources, and processing the n forecast data using the n machine learning processing modules respectively.
  • the final corrected forecast optical power is acquired by correcting the n reference forecast optical power via the optical power correcting layer, realizing a processing method of optical power forecast. Compared with a current processing method using a single machine learning processing module, the forecast level of each meteorological source is maximized and the advantages of the different meteorological sources are integrated, such that the accuracy in forecasting optical power is improved.
  • FIG. 1 is a schematic diagram of a system for forecasting optical power according to an exemplary embodiment of the present disclosure
  • FIG. 2 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure
  • FIG. 3 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure
  • FIG. 4 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure
  • FIG. 5 is a schematic diagram of a method for forecasting optical power according to an exemplary embodiment of the present disclosure
  • FIG. 6 is a block diagram of an apparatus for forecasting optical power according to an exemplary embodiment of the present disclosure.
  • FIG. 7 is a schematic structural diagram of a computer device according to an exemplary embodiment of the present disclosure.
  • PV power generation using solar energy has the characteristics of low energy density, intermittency, uncertainty, and the like.
  • the output power of PV power generation is closely related to meteorological conditions, which makes its power generation characteristics much different from that of conventional power generation.
  • Grid-connected access of PV power generation is an important form to realize large-scale efficient utilization of PV power generation. Due to the intermittency, uncertainty and uncontrollability of PV power generation, a great challenge may be posed to safe operation of the public supply system in the case that a large-scale and high-capacity PV power generation system is connected to the power grid.
  • optical power forecast mainly builds a single machine learning model based on forecast data of meteorological sources, with irradiation or optical power as optimization targets.
  • each meteorological source has its own limitations and structural deviations, how to give full play to the advantages of various meteorological sources and reduce deviations of each meteorological source is a great challenge for optical power forecast.
  • each meteorological source is modeled such that the advantages of each meteorological source are given into full play, the deviations of each meteorological source are reduced, and the accuracy in forecasting optical power is improved.
  • the method for forecasting optical power according to the embodiments of the present disclosure is described hereinafter by examples.
  • FIG. 1 is a schematic diagram of a system for forecasting optical power according to an embodiment of the present disclosure.
  • the system for forecasting optical power may include a meteorological source 110, a computer device 120, and a PV power generation enterprise 130.
  • the meteorological source 110 may release meteorological forecast data, such as forecast irradiation, temperature, rainfall, cloud cover, zenith angle, clearness index, and weather type.
  • meteorological sources 110 may include the weather research and forecasting model (WRF), deterministic forecasts provided by the European Centre for Medium-Range Weather Forecasts (ECMWF), ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the Weather Company of the International Business Machines Corporation (IBM), and the like.
  • WRF weather research and forecasting model
  • ECMWF European Centre for Medium-Range Weather Forecasts
  • IBM International Business Machines Corporation
  • the computer device 120 may be a device (such as a server with computing and storage capabilities) capable of transmitting data and conducting data processing, or a terminal (such as a cell phone, a tablet computer, a multimedia playback device and a wearable device), or other computer devices.
  • a terminal such as a cell phone, a tablet computer, a multimedia playback device and a wearable device
  • the computer device may be a single server, a server cluster consisting of a plurality of servers, or a cloud computing service center.
  • the computer device 120 may acquire forecast data provided by the plurality of meteorological sources 110, acquire corrected forecast optical power by calling a forecasting model to process these forecast data, and send the corrected forecast optical power to the PV power generation enterprise 130.
  • the PV power generation enterprise 130 is capable of forecasting optical power.
  • the computer device 120 may be provided in the PV power generation enterprise 130 or may be independent of the PV power generation enterprise 130, which is not limited in the embodiments of the present disclosure.
  • FIG. 2 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps.
  • n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
  • the computer device acquires the n forecast data for the target time period from the n meteorological sources.
  • the target time period is a future time period after a current moment, and the specific length of the target time period is not limited in the embodiments of the present disclosure.
  • the target period may be a day, a half day, or a midday period, such as 11:00 AM to 13:00 PM.
  • the meteorological source is a meteorological product capable of releasing meteorological forecast data.
  • the meteorological sources include at least two of the following meteorological sources: deterministic forecasts provided by the ECMWF, ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the International Business Machines Corporation (IBM), and the WRF.
  • the computer device acquires first day-ahead forecast data from the WRF and second day-ahead forecast data from the deterministic forecasts provided by the ECMWF.
  • the term "day-ahead" in the first day-ahead forecast data and the second day-ahead forecast data means that a target time period for the forecast data is 24 hours in the future.
  • step 204 a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
  • the computer device calls the forecasting model to process the acquired n forecast data in response to acquiring the n forecast data from the n meteorological sources.
  • the optical power forecasting layer in the forecasting model is a forecasting layer for outputting reference forecast optical power.
  • the optical power forecasting layer includes n machine learning processing modules that are configured to acquire n reference forecast optical power by processing the n forecast data respectively.
  • any one of the machine learning processing modules includes at least one machine learning model.
  • the optical power correcting layer in the forecasting model is a correcting layer for processing the n reference forecast optical power so as to output corrected forecast optical power.
  • i th reference forecast optical power for the target time period is acquired by inputting i th forecast data for the target time period into an i th machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n.
  • the computer device acquires first reference forecast optical power for the taiget time period by inputting first forecast data for the target time period into a first machine learning processing module in the optical power forecasting layer, and acquires second reference forecast optical power for the target time period by inputting second forecast data for the target time period into a second machine learning processing module in the optical power forecasting layer until it acquires the first reference forecast optical power to n th reference forecast optical power.
  • the reference forecast optical power is a reference forecast value of the forecast optical power acquired after being processed by the optical power forecasting layer. It can be understood that since the forecast data for the different meteorological sources is processed by the different machine learning processing modules, forecast characteristics of the different meteorological sources may be fully taken into account, which, compared with a current processing method using a single machine learning processing module, improves the accuracy in forecasting the reference forecast optical power corresponding to each meteorological source.
  • step 208 corrected forecast optical power is acquired by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the optical power correcting layer.
  • the computer device acquires the corrected forecast optical power for the target time period by inputting the n reference forecast optical power into the optical power correcting layer, in response to acquiring the n reference forecast optical power by the n machine learning processing modules in the optical power forecasting layer.
  • the corrected forecast optical power is a final forecast value of the forecast optical power acquired by corrected, in response to comprehensive consideration of the n reference forecast optical power by the optical power correcting layer.
  • the optical power correcting layer includes at least one machine learning model.
  • the n reference forecast optical power corresponding to the n meteorological sources is acquired by acquiring the n forecast data from the n meteorological sources and processing the n forecast data using the n machine learning processing modules respectively, and the final corrected forecast optical power is acquired by correcting the n reference forecast optical power via the optical power correcting layer, thereby practicing a processing method of optical power forecast.
  • the forecast level of each meteorological source is maximized and the advantages of the different meteorological sources are integrated, such that the accuracy in forecasting optical power is improved.
  • the optical power forecasting layer is divided into a first forecasting layer and a second forecasting layer; a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the first forecasting layer; and further, the more accurate corrected forecast parameter may be input into the second forecasting layer so as to forecast optical power.
  • FIG. 3 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps. [0058] In step 302, n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
  • step 202 For details about this step, reference may be made to step 202, which is not repeated herein.
  • step 304 a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules, and the optical power forecasting layer includes a first forecasting layer and a second forecasting layer.
  • the optical power forecasting layer includes the first forecasting layer and the second forecasting layer, and correspondingly, an i th machine learning processing module in the n machine learning processing modules includes an ii th machine learning model in the first forecasting layer, and an h th machine learning model in the second forecasting layer. It can be understood that the ii th machine learning model and the 12 th machine learning model are different machine learning models in the different forecasting layers.
  • the machine learning models in the n machine learning processing modules in the optical power forecasting layer are of at least one of the following types: an extreme gradient boosting (XGB) model; a light gradient boosting machine (light GBM) model, a gradient boosting decision tree (GBDT) model, a random forest (RF) model, and a neural network model.
  • XGB extreme gradient boosting
  • light GBM light gradient boosting machine
  • GBDT gradient boosting decision tree
  • RF random forest
  • the ii th machine learning model and the 12 th machine learning model may be of the same or different types.
  • both of the ii th machine learning model and the 12 th machine learning model are XGB models.
  • the ii th machine learning model is an XGB model and the 12 th machine learning model is a GBDT model.
  • step 306 an i th corrected forecast parameter for the target time period is acquired by inputting i th forecast data for the target time period into the ii th machine learning model in the first forecasting layer.
  • the computer device acquires a first corrected forecast parameter for the target time period by inputting first forecast data for the target time period into a first 1 machine learning model in the first forecasting layer, and acquires a second corrected forecast parameter for the target time period by inputting second forecast data for the target time period into a secondi machine learning model in the first forecasting layer until it acquires the first corrected forecast parameter to n th corrected forecast parameter.
  • the i th corrected forecast parameter is data acquired by optimizing a forecast parameter in the i th forecast data via the ii th machine learning model. It can be understood that a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the machine learning model in the first forecasting layer, such that the more accurate corrected forecast parameter may be used to forecast optical power.
  • the i th forecast data includes at least one of: i th forecast irradiation, an i th forecast temperature and an i th forecast moment label.
  • the i th forecast data further includes at least one of: i th forecast rainfall, i th forecast cloud cover, an i th forecast zenith angle, an i th forecast AM and PM label, an i th forecast clearness index, and an i th weather type.
  • the i th forecast data includes: i th forecast irradiation, an i th forecast temperature and an i th forecast moment label; and the i th forecast data include: i th forecast rainfall, i th forecast cloud cover, an i th forecast zenith angle, an i th forecast morning and afternoon label, an i th forecast clearness index, and an i th weather type.
  • the forecast irradiation refers to the radiation energy that reaches the unit area of the earth's surface within the target time period after solar radiation is absorbed, scattered, and emitted by the atmosphere, in units of watts per square meter (W/m 2 ).
  • the forecast moment label refers to a corresponding forecast moment within the target time period.
  • the forecast clearness index is intended to describe the influence on solar shortwave radiation from the atmosphere, and is a ratio of the total solar radiation incident onto a horizontal plane to the astronomical radiation within the taiget time period, in units of Joules per square meter (J/m 2 ).
  • the weathertype refers to the type of weather within the target time period, e.g., rainy, sunny, cloudy, etc.
  • the forecast temperature refers to the range of temperature within the target time period, in units of Celsius (°C).
  • the forecast rainfall refers to the depth to which rainfall accumulates on the water surface within the target time period.
  • the forecast cloud cover refers to the percentage of the sky view obscured by clouds within the target time period.
  • the forecast zenith angle refers to an angular distance between a celestial body and the zenith within the target time period, in units of °.
  • the forecast AM and PM label refer to AM and PM properties of the time within the target time period.
  • i th reference forecast optical power for the target time period is acquired by inputting an i th corrected forecast parameter and i th intermediate forecast data for the target time period into the 12 th machine learning model in the second forecasting layer.
  • the computer device acquires the following two types of data: the i th corrected forecast parameter for the target time period, optimized by the first forecasting layer, and the i th intermediate forecast data for the target time period, and acquires the i th reference forecast optical power for the target time period by inputting these two types of data into the 12 th machine learning model in the second forecasting layer.
  • the i th intermediate forecast data is data in the i th forecast data that is to be input into the i2 th machine learning model in the second forecasting layer. That is, the i th intermediate forecast data is part or all of the i th forecast data acquired from the meteorological source.
  • the i th corrected forecast parameter includes i th corrected forecast irradiation. It can be understood that since the accuracy of forecast irradiation has a great impact on the accuracy of the forecast optical power, an explanation is made by taking the i th corrected forecast parameter including the i th corrected forecast irradiation as an example. That is, in step 306, the computer device acquires more accurate corrected forecast irradiation by inputting the forecast data into the first forecasting layer and using the machine learning model in the first forecasting layer to process the optimal forecast irradiation.
  • the i th intermediate forecast data includes at least one of the following forecast data: i th forecast irradiation and an i th forecast moment label.
  • the i th intermediate forecast data further includes at least one of the following forecast data: an i th forecast clearness index and an i th weather type.
  • the i th intermediate forecast data includes: i th forecast irradiation and an i th forecast moment label; and the i th forecast data includes: an i th forecast clearness index and an i th weather type.
  • step 310 corrected forecast optical power is acquired by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the optical power correcting layer.
  • step 208 For details about this step, reference may be made to step 208, which is not repeated herein.
  • the optical power forecasting layer is divided into a first forecasting layer and a second forecasting layer, and a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the first forecasting layer, such that the more accurate corrected forecast parameter may be input into the second forecasting layer to forecast optical power with higher accuracy.
  • the optical power correcting layer includes a regression model.
  • FIG. 4 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps.
  • n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
  • step 202 For details about this step, reference may be made to step 202, which is not repeated herein.
  • step 404 a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
  • step 204 For details about this step, reference may be made to step 204, which is not repeated herein.
  • i th reference forecast optical power for the target time period is acquired by inputting i th forecast data for the target time period into an i th machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n.
  • i is a positive integer not greater than n.
  • step 408 corrected forecast optical power for the target time period is acquired by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the regression model of the optical power correcting layer for regression algorithm processing.
  • the computer device acquires the corrected forecast optical power for the target time period by inputting the n reference forecast optical power into the regression model in the optical power correcting layer for regression algorithm processing, in response to acquiring the n reference forecast optical power by the n machine learning processing modules in the optical power forecasting layer.
  • the regression model is of at least one of the following types: a linear regression model, a ridge regression model, and a least absolute shrinkage and selection operator (LASSO) regression model.
  • LASSO least absolute shrinkage and selection operator
  • the optical power correcting layer includes the regression model for regression algorithm processing of the n reference forecast optical power output by the optical power forecasting layer, such that the final corrected forecast optical power is acquired by synthesizing the n reference forecast optical power, which improves the accuracy in forecasting optical power.
  • the forecasting model includes three layers, namely a first layer of the forecasting model (also known as the first forecasting layer in the embodiments of the present disclosure), a second layer of the forecasting model (also known as the second forecasting layer in the embodiments of the present disclosure), and a third layer of the forecasting model (also known as the optical power correcting layer in the embodiments of the present disclosure).
  • a first layer of the forecasting model also known as the first forecasting layer in the embodiments of the present disclosure
  • a second layer of the forecasting model also known as the second forecasting layer in the embodiments of the present disclosure
  • a third layer of the forecasting model also known as the optical power correcting layer in the embodiments of the present disclosure
  • the first layer of the forecasting model includes n XGB models, n is a positive integer greater than 1, and the XGB models in the first layer of the forecasting model aim to optimize forecast radiation.
  • n meteorological sources namely a meteorological source 1 to a meteorological source n
  • each meteorological source inputs corresponding forecast data into the corresponding XGB model
  • the meteorological source 1 inputs first forecast data into the corresponding XGB model
  • the meteorological source n inputs n th forecast data into the corresponding XGB model.
  • Each XGB model outputs corrected forecast radiation, for example, the XGB model corresponding to the meteorological source 1 outputs first corrected forecast irradiation, and the XGB model corresponding to the meteorological source n outputs n th corrected forecast radiation.
  • the second layer of the forecasting model also includes n XGB models, wherein the XGB models in the second layer of the forecasting model are intended to optimize the forecast optical power.
  • An i th XGB model in the second layer of the forecasting model includes the following input data: i th corrected forecast irradiation acquired by the first layer of the forecasting model, and i th intermediate forecast data of i th forecast data corresponding to a meteorological source i.
  • a first XGB model in the second layer of the forecasting model includes the following input data: first corrected forecast irradiation and first intermediate forecast data, and an n th XGB model in the second layer of the forecasting model includes n th corrected forecast irradiation and n th intermediate forecast data.
  • Each XGB model outputs the reference forecast optical power, e.g., the XGB model corresponding to the meteorological source 1 outputs first reference forecast optical power, and the XGB model corresponding to the meteorological source n outputs n th reference forecast optical power.
  • the third layer of the forecasting model includes one linear regression model, wherein the linear regression model is configured to acquire final corrected forecast optical power as a result of an optical power forecast by conducting a linear regression operation on the n reference forecast optical power corresponding to the n meteorological sources acquired by the second layer of the forecasting model.
  • FIG. 6 is a schematic structural diagram of an apparatus for forecasting optical power according to an exemplary embodiment of the present disclosure.
  • the apparatus may be implemented as all or a part of a computer device through software, hardware or a combination of the two.
  • the apparatus includes a forecast data acquiring module 601, a model calling module 602, an optical power forecasting module 603, and an optical power correcting module 604.
  • the forecast data acquiring module 601 is configured to acquire n forecast data for a target time period from n meteorological sources respectively, wherein n is a positive integer greater than 1.
  • the model calling module 602 is configured to call a forecasting model, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
  • the optical power forecasting module 603 is configured to acquire i th reference forecast optical power for the target time period by inputting i th forecast data for the target time period into an i th machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n.
  • the optical power correcting module 604 is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the optical power correcting layer for regression algorithm processing.
  • the optical power forecasting layer includes a first forecasting layer and a second forecasting layer
  • the i th machine learning processing module in the optical power forecasting layer includes an ii th machine learning model in the first forecasting layer and an 12 th machine learning model in the second forecasting layer.
  • the optical power forecasting module 603 is configured to acquire an i th corrected forecast parameter for the target time period by inputting i th forecast data for the target time period into the ii th machine learning model in the first forecasting layer, and acquire i th reference forecast optical power for the target time period by inputting the i th corrected forecast parameter and i th intermediate forecast data for the target time period into the 12 th machine learning model in the second forecasting layer.
  • the i th intermediate forecast data is data in the i th forecast data that is to be input into the 12 th machine learning model in the second forecasting layer.
  • the i th corrected forecast parameter includes i th corrected forecast irradiation.
  • the i th intermediate forecast data includes at least one of: i th forecast irradiation and an i th forecast moment label.
  • the i th intermediate forecast data further includes at least one of: an i th forecast clearness index and an i th weather type.
  • the i th forecast data includes at least one of: i th forecast irradiation, an i th forecast temperature, and an i th forecast moment label.
  • the i th forecast data further includes at least one of: i th forecast rainfall, i th forecast cloud cover, an i th forecast zenith angle, an i th forecast AM and PM label, an i th forecast clearness index, and an i th weather type.
  • the optical power correcting layer includes a regression model.
  • the optical power correcting module 604 is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to n th reference forecast optical power for the target time period into the regression model of the optical power correcting layer for regression algorithm processing.
  • the regression model is at least one of: a linear regression model, a ridge regression model, and a least absolute shrinkage and selection operator (LASSO) regression model.
  • LASSO least absolute shrinkage and selection operator
  • the meteorological sources include at least two of: deterministic forecasts provided by the ECMWF, ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the International Business Machines Corporation (IBM), and the WRF.
  • the machine learning models in the n machine learning processing modules in the optical power forecasting layer may be of at least one of: an XGB model; a light GBM model, a GBDT model, an RF model, and a neural network model.
  • FIG. 7 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure.
  • the computer device 700 includes a central processing unit (CPU) 701, a system memory 704 including a random-access memory (RAM) 702 and a read-only memory (ROM) 703, and a system bus 705 connecting the system memory 704 and the CPU 701.
  • the computer device 700 further includes a basic input/output (FO) system 706 configured to help transmit information among various components within the computer device, and a mass storage device 707 configured to store an operating system 713, an application 714, and other program modules 715.
  • CPU central processing unit
  • system memory 704 including a random-access memory (RAM) 702 and a read-only memory (ROM) 703
  • a system bus 705 connecting the system memory 704 and the CPU 701.
  • the computer device 700 further includes a basic input/output (FO) system 706 configured to help transmit information among various components within the computer device, and a mass storage device 707 configured to store an operating
  • the basic I/O system 706 includes a display 708 configured to display information and an input device 709, for example, a mouse, a keyboard, and the like, configured for the user to input information. Both the display 708 and the input device 709 are connected to the CPU 701 by an input/output controller 710 connected to the system bus 705.
  • the basic I/O system 706 may also include the input/output controller 710 configured to receive and processing inputs from a plurality of other devices, such as the keyboard, the mouse, or an electronic stylus. Similarly, the input/output controller 710 is further configured to provide outputs to the display, a printer or other types of output devices.
  • the mass storage device 707 is connected to the CPU 701 by a mass storage controller (not shown) connected to the system bus 705.
  • the mass storage device 707 and a computer-readable medium associated therewith provide non-volatile storage for the computer device 700. That is, the mass storage device 707 may include the computer-readable medium (not shown), such as a hard disk or a compact disc read-only memory (CD-ROM) driver.
  • CD-ROM compact disc read-only memory
  • the computer-readable medium may include a computer storage medium and a communication medium.
  • the computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as a computer-readable instruction, a data structure, a program module or other data.
  • the computer storage medium includes a RAM, a ROM, an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory or other solid-state memory technologies; a CD-ROM, a digital versatile disc (DVD) or other optical storage; and a tape cassette, a magnetic tape, a magnetic disk storage or other magnetic storage devices.
  • the computer storage medium is not limited to above.
  • the above system memory 704 and the mass storage device 707 may be collectively referred to as the memory.
  • the computer device 700 may also be operated by being connected by a network such as the Internet to a remote computer. That is, the computer device 700 may be connected to the network 712 by a network interface unit 711 connected to the system bus 705, or that is, the computer device 700 may be connected to other types of networks or remote computer systems (not shown) by using the network interface unit 711.
  • An embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above method embodiments.
  • An embodiment of the present disclosure further provides a computer program product or a computer program.
  • the computer program product or the computer program includes one or more computer instructions stored in a computer-readable storage medium, wherein the one or more computer instructions, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above embodiments.
  • the term "plurality” herein refers to two or more.
  • the term “and/or” herein describes the associated relationship of the associated objects, indicating three relationships. For example, A and/or B may indicate that: A exists alone, A and B exist concurrently, B exists alone.
  • the symbol “/” generally indicates that the contextual objects are in an "or” relationship.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Human Resources & Organizations (AREA)
  • Computing Systems (AREA)
  • Tourism & Hospitality (AREA)
  • Mathematical Physics (AREA)
  • Medical Informatics (AREA)
  • Power Engineering (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Marketing (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Secondary Cells (AREA)

Abstract

Disclosed are a method and apparatus for forecasting optical power, and a computer device and a storage medium thereof. The method includes: acquiring n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1; calling a forecasting model, the forecasting model including an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer including n machine learning processing modules; acquiring ith reference forecast optical power for the target time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n; and acquiring corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.

Description

METHOD AND APPARATUS FOR FORECASTING OPTICAE POWER, COMPUTER DEVICE AND STORAGE MEDIUM
TECHNICAU FIEUD
[0001] The present disclosure relates to the technical field of photovoltaic (PV) power generation, and in particular, relates to a method and apparatus for forecasting optical power, and a computer device and a storage medium thereof.
BACKGROUND
[0002] PV power generation is a power generation method in which output optical power is closely related to meteorological conditions.
[0003] At present, PV power generation enterprises generally input forecast data from meteorological sources into a machine learning model based on this single machine learning model to output forecasted optical power, which is not sufficiently accurate in forecasting.
SUMMARY
[0004] Embodiments of the present disclosure provide a method and apparatus for forecasting optical power, and a computer device and a storage medium thereof. In these solutions, the forecasting level of each meteorological source may be maximized, such that the accuracy in forecasting optical power is improved.
[0005] According to one aspect of the embodiments of the present disclosure, a method for forecasting optical power is provided. The method includes:
[0006] acquiring n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1 ;
[0007] calling a forecasting model, the forecasting model including an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer including n machine learning processing modules;
[0008] acquiring ith reference forecast optical power for the target time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n; and [0009] acquiring corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer. [0010] According to another aspect of the embodiments of the present disclosure, an apparatus for forecasting optical power is provided. The apparatus includes: a forecast data acquiring module, a model calling module, an optical power forecasting module, and an optical power correcting module; wherein
[0011] the forecast data acquiring module is configured to acquire n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1; [0012] the model calling module is configured to call a forecasting model, the forecasting model including an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer including n machine learning processing modules;
[0013] the optical power forecasting module is configured to acquire ith reference forecast optical power for the target time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n; and
[0014] the optical power correcting module is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.
[0015] According to yet another aspect of the embodiments of the present disclosure, a computer device is provided. The computer device includes: a processor and a memory storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the processor, when loading and executing the at least one instruction, the at least one program, the code set, or the instruction set, is caused to perform the method for forecasting optical power according to the above aspect.
[0016] According to still another aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided. The storage medium stores at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above aspect.
[0017] According to yet still another aspect of the embodiments of the present disclosure, a computer program product or a computer program is provided. The computer program product or the computer program includes one or more computer instructions stored in a computer-readable storage medium, wherein the one or more computer instructions, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above aspect. [0018] The technical solutions according to the embodiments of the present disclosure at least may achieve the following beneficial effects.
[0019] The n reference forecast optical power corresponding to the n meteorological sources is acquired by acquiring the n forecast data from the n meteorological sources, and processing the n forecast data using the n machine learning processing modules respectively. The final corrected forecast optical power is acquired by correcting the n reference forecast optical power via the optical power correcting layer, realizing a processing method of optical power forecast. Compared with a current processing method using a single machine learning processing module, the forecast level of each meteorological source is maximized and the advantages of the different meteorological sources are integrated, such that the accuracy in forecasting optical power is improved.
BRIEF DESCRIPTION OF THE DRAWINGS [0020] For clearer descriptions of the technical solutions in the embodiments of the present disclosure, the following briefly introduces the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and a person of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.
[0021] FIG. 1 is a schematic diagram of a system for forecasting optical power according to an exemplary embodiment of the present disclosure;
[0022] FIG. 2 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure;
[0023] FIG. 3 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure;
[0024] FIG. 4 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure;
[0025] FIG. 5 is a schematic diagram of a method for forecasting optical power according to an exemplary embodiment of the present disclosure;
[0026] FIG. 6 is a block diagram of an apparatus for forecasting optical power according to an exemplary embodiment of the present disclosure; and
[0027] FIG. 7 is a schematic structural diagram of a computer device according to an exemplary embodiment of the present disclosure.
DETAILED DESCRIPTION [0028] The embodiments of the present disclosure are described hereinafter in further detail with reference to the accompanying drawings, to present the objectives, technical solutions, and advantages of the present disclosure more clearly.
[0029] PV power generation using solar energy has the characteristics of low energy density, intermittency, uncertainty, and the like. In particular, the output power of PV power generation is closely related to meteorological conditions, which makes its power generation characteristics much different from that of conventional power generation.
[0030] Grid-connected access of PV power generation is an important form to realize large-scale efficient utilization of PV power generation. Due to the intermittency, uncertainty and uncontrollability of PV power generation, a great challenge may be posed to safe operation of the public supply system in the case that a large-scale and high-capacity PV power generation system is connected to the power grid.
[0031] Therefore, the acceptance and digestion of such unstable energy by the power grid can be promoted if the optical power of the PV power generation system is forecast accurately to alleviate the uncertainty of the output power of the PV power generation system, which is of great significance to the safety and stability of grid-connected access operation of the PV power generation system, as well as economic dispatch of the power grid.
[0032] Currently, optical power forecast mainly builds a single machine learning model based on forecast data of meteorological sources, with irradiation or optical power as optimization targets. [0033] Since each meteorological source has its own limitations and structural deviations, how to give full play to the advantages of various meteorological sources and reduce deviations of each meteorological source is a great challenge for optical power forecast.
[0034] In the embodiments of the present disclosure, each meteorological source is modeled such that the advantages of each meteorological source are given into full play, the deviations of each meteorological source are reduced, and the accuracy in forecasting optical power is improved. The method for forecasting optical power according to the embodiments of the present disclosure is described hereinafter by examples.
[0035] FIG. 1 is a schematic diagram of a system for forecasting optical power according to an embodiment of the present disclosure. As illustrated in FIG. 1, the system for forecasting optical power may include a meteorological source 110, a computer device 120, and a PV power generation enterprise 130.
[0036] The meteorological source 110 may release meteorological forecast data, such as forecast irradiation, temperature, rainfall, cloud cover, zenith angle, clearness index, and weather type. In an embodiment of the present disclosure, at least two meteorological sources 110 are provided, and FIG. 1 illustrates only by taking 3 meteorological sources 110 as an example. In some embodiments, the meteorological sources 110 may include the weather research and forecasting model (WRF), deterministic forecasts provided by the European Centre for Medium-Range Weather Forecasts (ECMWF), ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the Weather Company of the International Business Machines Corporation (IBM), and the like.
[0037] The computer device 120 may be a device (such as a server with computing and storage capabilities) capable of transmitting data and conducting data processing, or a terminal (such as a cell phone, a tablet computer, a multimedia playback device and a wearable device), or other computer devices. Optionally, in the case that the computer device is a server, the computer device may be a single server, a server cluster consisting of a plurality of servers, or a cloud computing service center. In an embodiment of the present disclosure, the computer device 120 may acquire forecast data provided by the plurality of meteorological sources 110, acquire corrected forecast optical power by calling a forecasting model to process these forecast data, and send the corrected forecast optical power to the PV power generation enterprise 130.
[0038] The PV power generation enterprise 130 is capable of forecasting optical power. Optionally, the computer device 120 may be provided in the PV power generation enterprise 130 or may be independent of the PV power generation enterprise 130, which is not limited in the embodiments of the present disclosure.
[0039] FIG. 2 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps.
[0040] In step 202, n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
[0041] In some embodiments, the computer device acquires the n forecast data for the target time period from the n meteorological sources.
[0042] The target time period is a future time period after a current moment, and the specific length of the target time period is not limited in the embodiments of the present disclosure. Exemplarily, the target period may be a day, a half day, or a midday period, such as 11:00 AM to 13:00 PM.
[0043] The meteorological source is a meteorological product capable of releasing meteorological forecast data. In an embodiment of the present disclosure, there may be a plurality of meteorological sources that supports the release of multiple forecast data, and the meteorological sources and the forecast data are in one-to-one correspondence. Optionally, the meteorological sources include at least two of the following meteorological sources: deterministic forecasts provided by the ECMWF, ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the International Business Machines Corporation (IBM), and the WRF.
[0044] In some embodiments, the computer device acquires first day-ahead forecast data from the WRF and second day-ahead forecast data from the deterministic forecasts provided by the ECMWF. The term "day-ahead" in the first day-ahead forecast data and the second day-ahead forecast data means that a target time period for the forecast data is 24 hours in the future.
[0045] In step 204, a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
[0046] In some embodiments, the computer device calls the forecasting model to process the acquired n forecast data in response to acquiring the n forecast data from the n meteorological sources.
[0047] The optical power forecasting layer in the forecasting model is a forecasting layer for outputting reference forecast optical power. In an embodiment of the present disclosure, the optical power forecasting layer includes n machine learning processing modules that are configured to acquire n reference forecast optical power by processing the n forecast data respectively. In some embodiments, any one of the machine learning processing modules includes at least one machine learning model.
[0048] The optical power correcting layer in the forecasting model is a correcting layer for processing the n reference forecast optical power so as to output corrected forecast optical power. [0049] In step 206, ith reference forecast optical power for the target time period is acquired by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n. [0050] In some embodiments, the computer device acquires first reference forecast optical power for the taiget time period by inputting first forecast data for the target time period into a first machine learning processing module in the optical power forecasting layer, and acquires second reference forecast optical power for the target time period by inputting second forecast data for the target time period into a second machine learning processing module in the optical power forecasting layer until it acquires the first reference forecast optical power to nth reference forecast optical power.
[0051] The reference forecast optical power is a reference forecast value of the forecast optical power acquired after being processed by the optical power forecasting layer. It can be understood that since the forecast data for the different meteorological sources is processed by the different machine learning processing modules, forecast characteristics of the different meteorological sources may be fully taken into account, which, compared with a current processing method using a single machine learning processing module, improves the accuracy in forecasting the reference forecast optical power corresponding to each meteorological source.
[0052] In step 208, corrected forecast optical power is acquired by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.
[0053] In some embodiments, the computer device acquires the corrected forecast optical power for the target time period by inputting the n reference forecast optical power into the optical power correcting layer, in response to acquiring the n reference forecast optical power by the n machine learning processing modules in the optical power forecasting layer.
[0054] The corrected forecast optical power is a final forecast value of the forecast optical power acquired by corrected, in response to comprehensive consideration of the n reference forecast optical power by the optical power correcting layer. In some embodiments, the optical power correcting layer includes at least one machine learning model.
[0055] In summary, according to the method according to the embodiment of the present disclosure, the n reference forecast optical power corresponding to the n meteorological sources is acquired by acquiring the n forecast data from the n meteorological sources and processing the n forecast data using the n machine learning processing modules respectively, and the final corrected forecast optical power is acquired by correcting the n reference forecast optical power via the optical power correcting layer, thereby practicing a processing method of optical power forecast. In this way, compared with a current processing method using a single machine learning processing module, the forecast level of each meteorological source is maximized and the advantages of the different meteorological sources are integrated, such that the accuracy in forecasting optical power is improved.
[0056] In some embodiments, for improvement of the accuracy in forecasting the reference forecast optical power output by the optical power forecasting layer, the optical power forecasting layer is divided into a first forecasting layer and a second forecasting layer; a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the first forecasting layer; and further, the more accurate corrected forecast parameter may be input into the second forecasting layer so as to forecast optical power.
[0057] FIG. 3 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps. [0058] In step 302, n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
[0059] For details about this step, reference may be made to step 202, which is not repeated herein.
[0060] In step 304, a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules, and the optical power forecasting layer includes a first forecasting layer and a second forecasting layer.
[0061] In this embodiment, the optical power forecasting layer includes the first forecasting layer and the second forecasting layer, and correspondingly, an ith machine learning processing module in the n machine learning processing modules includes an iith machine learning model in the first forecasting layer, and an hth machine learning model in the second forecasting layer. It can be understood that the iith machine learning model and the 12th machine learning model are different machine learning models in the different forecasting layers.
[0062] In some embodiments, the machine learning models in the n machine learning processing modules in the optical power forecasting layer are of at least one of the following types: an extreme gradient boosting (XGB) model; a light gradient boosting machine (light GBM) model, a gradient boosting decision tree (GBDT) model, a random forest (RF) model, and a neural network model.
[0063] In this embodiment, the iith machine learning model and the 12th machine learning model may be of the same or different types. Exemplarily, both of the iith machine learning model and the 12th machine learning model are XGB models. Exemplarily, the iith machine learning model is an XGB model and the 12th machine learning model is a GBDT model.
[0064] In step 306, an ith corrected forecast parameter for the target time period is acquired by inputting ith forecast data for the target time period into the iith machine learning model in the first forecasting layer.
[0065] In some embodiments, the computer device acquires a first corrected forecast parameter for the target time period by inputting first forecast data for the target time period into a first 1 machine learning model in the first forecasting layer, and acquires a second corrected forecast parameter for the target time period by inputting second forecast data for the target time period into a secondi machine learning model in the first forecasting layer until it acquires the first corrected forecast parameter to nth corrected forecast parameter.
[0066] The ith corrected forecast parameter is data acquired by optimizing a forecast parameter in the ith forecast data via the iith machine learning model. It can be understood that a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the machine learning model in the first forecasting layer, such that the more accurate corrected forecast parameter may be used to forecast optical power.
[0067] In some embodiments, the ith forecast data includes at least one of: ith forecast irradiation, an ith forecast temperature and an ith forecast moment label. Optionally, the ith forecast data further includes at least one of: ith forecast rainfall, ith forecast cloud cover, an ith forecast zenith angle, an ith forecast AM and PM label, an ith forecast clearness index, and an ith weather type. [0068] In some embodiments, the ith forecast data includes: ith forecast irradiation, an ith forecast temperature and an ith forecast moment label; and the ith forecast data include: ith forecast rainfall, ith forecast cloud cover, an ith forecast zenith angle, an ith forecast morning and afternoon label, an ith forecast clearness index, and an ith weather type.
[0069] The forecast irradiation refers to the radiation energy that reaches the unit area of the earth's surface within the target time period after solar radiation is absorbed, scattered, and emitted by the atmosphere, in units of watts per square meter (W/m2). The forecast moment label refers to a corresponding forecast moment within the target time period. The forecast clearness index is intended to describe the influence on solar shortwave radiation from the atmosphere, and is a ratio of the total solar radiation incident onto a horizontal plane to the astronomical radiation within the taiget time period, in units of Joules per square meter (J/m2). The weathertype refers to the type of weather within the target time period, e.g., rainy, sunny, cloudy, etc. The forecast temperature refers to the range of temperature within the target time period, in units of Celsius (°C). The forecast rainfall refers to the depth to which rainfall accumulates on the water surface within the target time period. The forecast cloud cover refers to the percentage of the sky view obscured by clouds within the target time period. The forecast zenith angle refers to an angular distance between a celestial body and the zenith within the target time period, in units of °. The forecast AM and PM label refer to AM and PM properties of the time within the target time period.
[0070] In step 308, ith reference forecast optical power for the target time period is acquired by inputting an ith corrected forecast parameter and ith intermediate forecast data for the target time period into the 12th machine learning model in the second forecasting layer.
[0071] In some embodiments, the computer device acquires the following two types of data: the ith corrected forecast parameter for the target time period, optimized by the first forecasting layer, and the ith intermediate forecast data for the target time period, and acquires the ith reference forecast optical power for the target time period by inputting these two types of data into the 12th machine learning model in the second forecasting layer. [0072] The ith intermediate forecast data is data in the ith forecast data that is to be input into the i2th machine learning model in the second forecasting layer. That is, the ith intermediate forecast data is part or all of the ith forecast data acquired from the meteorological source.
[0073] In some embodiments, the ith corrected forecast parameter includes ith corrected forecast irradiation. It can be understood that since the accuracy of forecast irradiation has a great impact on the accuracy of the forecast optical power, an explanation is made by taking the ith corrected forecast parameter including the ith corrected forecast irradiation as an example. That is, in step 306, the computer device acquires more accurate corrected forecast irradiation by inputting the forecast data into the first forecasting layer and using the machine learning model in the first forecasting layer to process the optimal forecast irradiation.
[0074] In some embodiments, the ith intermediate forecast data includes at least one of the following forecast data: ith forecast irradiation and an ith forecast moment label. Optionally, the ith intermediate forecast data further includes at least one of the following forecast data: an ith forecast clearness index and an ith weather type.
[0075] In some embodiments, the ith intermediate forecast data includes: ith forecast irradiation and an ith forecast moment label; and the ith forecast data includes: an ith forecast clearness index and an ith weather type.
[0076] Understandably, those skilled in the art who understand the technical solutions according to the embodiments of the present disclosure may readily contemplate that there may be a plurality of first forecasting layers, and a plurality of more accurate corrected forecast parameters may be acquired by optimizing various forecast parameters in the forecast data using the plurality of first forecasting layers, so as to complete all or part of the functions described in the embodiments of the present disclosure, all of which should fall within the scope of protection of the present disclosure.
[0077] In step 310, corrected forecast optical power is acquired by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.
[0078] For details about this step, reference may be made to step 208, which is not repeated herein.
[0079] In summary, in the method according to the embodiment of the present disclosure, the optical power forecasting layer is divided into a first forecasting layer and a second forecasting layer, and a more accurate corrected forecast parameter is acquired by optimizing a forecast parameter in the forecast data via the first forecasting layer, such that the more accurate corrected forecast parameter may be input into the second forecasting layer to forecast optical power with higher accuracy. [0080] In some embodiments, the optical power correcting layer includes a regression model. [0081] FIG. 4 is a flowchart of a method for forecasting optical power according to an exemplary embodiment of the present disclosure. Description is made by taking the method being applied to the computer device in the system for forecasting optical power shown in FIG. 1 as an example. The method includes the following steps.
[0082] In step 402, n forecast data for a target time period is acquired from n meteorological sources respectively, wherein n is a positive integer greater than 1.
[0083] For details about this step, reference may be made to step 202, which is not repeated herein.
[0084] In step 404, a forecasting model is called, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
[0085] For details about this step, reference may be made to step 204, which is not repeated herein.
[0086] In step 406, ith reference forecast optical power for the target time period is acquired by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n. [0087] For details about this step, reference may be made to step 206, which is not repeated herein.
[0088] In step 408, corrected forecast optical power for the target time period is acquired by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the regression model of the optical power correcting layer for regression algorithm processing.
[0089] In some embodiments, the computer device acquires the corrected forecast optical power for the target time period by inputting the n reference forecast optical power into the regression model in the optical power correcting layer for regression algorithm processing, in response to acquiring the n reference forecast optical power by the n machine learning processing modules in the optical power forecasting layer.
[0090] In some embodiments, the regression model is of at least one of the following types: a linear regression model, a ridge regression model, and a least absolute shrinkage and selection operator (LASSO) regression model.
[0091] In summary, in the method according to the embodiment of the present disclosure, the optical power correcting layer includes the regression model for regression algorithm processing of the n reference forecast optical power output by the optical power forecasting layer, such that the final corrected forecast optical power is acquired by synthesizing the n reference forecast optical power, which improves the accuracy in forecasting optical power.
[0092] Hereinafter, the method for forecasting optical power according to the present disclosure is exemplarily described with reference to the following embodiment.
[0093] In this embodiment, the forecasting model includes three layers, namely a first layer of the forecasting model (also known as the first forecasting layer in the embodiments of the present disclosure), a second layer of the forecasting model (also known as the second forecasting layer in the embodiments of the present disclosure), and a third layer of the forecasting model (also known as the optical power correcting layer in the embodiments of the present disclosure).
[0094] As shown in FIG. 5, the first layer of the forecasting model includes n XGB models, n is a positive integer greater than 1, and the XGB models in the first layer of the forecasting model aim to optimize forecast radiation. There are n meteorological sources, namely a meteorological source 1 to a meteorological source n, and each meteorological source inputs corresponding forecast data into the corresponding XGB model, for example, the meteorological source 1 inputs first forecast data into the corresponding XGB model, and the meteorological source n inputs nth forecast data into the corresponding XGB model. Each XGB model outputs corrected forecast radiation, for example, the XGB model corresponding to the meteorological source 1 outputs first corrected forecast irradiation, and the XGB model corresponding to the meteorological source n outputs nth corrected forecast radiation.
[0095] The second layer of the forecasting model also includes n XGB models, wherein the XGB models in the second layer of the forecasting model are intended to optimize the forecast optical power. An ith XGB model in the second layer of the forecasting model includes the following input data: ith corrected forecast irradiation acquired by the first layer of the forecasting model, and ith intermediate forecast data of ith forecast data corresponding to a meteorological source i. For example, a first XGB model in the second layer of the forecasting model includes the following input data: first corrected forecast irradiation and first intermediate forecast data, and an nth XGB model in the second layer of the forecasting model includes nth corrected forecast irradiation and nth intermediate forecast data. Each XGB model outputs the reference forecast optical power, e.g., the XGB model corresponding to the meteorological source 1 outputs first reference forecast optical power, and the XGB model corresponding to the meteorological source n outputs nth reference forecast optical power.
[0096] The third layer of the forecasting model includes one linear regression model, wherein the linear regression model is configured to acquire final corrected forecast optical power as a result of an optical power forecast by conducting a linear regression operation on the n reference forecast optical power corresponding to the n meteorological sources acquired by the second layer of the forecasting model.
[0097] It can be understood that the above method embodiments may be practiced individually or in combination, which is not limited herein.
[0098] Described hereinafter are apparatus embodiments of the present disclosure. For details that are not described in detail in the apparatus embodiments, reference may be made to the corresponding records in the above method embodiments, which is not repeated herein.
[0099] FIG. 6 is a schematic structural diagram of an apparatus for forecasting optical power according to an exemplary embodiment of the present disclosure. The apparatus may be implemented as all or a part of a computer device through software, hardware or a combination of the two. The apparatus includes a forecast data acquiring module 601, a model calling module 602, an optical power forecasting module 603, and an optical power correcting module 604. [00100] The forecast data acquiring module 601 is configured to acquire n forecast data for a target time period from n meteorological sources respectively, wherein n is a positive integer greater than 1.
[00101] The model calling module 602 is configured to call a forecasting model, wherein the forecasting model includes an optical power forecasting layer and an optical power correcting layer, wherein the optical power forecasting layer includes n machine learning processing modules.
[00102] The optical power forecasting module 603 is configured to acquire ith reference forecast optical power for the target time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, wherein i is a positive integer not greater than n.
[00103] The optical power correcting module 604 is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer for regression algorithm processing.
[00104] In some embodiments, the optical power forecasting layer includes a first forecasting layer and a second forecasting layer, and the ith machine learning processing module in the optical power forecasting layer includes an iith machine learning model in the first forecasting layer and an 12th machine learning model in the second forecasting layer.
[00105] The optical power forecasting module 603 is configured to acquire an ith corrected forecast parameter for the target time period by inputting ith forecast data for the target time period into the iith machine learning model in the first forecasting layer, and acquire ith reference forecast optical power for the target time period by inputting the ith corrected forecast parameter and ith intermediate forecast data for the target time period into the 12th machine learning model in the second forecasting layer. The ith intermediate forecast data is data in the ith forecast data that is to be input into the 12th machine learning model in the second forecasting layer.
[00106] In some embodiments, the ith corrected forecast parameter includes ith corrected forecast irradiation.
[00107] In some embodiments, the ith intermediate forecast data includes at least one of: ith forecast irradiation and an ith forecast moment label.
[00108] In some embodiments, the ith intermediate forecast data further includes at least one of: an ith forecast clearness index and an ith weather type.
[00109] In some embodiments, the ith forecast data includes at least one of: ith forecast irradiation, an ith forecast temperature, and an ith forecast moment label.
[00110] In some embodiments, the ith forecast data further includes at least one of: ith forecast rainfall, ith forecast cloud cover, an ith forecast zenith angle, an ith forecast AM and PM label, an ith forecast clearness index, and an ith weather type.
[00111] In some embodiments, the optical power correcting layer includes a regression model. [00112] The optical power correcting module 604 is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the regression model of the optical power correcting layer for regression algorithm processing.
[00113] In some embodiments, the regression model is at least one of: a linear regression model, a ridge regression model, and a least absolute shrinkage and selection operator (LASSO) regression model.
[00114] In some embodiments, the meteorological sources include at least two of: deterministic forecasts provided by the ECMWF, ensemble forecasts provided by the ECMWF, deterministic forecasts provided by the International Business Machines Corporation (IBM), and the WRF. [00115] In some embodiments, the machine learning models in the n machine learning processing modules in the optical power forecasting layer may be of at least one of: an XGB model; a light GBM model, a GBDT model, an RF model, and a neural network model.
[00116] FIG. 7 is a schematic structural diagram of a computer device according to an embodiment of the present disclosure. Specifically, the computer device 700 includes a central processing unit (CPU) 701, a system memory 704 including a random-access memory (RAM) 702 and a read-only memory (ROM) 703, and a system bus 705 connecting the system memory 704 and the CPU 701. The computer device 700 further includes a basic input/output (FO) system 706 configured to help transmit information among various components within the computer device, and a mass storage device 707 configured to store an operating system 713, an application 714, and other program modules 715.
[00117] The basic I/O system 706 includes a display 708 configured to display information and an input device 709, for example, a mouse, a keyboard, and the like, configured for the user to input information. Both the display 708 and the input device 709 are connected to the CPU 701 by an input/output controller 710 connected to the system bus 705. The basic I/O system 706 may also include the input/output controller 710 configured to receive and processing inputs from a plurality of other devices, such as the keyboard, the mouse, or an electronic stylus. Similarly, the input/output controller 710 is further configured to provide outputs to the display, a printer or other types of output devices.
[00118] The mass storage device 707 is connected to the CPU 701 by a mass storage controller (not shown) connected to the system bus 705. The mass storage device 707 and a computer-readable medium associated therewith provide non-volatile storage for the computer device 700. That is, the mass storage device 707 may include the computer-readable medium (not shown), such as a hard disk or a compact disc read-only memory (CD-ROM) driver.
[00119] Without loss of generality, the computer-readable medium may include a computer storage medium and a communication medium. The computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as a computer-readable instruction, a data structure, a program module or other data. The computer storage medium includes a RAM, a ROM, an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory or other solid-state memory technologies; a CD-ROM, a digital versatile disc (DVD) or other optical storage; and a tape cassette, a magnetic tape, a magnetic disk storage or other magnetic storage devices. It will be known by a person skilled in the art that the computer storage medium is not limited to above. The above system memory 704 and the mass storage device 707 may be collectively referred to as the memory. [00120] According to various embodiments of the present disclosure, the computer device 700 may also be operated by being connected by a network such as the Internet to a remote computer. That is, the computer device 700 may be connected to the network 712 by a network interface unit 711 connected to the system bus 705, or that is, the computer device 700 may be connected to other types of networks or remote computer systems (not shown) by using the network interface unit 711.
[00121] An embodiment of the present disclosure further provides a non-transitory computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above method embodiments.
[00122] An embodiment of the present disclosure further provides a computer program product or a computer program. The computer program product or the computer program includes one or more computer instructions stored in a computer-readable storage medium, wherein the one or more computer instructions, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power according to the above embodiments.
[00123] The term "plurality" herein refers to two or more. The term "and/or" herein describes the associated relationship of the associated objects, indicating three relationships. For example, A and/or B may indicate that: A exists alone, A and B exist concurrently, B exists alone. The symbol "/" generally indicates that the contextual objects are in an "or" relationship.
[00124] A person of ordinary skill in the art may understand that all or part of the steps in the above embodiments may be completed through hardware, or through relevant hardware instructed by a program stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disc.
[00125] Described above are only optional embodiments of the present disclosure, and are not intended to limit the present disclosure. Any modifications, equivalent substitutions, improvements and the like made should be included within the scope of protection of the present disclosure, without departing from the spirit and principles of the present disclosure.

Claims

CLAIMS What is claimed is:
1. A method for forecasting optical power, comprising: acquiring n forecast data for a target time period from n meteorological sources respectively, n being a positive integer greater than 1 ; calling a forecasting model, the forecasting model comprising an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer comprising n machine learning processing modules; acquiring ith reference forecast optical power for the taiget time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n; and acquiring corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.
2. The method according to claim 1, wherein the optical power forecasting layer comprises a first forecasting layer and a second forecasting layer, and the ith machine learning processing module in the optical power forecasting layer comprises an iith machine learning model in the first forecasting layer and an 12th machine learning model in the second forecasting layer; and acquiring the ith reference forecast optical power for the target time period by inputting the ith forecast data for the target time period into the ith machine learning processing module in the optical power forecasting layer comprises: acquiring an ith corrected forecast parameter for the target time period by inputting ith forecast data for the target time period into the iith machine learning model in the first forecasting layer; and acquiring ith reference forecast optical power for the target time period by inputting the ith corrected forecast parameter and ith intermediate forecast data for the target time period into the 12th machine learning model in the second forecasting layer, the ith intermediate forecast data being data in the ith forecast data that is to be input into the 12th machine learning model in the second forecasting layer.
3. The method according to claim 2, wherein the ith corrected forecast parameter comprises ith corrected forecast irradiation.
4. The method according to claim 2, wherein the ith intermediate forecast data comprises at least one of: ith forecast irradiation and an ith forecast moment label.
5. The method according to claim 4, wherein the ith intermediate forecast data further comprises at least one of: an ith forecast clearness index and an ith weather type.
6. The method according to any one of claims 1 to 5, wherein the ith forecast data comprises at least one of: ith forecast irradiation, an ith forecast temperature, an ith forecast moment label, ith forecast rainfall, ith forecast cloud cover, an ith forecast zenith angle, an ith forecast AM and PM label, an ith forecast clearness index, and an ith weather type.
7. The method according to any one of claims 1 to 5, wherein the optical power correcting layer comprises a regression model; and acquiring the corrected forecast optical power by inputting the first reference forecast optical power to the nth reference forecast optical power for the target time period into the optical power correcting layer comprises: acquiring corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the regression model of the optical power correcting layer for regression algorithm processing.
8. An apparatus for forecasting optical power, comprising: a forecast data acquiring module, a model calling module, an optical power forecasting module, and an optical power correcting module; wherein the forecast data acquiring module is configured to acquire n forecast data for a taiget time period from n meteorological sources respectively, n being a positive integer greater than 1 ; the model calling module is configured to call a forecasting model, the forecasting model comprising an optical power forecasting layer and an optical power correcting layer, the optical power forecasting layer comprising n machine learning processing modules; the optical power forecasting module is configured to acquire ith reference forecast optical power for the target time period by inputting ith forecast data for the target time period into an ith machine learning processing module in the optical power forecasting layer, i being a positive integer not greater than n; and the optical power correcting module is configured to acquire corrected forecast optical power by inputting first reference forecast optical power to nth reference forecast optical power for the target time period into the optical power correcting layer.
9. A computer device, comprising a processor and a memory storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the processor, when loading and executing the at least one instruction, at least one program, the code set, or the instruction set, is caused to perform the method for forecasting optical power as defined in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium storing at least one instruction, at least one program, a code set, or an instruction set therein, wherein the at least one instruction, the at least one program, the code set or the instruction set, when loaded and executed by a processor of a computer device, causes the computer device to perform the method for forecasting optical power as defined in any one of claims 1 to 7.
PCT/SG2022/050533 2021-07-27 2022-07-26 Method and apparatus for forecasting optical power, computer device and storage medium WO2023009070A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202110851492.5 2021-07-27
CN202110851492.5A CN113610287A (en) 2021-07-27 2021-07-27 Optical power forecasting method and device, computer equipment and storage medium

Publications (2)

Publication Number Publication Date
WO2023009070A2 true WO2023009070A2 (en) 2023-02-02
WO2023009070A3 WO2023009070A3 (en) 2023-05-25

Family

ID=78305622

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/SG2022/050533 WO2023009070A2 (en) 2021-07-27 2022-07-26 Method and apparatus for forecasting optical power, computer device and storage medium

Country Status (2)

Country Link
CN (1) CN113610287A (en)
WO (1) WO2023009070A2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115079306B (en) * 2022-05-23 2023-12-05 远景智能国际私人投资有限公司 Lightning prediction method, device, apparatus, storage medium and computer program product

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101769788B (en) * 2009-12-29 2012-01-04 青海国泰节能技术研究院 Method for forecasting optical output power and electric energy production of photovoltaic power station
CN102411729B (en) * 2011-11-04 2013-10-09 国电南京自动化股份有限公司 Wind power prediction method based on adaptive linear logic network
CN107169683A (en) * 2017-06-30 2017-09-15 上海电力学院 A kind of grid-connected photovoltaic power station short term power combination forecasting method of variable weight coefficient
CN109583467A (en) * 2018-10-11 2019-04-05 中国电力科学研究院有限公司 A kind of electric power meteorological data fusion method and system
CN109543721B (en) * 2018-11-05 2023-07-04 中国科学院西北生态环境资源研究院 Solar irradiance ultra-short-term forecasting method under sunny cloudy weather condition
CN110263998B (en) * 2019-06-19 2024-02-13 大唐（赤峰）新能源有限公司 Double-layer correction method for multisource numerical weather forecast set
KR102063358B1 (en) * 2019-10-22 2020-03-02 (주)한국해양기상기술 Learning method and testing method for generating high-resolution weather and climate data, and learning device and testing device using the same
CN111091138A (en) * 2019-11-14 2020-05-01 远景智能国际私人投资有限公司 Irradiation forecast processing method and stacked generalization model training method and device
CN111369070B (en) * 2020-03-13 2023-06-27 西安理工大学 Multimode fusion photovoltaic power prediction method based on envelope clustering
CN111507512B (en) * 2020-04-10 2023-11-24 阳光慧碳科技有限公司 Correction method and device for photovoltaic ultra-short-term predicted power
CN112418558B (en) * 2020-12-07 2022-03-11 天津大学 Total radiation correction method based on multi-source meteorological forecast
CN112906757A (en) * 2021-01-29 2021-06-04 广东晨兴智能科技有限公司 Model training method, optical power prediction method, device, equipment and storage medium
CN113408785B (en) * 2021-05-20 2023-04-07 上海晨翘智能科技有限公司 Method, device, equipment and storage medium for predicting optical power

Also Published As

Publication number Publication date
CN113610287A (en) 2021-11-05
WO2023009070A3 (en) 2023-05-25

Similar Documents

Publication Publication Date Title
US9389630B2 (en) Predictive smart grid re-structuring based on expected load and power generation
US10326280B1 (en) Distributed renewable energy grid controller
US20200067316A1 (en) Virtualization of large-scale energy storage
CN113075940B (en) Photovoltaic string tracking support control method and related device
US20220376499A1 (en) System and method for load and source forecasting for increasing electrical grid component longevity
Shi et al. Multistage robust energy management for microgrids considering uncertainty
CN113570126A (en) Method, device and system for predicting power generation power of photovoltaic power station
WO2023009070A2 (en) Method and apparatus for forecasting optical power, computer device and storage medium
Saberi et al. Probabilistic congestion driven network expansion planning using point estimate technique
Nikoobakht et al. Integrated transmission and storage systems investment planning hosting wind power generation: continuous‐time hybrid stochastic/robust optimisation
US11842303B2 (en) Method for processing irradiation forecast, method for training stacked generalization model, and apparatuses thereof
CN113283492B (en) Photovoltaic output scene generation method and scene generation device based on Catboost algorithm
CN113408785B (en) Method, device, equipment and storage medium for predicting optical power
Zhu et al. Multi‐time‐scale robust economic dispatching method for the power system with clean energy
Pérez et al. Reduction of energy losses through the integration of photovoltaic power plants in distribution networks
CN116070758A (en) Photovoltaic power generation short-term power prediction method, device and storage medium
CN113178896B (en) Method and system for configuring installed capacity of fixed-output light-storage combined power station
Hasan et al. Neural network-based solar irradiance forecast for peak load management of grid-connected microgrid with photovoltaic distributed generation
WO2020125984A1 (en) System and method for power outage prediction
US20210351612A1 (en) Solar inverter power output communications methods, and related computer program products
Wang et al. A distributed big data storage and data mining framework for solar-generated electricity quantity forecasting
Jiang et al. A dual timescale active power coordinated scheduling framework for wind integrated power system in the presence of storage and wind forecast uncertainties
CN116402324B (en) Renewable energy management method, device, medium and electronic equipment
CN113988491B (en) Photovoltaic short-term power prediction method and device, electronic equipment and storage medium
CN116470574A (en) Energy large-base stable power transmission method, system, device and medium

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE