Photovoltaic power prediction methods, devices, equipment, dielectrics and processes products

Abstract

This disclosure provides an exemplary embodiment of a photovoltaic power prediction method, apparatus, device, medium, and program product. The method includes: acquiring and preprocessing historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power plant structure data for the area where the target photovoltaic power plant is located; constructing a cloud shading state vector; predicting the evolution process of the cloud shading state to generate a cloud shading propagation state sequence; determining the effective irradiance state and power response for the future time period to obtain a future power response sequence; and performing physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate an ultra-short-term photovoltaic power prediction output result. This disclosure integrates multi-source data, improving the quality of photovoltaic power prediction, and improves the accuracy of ultra-short-term photovoltaic power prediction by performing photovoltaic power prediction based on cloud shading propagation.

Description

Technical Field

[0001] This disclosure relates to the field of photovoltaic power prediction technology, and in particular to a photovoltaic power prediction method, apparatus, equipment, medium, and program product. Background Technology

[0002] Currently, photovoltaic power prediction mainly relies on statistical analysis of historical power, data-driven prediction through machine learning, and prediction by integrating meteorological elements. However, in typical ultra-short-term scenarios such as cloudy skies, rapid local cloud passage, and cloud edge enhancement, there are problems such as a lack of characterization of photovoltaic power mutation causes, difficulty in effectively describing the spatial propagation process of cloud cover, insufficient rapid fluctuation response capability in ultra-short-term scenarios, and unclear prediction chains. Summary of the Invention

[0003] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a photovoltaic power prediction method, apparatus, equipment, medium, and program product.

[0004] An exemplary embodiment of this disclosure provides a photovoltaic power prediction method, the method comprising: Acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data, and power station structure data of the area where the target photovoltaic power station is located; Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, feature extraction is performed to construct a cloud obscuring state vector; Based on the cloud occlusion state vector of historical time periods, the evolution process of cloud occlusion state in future time periods is predicted, and a cloud occlusion propagation state sequence for the future time periods is generated. Based on the historical photovoltaic power data, the cloud obstruction propagation state sequence, the future period's irradiance meteorological data, astronomical location data, and power plant structure data, the effective irradiance state and power response for the future period are determined, and a future power response sequence is obtained. The future power response sequence is corrected by physical boundary constraints and temporal continuity verification to generate the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station.

[0005] In some exemplary embodiments, the historical photovoltaic power data includes the station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power plant; The ground-based cloud image data includes a continuous sequence of sky images; The irradiance meteorological data shall include at least total irradiance, diffuse irradiance, direct irradiance, ambient temperature, component temperature, relative humidity, wind speed, and wind direction; The astronomical location data includes at least the solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance values ​​for the predicted time and future periods. The power station structure data includes the location coordinates of the photovoltaic array, the arrangement of the components, the tilt angle and orientation of the components, the row and column relationship of the array, the mapping relationship between the inverter and the array, and the spatial information of the station area; The acquisition and preprocessing of historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power station structure data for the target photovoltaic power station area includes: Based on five types of data channels, the historical photovoltaic power data, the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data are acquired; Based on the preset standardized data preprocessing, the historical photovoltaic power data, the ground cloud map data, the irradiance meteorological data, the astronomical location data and the power station structure data are preprocessed. The standardized data preprocessing includes data format unification, time alignment, spatial alignment and data quality labeling.

[0006] In some exemplary embodiments, the step of extracting features based on the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data to construct a cloud obscuring state vector includes: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, cloud obstruction intensity characteristics, cloud boundary morphology characteristics, cloud movement and propagation characteristics, and cloud shadow projection coverage characteristics are extracted. Based on the cloud occlusion intensity characteristics, cloud boundary morphology characteristics, cloud motion propagation characteristics, and cloud shadow projection coverage characteristics, a cloud occlusion state vector is constructed.

[0007] In some exemplary embodiments, the prediction of the evolution of cloud occlusion state in future periods based on the cloud occlusion state vector of historical time periods, generating a cloud occlusion propagation state sequence for the future periods, includes: Construct a historical state sequence based on the cloud occlusion state vectors of historical time periods; Based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed to determine the propagation trend in future periods; Based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is extrapolated step by step to obtain the future cloud occlusion state vector at the predicted time. Based on the solar position parameters for future time periods, the cloud shadow projection coverage features in the future cloud occlusion state vector are dynamically corrected to generate the cloud occlusion propagation state sequence for the future time periods.

[0008] In some exemplary embodiments, determining the effective irradiance state and power response for the future period based on the historical photovoltaic power data, the cloud cover propagation state sequence, the future irradiance meteorological data, astronomical location data, and power plant structure data, to obtain the future power response sequence, includes: Based on the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, and the astronomical location data for the future period, the effective irradiance state for the future period is calculated. Based on the power station structure data, the spatial sub-regions of the photovoltaic power station are determined, and based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of the spatial sub-regions in the future time period is determined. Based on the effective irradiance state, the irradiated state, and the historical photovoltaic power data, the effective irradiance state and power response for the future time period are determined through comprehensive mapping or partition aggregation, thereby obtaining the future power response sequence for the future time period.

[0009] In some exemplary embodiments, the physical boundary constraints include nonnegativity constraints, nominal capacity constraints, solar position constraints, and irradiance consistency constraints; The process of performing physical boundary constraint correction and temporal continuity verification on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station includes: Based on the physical boundary constraints, the future power response sequence is modified according to the physical boundary constraints. Based on the preset power change range, the power change amplitude between adjacent prediction times is checked and corrected for temporal continuity. Based on the future power response sequence, a time-series verification and correction are performed on abnormal jumps that do not match the cloud obstruction propagation state sequence. Based on the physical boundary constraint correction and the temporal continuity verification correction, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station is generated.

[0010] An exemplary embodiment of this disclosure also provides a photovoltaic power prediction device, the device comprising: The acquisition module is configured to acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data, and power station structure data of the area where the target photovoltaic power station is located; The construction module is configured to extract features based on the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data to construct a cloud obscuring state vector; The first generation module is configured to predict the evolution of cloud occlusion state in future periods based on cloud occlusion state vectors in historical periods, and generate a cloud occlusion propagation state sequence for the future periods. The module is configured to determine the effective irradiance state and power response for the future period based on the historical photovoltaic power data, the cloud shading propagation state sequence, the irradiance meteorological data for the future period, the astronomical location data, and the power station structure data, and obtain the future power response sequence. The second generation module is configured to perform physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output result of the target photovoltaic power station.

[0011] The apparatus provided in the exemplary embodiments of this disclosure includes historical photovoltaic power data, which includes station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power plant. The ground-based cloud image data includes a continuous sequence of sky images; The irradiance meteorological data shall include at least total irradiance, diffuse irradiance, direct irradiance, ambient temperature, component temperature, relative humidity, wind speed, and wind direction; The astronomical location data includes at least the solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance values ​​for the predicted time and future periods. The power station structure data includes the location coordinates of the photovoltaic array, the arrangement of the components, the tilt angle and orientation of the components, the row and column relationship of the array, the mapping relationship between the inverter and the array, and the spatial information of the station area; The acquisition module is specifically used for: Based on five types of data channels, the historical photovoltaic power data, the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data are acquired; Based on the preset standardized data preprocessing, the historical photovoltaic power data, the ground cloud map data, the irradiance meteorological data, the astronomical location data and the power station structure data are preprocessed. The standardized data preprocessing includes data format unification, time alignment, spatial alignment and data quality labeling.

[0012] The apparatus provided in the exemplary embodiments of this disclosure, wherein the construction module is specifically used for: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, cloud obstruction intensity characteristics, cloud boundary morphology characteristics, cloud movement and propagation characteristics, and cloud shadow projection coverage characteristics are extracted. Based on the cloud occlusion intensity characteristics, cloud boundary morphology characteristics, cloud motion propagation characteristics, and cloud shadow projection coverage characteristics, a cloud occlusion state vector is constructed.

[0013] The apparatus provided in the exemplary embodiments of this disclosure includes a first generation module specifically used for: Construct a historical state sequence based on the cloud occlusion state vectors of historical time periods; Based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed to determine the propagation trend in future periods; Based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is extrapolated step by step to obtain the future cloud occlusion state vector at the predicted time. Based on the solar position parameters for future time periods, the cloud shadow projection coverage features in the future cloud occlusion state vector are dynamically corrected to generate the cloud occlusion propagation state sequence for the future time periods.

[0014] The apparatus provided in this exemplary embodiment of the present disclosure, wherein the obtaining module is specifically used for: Based on the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, and the astronomical location data for the future period, the effective irradiance state for the future period is calculated. Based on the power station structure data, the spatial sub-regions of the photovoltaic power station are determined, and based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of the spatial sub-regions in the future time period is determined. Based on the effective irradiance state, the irradiated state, and the historical photovoltaic power data, the effective irradiance state and power response for the future time period are determined through comprehensive mapping or partition aggregation, thereby obtaining the future power response sequence for the future time period.

[0015] The apparatus provided in the exemplary embodiments of this disclosure includes physical boundary constraints such as nonnegativity constraints, rated capacity constraints, solar position constraints, and irradiance consistency constraints. The second generation module is specifically used for: Based on the physical boundary constraints, the future power response sequence is modified according to the physical boundary constraints. Based on the preset power change range, the power change amplitude between adjacent prediction times is checked and corrected for temporal continuity. Based on the future power response sequence, a time-series verification and correction are performed on abnormal jumps that do not match the cloud obstruction propagation state sequence. Based on the physical boundary constraint correction and the temporal continuity verification correction, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station is generated.

[0016] An exemplary embodiment of this disclosure also provides an electronic device, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the photovoltaic power prediction method as provided in the exemplary embodiment of this disclosure.

[0017] This disclosure also provides a computer-readable storage medium storing a computer program for performing a photovoltaic power prediction method as provided in this disclosure.

[0018] The technical solution provided by the exemplary embodiments of this disclosure has the following advantages compared with the prior art: The photovoltaic power prediction method provided in this exemplary embodiment acquires and preprocesses historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power station structure data for the target photovoltaic power station area; constructs a cloud shading state vector; predicts the evolution of the cloud shading state to generate a cloud shading propagation state sequence; determines the effective irradiance state and power response for the future time period to obtain a future power response sequence; and performs physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate an ultra-short-term photovoltaic power prediction output result. This disclosure integrates multi-source data, improves the quality of photovoltaic power prediction, and improves the accuracy of ultra-short-term photovoltaic power prediction based on cloud shading propagation. Attached Figure Description

[0019] The above and other features, advantages, and aspects of the exemplary embodiments of this disclosure will become more apparent when taken in conjunction with the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0020] Figure 1 A schematic flowchart of a photovoltaic power prediction method provided as an exemplary embodiment of the present disclosure; Figure 2 A schematic diagram of a photovoltaic power prediction device provided as an exemplary embodiment of the present disclosure; Figure 3 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of the present disclosure. Detailed Implementation

[0021] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While some exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be construed as limited to the exemplary embodiments set forth herein. Rather, these exemplary embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the accompanying drawings and exemplary embodiments of the present disclosure are for illustrative purposes only and are not intended to limit the scope of protection of the present disclosure.

[0022] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0023] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one exemplary embodiment" means "at least one exemplary embodiment"; the term "another exemplary embodiment" means "at least one additional exemplary embodiment"; the term "some exemplary embodiments" means "at least some exemplary embodiments". Definitions of other terms will be given in the description below.

[0024] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0025] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0026] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0027] To address the aforementioned issues, this disclosure provides an exemplary method for predicting photovoltaic power, which will be described below with reference to specific exemplary embodiments.

[0028] Figure 1 This is a flowchart illustrating a photovoltaic power prediction method provided as an exemplary embodiment of the present disclosure. The method can be executed by a photovoltaic power prediction device, which can be implemented using software and / or hardware and is generally integrated into an electronic device.

[0029] like Figure 1As shown in the exemplary embodiment of this disclosure, a photovoltaic power prediction method includes: Step S101: Acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data and power station structure data of the area where the target photovoltaic power station is located.

[0030] In some exemplary embodiments, the historical photovoltaic power data includes the station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power plant; The ground-based cloud image data includes a continuous sequence of sky images; The irradiance meteorological data shall include at least total irradiance, diffuse irradiance, direct irradiance, ambient temperature, component temperature, relative humidity, wind speed, and wind direction; The astronomical location data includes at least the solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance values ​​for the predicted time and future periods. The power station structure data includes the location coordinates of the photovoltaic array, the arrangement of the components, the tilt angle and orientation of the components, the row and column relationship of the array, the mapping relationship between the inverter and the array, and the spatial information of the station area; The acquisition and preprocessing of historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power station structure data for the target photovoltaic power station area includes: Based on five types of data channels, the historical photovoltaic power data, the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data are acquired; Based on the preset standardized data preprocessing, the historical photovoltaic power data, the ground cloud map data, the irradiance meteorological data, the astronomical location data and the power station structure data are preprocessed. The standardized data preprocessing includes data format unification, time alignment, spatial alignment and data quality labeling.

[0031] In some exemplary embodiments, each type of data channel corresponds to different data sources, sampling methods, data formats, and time resolutions.

[0032] In some exemplary embodiments, historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power plant structure data differ in terms of data structure, sampling frequency, and data format. Therefore, they are processed uniformly based on preset standardized data preprocessing.

[0033] In some exemplary embodiments, historical photovoltaic power data is used to characterize the actual output response of a photovoltaic power plant under different irradiation conditions and different operating states.

[0034] In some exemplary embodiments, ground-based cloud image data is the core data characterizing the propagation state of cloud obstruction; it can provide information on the spatial distribution of clouds above photovoltaic power stations, cloud boundary morphology, solar eclipse status, cloud thickness variations, and continuous cloud movement.

[0035] In some exemplary embodiments, irradiance meteorological data are used to reflect the surface light reception status under cloud cover and the external environmental conditions for component operation.

[0036] In some exemplary embodiments, astronomical location data are used to reflect the theoretical light reception conditions under solar radiation incidence and cloudless conditions.

[0037] In some exemplary embodiments, power plant structure data is used to represent the relative positional relationship and the order of obstruction of different areas within the power plant during the cloud shadow propagation process.

[0038] In some exemplary embodiments, historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power plant structure data are uniformly converted into standard data formats through data format unification, which facilitates subsequent unified processing and correlation analysis.

[0039] In some exemplary embodiments, data whose data quality labeling results are unqualified are not included in subsequent processing.

[0040] In some exemplary embodiments, historical photovoltaic power data is acquired through data channel A. The historical photovoltaic power data includes the station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power station. The data is acquired in real time through the power station SCADA system, inverter monitoring platform, or edge acquisition terminal.

[0041] In some exemplary embodiments, ground-based cloud image data is acquired through data channel B. The ground-based cloud image data includes a continuous sequence of sky images acquired by a ground-based all-sky imager, a ground-based cloud camera, or a cloud observation device.

[0042] In some exemplary embodiments, irradiance meteorological data is acquired through data channel C. The irradiance meteorological data includes total irradiance (GHI), diffuse irradiance (DHI), direct irradiance (DNI), ambient temperature, component temperature, relative humidity, wind speed, wind direction, and other data collected in real time by the on-site meteorological station.

[0043] In some exemplary embodiments, astronomical position data is acquired through data channel D. The astronomical position data includes information such as solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance during the predicted time and future predicted period, and is obtained by astronomical calculation models, etc.

[0044] In some exemplary embodiments, power plant structure data is acquired via data channel E.

[0045] In some exemplary embodiments, to predict time Based on this, a unified time window is constructed. Align data from different data channels to a uniform time resolution. ; Indicates a historical period; To represent future time periods, the length of historical time periods is determined based on the length of the future time periods. For historical photovoltaic power data and irradiance meteorological data, they are aligned to a uniform time resolution using nearest neighbor or difference methods. For ground-based cloud map data, by selecting distance Recent Consecutive Frame image sequences are used as input; for astronomical location data, through... Calculate the future sequence directly.

[0046] In some exemplary embodiments, a mapping relationship is established between the ground-based cloud image coordinates, the solar projection direction, and the power station GIS coordinates, so that the cloud areas, solar neighbor occlusion areas, and cloud boundary areas identified in the cloud image can correspond to the internal spatial areas of the power station, and further obtain the coverage position and propagation path of the cloud shadow relative to the power station array and string area.

[0047] As an example, the sampling frequency for historical photovoltaic power data can be 1 minute, 5 minutes, or 15 minutes; the data format for ground-based cloud map data can be JPEG, PNG, TIFF, etc., with a time resolution of 5 minutes; and the sampling frequency for irradiance meteorological data can be 1 to 5 minutes.

[0048] Step S102: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, feature extraction is performed to construct a cloud obscuring state vector.

[0049] In some exemplary embodiments, the step of extracting features based on the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data to construct a cloud obscuring state vector includes: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, cloud obstruction intensity characteristics, cloud boundary morphology characteristics, cloud movement and propagation characteristics, and cloud shadow projection coverage characteristics are extracted. Based on the cloud occlusion intensity characteristics, cloud boundary morphology characteristics, cloud motion propagation characteristics, and cloud shadow projection coverage characteristics, a cloud occlusion state vector is constructed.

[0050] In some exemplary embodiments, based on ground-based cloud image data, irradiance meteorological data, astronomical location data, and power plant structure data after standardized data preprocessing, state variables reflecting cloud shading behavior and its propagation characteristics are extracted to construct a cloud shading state vector.

[0051] In some exemplary embodiments, the ground cloud map data after standardized data preprocessing is subjected to ground cloud map preprocessing to obtain a standardized ground cloud map sequence suitable for subsequent cloud area identification and cloud occlusion state analysis. Ground cloud map preprocessing includes image denoising and enhancement, distortion correction, invalid region removal, solar glare suppression, and temporal continuity check.

[0052] In some exemplary embodiments, after ground-based cloud image preprocessing is completed, cloud areas, clear sky areas, and solar neighborhood areas are identified from the standardized ground-based cloud image sequence, specifically including: Sky region extraction: Determine the boundaries of the effective sky region in the image to form a sky region mask; Cloud and clear sky classification: Based on the color, brightness, and texture features of image pixels or predefined cloud classification rules, the pixels in the sky region are classified to distinguish between cloud areas and clear sky areas. Solar Neighborhood Recognition: Based on the projection position of the sun's position parameters in the image coordinates, determine the solar neighborhood region and analyze whether the region is covered by clouds; Local cloud cluster segmentation: Connectivity analysis is performed on continuously distributed cloud regions to identify independent cloud clusters and extract their area, centroid, bounding box, and outer contour information.

[0053] Based on this, we can obtain spatial distribution information of clouds, which lays the foundation for subsequent extraction of cloud occlusion intensity features, cloud boundary morphology features, cloud motion propagation features, and cloud shadow projection coverage features.

[0054] In some exemplary embodiments, cloud shading intensity is used to characterize the ability of clouds to reduce solar irradiance on the Earth's surface at the current moment. The impact of clouds on photovoltaic power is primarily manifested as irradiance attenuation. Therefore, the spatial distribution information of clouds is further quantified into the degree of shading, specifically including: Solar Neighborhood Occlusion: A circular area with a preset neighborhood radius centered on the projection point of the sun onto the ground-based cloud map is defined as the solar neighborhood area. It is determined whether the solar neighborhood area is covered by clouds. If it is covered by clouds, the cloud coverage ratio of the solar neighborhood is calculated based on the ratio of the number of cloud pixels in the solar neighborhood area to the total pixel vector in the solar neighborhood area. Image brightness attenuation characteristics: The ground-based cloud map is converted into a grayscale cloud map. The area reflecting changes in cloud layer and illumination is defined as the main sky area. The brightness distribution changes of the solar neighborhood area, the main sky area and the overall sky area are statistically analyzed to determine the image brightness attenuation characteristics, which are used to reflect the thickness of the cloud layer and the degree of light transmission. Irradiance contrast feature extraction: The measured irradiance values ​​at the station are compared with the theoretical irradiance values ​​under clear skies to construct irradiance contrast features that reflect the current cloud cover intensity. Cloud occlusion intensity characteristics: Based on cloud coverage ratio, image brightness attenuation characteristics, and irradiance contrast characteristics, cloud occlusion intensity characteristics are determined using the following formula: ; in, Indicates the intensity characteristics of cloud cover; Indicates the percentage of cloud coverage; Weights representing the proportion of cloud coverage; Indicates the characteristics of image brightness decay; The weights representing the image brightness decay characteristics; Indicates the characteristics of irradiation contrast; This represents the weight of the irradiation contrast feature.

[0055] In the above formula, the greater the cloud shading intensity characteristic, the stronger the cloud shading, and the more obvious the reduction in the effective light reception of the photovoltaic array.

[0056] In some exemplary embodiments, photovoltaic power is prone to rapid fluctuations in ultra-short-term scenarios. These fluctuations are mainly caused by cloud boundaries crossing the sun's projected path or passing over the power station. Therefore, cloud boundary morphology features are extracted to describe the triggering effect of cloud edges on sudden power spikes and drops, specifically including: Cloud boundary contour extraction: Edge detection is performed on the identified cloud contours to obtain cloud boundary curves; Boundary position calculation: Calculate the spatial position of the cloud boundary relative to the sun's projection direction and relative to the power station's orientation; Boundary clarity quantification: The clarity of cloud boundaries is quantified by edge gradient, grayscale transition degree or texture abrupt change. The higher the boundary clarity quantification value, the clearer the boundary and the stronger the response to lighting switching. Boundary length and complexity analysis: Extract features such as boundary length, boundary curvature, and shape irregularity to reflect the complexity of cloud boundary morphology; Boundary Advancement Trend Extraction: Analyze the changes in boundary position over consecutive time periods to determine the motion pattern of the cloud boundary.

[0057] Based on this, the time The cloud boundary morphological features are denoted as The more pronounced the cloud boundary morphology and the faster the cloud boundary advances, the higher the likelihood of rapid irradiance fluctuations and power abrupt changes in the near future.

[0058] In some exemplary embodiments, cloud motion propagation features are extracted to predict subsequent future cloud occlusion propagation state sequences, specifically including: Continuous frame matching analysis: matching adjacent time points or consecutive frames The cloud map at different times is used for region matching to identify the correspondence of the same cloud cluster at different times; Displacement vector calculation: Calculate the displacement of the cloud's centroid, main boundary, or characteristic region at consecutive time intervals to obtain the cloud's direction of motion and speed of movement; Expansion and contraction analysis: By comparing the changes in cloud area over consecutive time periods, the expansion, contraction, splitting, or merging processes of the cloud body can be determined. Morphological evolution trend analysis: Statistical analysis of changes in cloud shape, boundary complexity, and expansion of local thick cloud areas to form characteristics of cloud morphological evolution trends; Main propagation path identification: If several cloud clusters exist at the same time, identify the dominant cloud cluster that has the greatest impact on the direction of the sun or the power plant area, extract its propagation path, and determine the main propagation path.

[0059] Based on this, the time The characteristics of cloud propagation direction are denoted as The cloud propagation speed characteristic is denoted as The characteristics of cloud morphological evolution are denoted as ,in, Indicates the main direction of cloud movement. This represents the displacement velocity of the cloud cluster per unit time. This indicates the expansion, weakening, aggregation, or breakup trends of cloud clusters during their movement.

[0060] In some exemplary embodiments, cloud shadow projection coverage features are extracted based on solar position parameters and power plant structure data from astronomical location data to reflect the impact of cloud shadows on photovoltaic power, specifically including: Calculation of solar projection direction: Based on the current solar altitude angle and solar azimuth angle, calculate the projection direction of the cloud layer relative to the Earth's surface; Cloud shadow location mapping: The location of cloud areas in the cloud map is mapped to the coordinate system of the power station area through the solar projection relationship; Coverage ratio calculation: Calculate the coverage ratio of the cloud shadow on the power station as a whole and in each local area; Coverage path identification: Analyze the propagation path of cloud shadows within the power plant to identify the area it first affects, the main areas it passes through, and the area it is expected to leave. Key area analysis: If the cloud cover ratio in a local area is higher than the ratio threshold or is located on the main propagation path, it will be marked as a key coverage area separately.

[0061] In some exemplary embodiments, cloud occlusion intensity features, cloud boundary morphology features, cloud motion propagation features, and cloud shadow projection coverage features are uniformly represented as a structured cloud occlusion state vector: ; in, Indicates time The cloud occlusion state vector; Indicates time The characteristics of cloud obstruction intensity; Indicates time Cloud boundary morphological characteristics; Indicates time The characteristics of cloud propagation direction; Indicates time The characteristics of cloud propagation speed; Indicates time The characteristics of cloud morphological evolution; Indicates time The cloud shadow projection coverage features.

[0062] Step S103: Based on the cloud occlusion state vector of historical time periods, predict the evolution process of cloud occlusion state in future time periods, and generate the cloud occlusion propagation state sequence for the future time periods.

[0063] In some exemplary embodiments, the prediction of the evolution of cloud occlusion state in future periods based on the cloud occlusion state vector of historical time periods, generating a cloud occlusion propagation state sequence for the future periods, includes: Construct a historical state sequence based on the cloud occlusion state vectors of historical time periods; Based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed to determine the propagation trend in future periods; Based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is extrapolated step by step to obtain the future cloud occlusion state vector at the predicted time. Based on the solar position parameters for future time periods, the cloud shadow projection coverage features in the future cloud occlusion state vector are dynamically corrected to generate the cloud occlusion propagation state sequence for the future time periods.

[0064] In some exemplary embodiments, a historical state sequence is constructed based on the cloud occlusion state vectors of historical time periods. , used to indicate at the prediction time Previous consecutive The continuous change process of cloud cover status at any given moment.

[0065] In some exemplary embodiments, based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed, including the propagation direction and stability of the cloud cluster, the propagation speed and its changing trend, the movement trend of the cloud boundary, and the movement trend of the cloud shadow, to determine the propagation trend in future periods.

[0066] In some exemplary embodiments, based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is progressively extrapolated to obtain the future cloud occlusion state vector at the predicted time: ; in, Indicates the first The future cloud occlusion state vector at each predicted time; Indicates the first Cloud occlusion intensity characteristics at each predicted time; Indicates the first Cloud boundary morphology characteristics at each predicted time point; Indicates the first Characteristics of cloud propagation direction at each predicted time; Indicates the first Cloud propagation speed characteristics at each predicted time point; Indicates the first Cloud morphology evolution characteristics at each predicted moment; Indicates the first Cloud shadow projection coverage features at each prediction time.

[0067] Based on this, the stepwise extrapolation not only considers the current displacement relationship of the cloud cluster, but also the changes in the shape of the cloud cluster and the changes in the intensity of shading, effectively reflecting the evolution process of the cloud cluster.

[0068] In some exemplary embodiments, the solar altitude angle and solar azimuth angle will continue to change in the future, which will cause the cloud shadow projection coverage on the ground and photovoltaic power station area to change accordingly. Therefore, in the process of predicting the future cloud shading state vector, the cloud shadow projection coverage characteristics at each prediction time are dynamically corrected by the solar position parameters in the future.

[0069] In some exemplary embodiments, a sequence of cloud obstruction propagation states for future time periods is generated. ,in, Indicates the first term in the future time period A sequence of cloud obstruction propagation states at each predicted time; This represents the total number of moments in the future time period. The cloud shading propagation state sequence is used to represent the impact of cloud shading on the area above and inside the photovoltaic power station in the future time period. The future cloud shading propagation state sequence reflects the changes in cloud shading intensity, boundary advancement, propagation trend, and projection coverage of the photovoltaic power station area in a short period of time (1-5 minutes), providing basic input for subsequent effective irradiance status and power response.

[0070] Step S104: Based on the historical photovoltaic power data, the cloud obstruction propagation state sequence, the future period's irradiance meteorological data, astronomical location data, and power station structure data, determine the effective irradiance state and power response for the future period, and obtain the future power response sequence.

[0071] In some exemplary embodiments, determining the effective irradiance state and power response for the future period based on the historical photovoltaic power data, the cloud cover propagation state sequence, the future irradiance meteorological data, astronomical location data, and power plant structure data, to obtain the future power response sequence, includes: Based on the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, and the astronomical location data for the future period, the effective irradiance state for the future period is calculated. Based on the power station structure data, the spatial sub-regions of the photovoltaic power station are determined, and based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of the spatial sub-regions in the future time period is determined. Based on the effective irradiance state, the irradiated state, and the historical photovoltaic power data, the effective irradiance state and power response for the future time period are determined through comprehensive mapping or partition aggregation, thereby obtaining the future power response sequence for the future time period.

[0072] In some exemplary embodiments, for the future time period... At each predicted time, based on the future cloud occlusion state vector By using the solar altitude angle, solar azimuth angle, and clear sky theoretical irradiance conditions at the corresponding time, the effective irradiance state for the future period is calculated; this is used to convert the future cloud shading intensity, boundary position, and cloud shadow coverage state into the degree of attenuation of solar radiation on the Earth's surface, thereby obtaining the effective irradiance level that the photovoltaic array may actually receive at the future time.

[0073] The formula for calculating the effective irradiance state in the future period is as follows: ; in, Indicates the first term in the future time period Effective irradiance status at each predicted time; Indicates the first term in the future time period The future cloud occlusion state vector at each predicted time; Indicates the first term in the future time period Solar position parameters at each predicted time; Indicates the first term in the future time period Theoretical clear-sky irradiance at a predicted moment; Indicates the first term in the future time period Future environmental correction parameters for each predicted time point; This represents the mapping function from the cloud obstruction propagation state sequence to the effective irradiance state.

[0074] Based on this, the effective irradiance state results after cloud obstruction at each predicted time in the future period are obtained, which are used to represent the actual impact of cloud propagation on the light-receiving conditions of the photovoltaic array.

[0075] In some exemplary embodiments, different areas within a photovoltaic power station experience varying degrees and times of shading during cloud shadow propagation. Therefore, based on the power station's structural data, a regional analysis of the future illumination status is performed, dividing the power station into several spatial sub-regions. Based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of each spatial sub-region at each predicted time in the future is determined.

[0076] In some exemplary embodiments, the effective irradiance state and power response for future time periods are determined by comprehensive mapping and partition aggregation, resulting in a future power response sequence for future time periods, as shown in the following formula: ; in, Indicates the first term in the future time period Power response at each predicted time point; Indicates the first term in the future time period Effective irradiance status at each predicted time; Indicates historical power response; Indicates the first term in the future time period Future environmental correction parameters for each predicted time point; Indicates the first term in the future time period Cloud shadow projection coverage features at each predicted time; The mapping function representing the effective irradiance state to the power response; This represents the total number of spatial sub-regions into which a photovoltaic power station is divided; Indicates the first The spatial sub-region in the future time period Power response at each predicted time point; Indicates the first The weights of each spatial sub-region; This represents the future power response sequence for a future time period.

[0077] In the above formula, if a comprehensive mapping is used, the power response of the future period is represented as a comprehensive mapping of the effective irradiance state, historical power response, future environmental correction parameters, and cloud shadow projection coverage characteristics; if a partitioned aggregation is used, the power response of the future period is represented as a weighted sum of the power responses of each spatial sub-region.

[0078] In some exemplary embodiments, in typical ultra-short-term scenarios such as rapid cloud boundary advance, sudden enhancement of local cloud clusters, or rapid changes in the solar neighborhood cloud layer, the power of the power plant often experiences short-term rapid rises and falls. Therefore, the power response at future moments can be locally corrected based on the cloud boundary advance speed, boundary clarity, cloud shadow coverage change rate, and solar neighborhood shading change magnitude at future moments. For example, if a significant rapid boundary crossing or sudden change in shading intensity is detected near a certain future moment, the power change amplitude at that moment and adjacent moments can be enhanced, making the prediction results more consistent with the sudden rises and falls caused by the rapid propagation of cloud shadows in actual operation.

[0079] Step S105: Perform physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output result of the target photovoltaic power station.

[0080] In some exemplary embodiments, the physical boundary constraints include nonnegativity constraints, nominal capacity constraints, solar position constraints, and irradiance consistency constraints; The process of performing physical boundary constraint correction and temporal continuity verification on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station includes: Based on the physical boundary constraints, the future power response sequence is modified according to the physical boundary constraints. Based on the preset power change range, the power change amplitude between adjacent prediction times is checked and corrected for temporal continuity. Based on the future power response sequence, a time-series verification and correction are performed on abnormal jumps that do not match the cloud obstruction propagation state sequence. Based on the physical boundary constraint correction and the temporal continuity verification correction, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station is generated.

[0081] In some exemplary embodiments, physical boundary constraint correction is applied to the future power response sequence, specifically including: Non-negativity constraint: the power at any prediction time must not be less than 0; Rated capacity constraints apply; the power output at any predicted time must not exceed the installed capacity of the photovoltaic power station or the theoretical maximum power output at any predicted time.

[0082] The solar position constraint automatically applies a zero-value constraint to the predicted power before sunrise, after sunset, or when the solar altitude angle is below a preset angle threshold. Irradiance consistency constraint: if the effective irradiance state in the future period is low, the power at the corresponding predicted time should remain low; if the effective irradiance state in the future period is close to clear sky, the power at the corresponding predicted time should be consistent with the theoretical upper limit of light received.

[0083] Based on this physical boundary constraint correction, excessively high, low, or non-physical power prediction results caused by local errors are eliminated.

[0084] In some exemplary embodiments, photovoltaic power may fluctuate rapidly in ultra-short-term scenarios, but its changes must satisfy temporal continuity. Therefore, temporal continuity verification and correction are performed on the power change amplitude between adjacent prediction times, specifically including: Check whether the power variation between adjacent prediction points exceeds a reasonable range; Check for any abnormal jumps within a local time period that do not match the cloud cover propagation status; Correct spikes or depressions caused by image anomalies, misjudgments of state, or local calculation errors; While preserving the characteristics of power surges and drops caused by the rapid propagation of real cloud shadows, non-physical abnormal fluctuations are smoothed out.

[0085] Based on this, the final power prediction results can reflect the real power output fluctuations caused by rapid changes in cloud clusters in ultra-short-term scenarios, and avoid abnormal oscillations that do not conform to the actual operating characteristics of photovoltaic power plants.

[0086] In some exemplary embodiments, the following auxiliary information is provided when generating the ultra-short-term photovoltaic power prediction output of the target photovoltaic power plant: Periods with rapid changes in occlusion in the future are marked as high volatility risk periods; Reduce prediction confidence for periods with poor image quality and high uncertainty in state inference; Improve prediction confidence for stable scenarios with consistent support from continuous multi-source information; The prediction results are supplemented with cloud cover intensity, coverage ratio and prediction confidence labels.

[0087] In some exemplary embodiments, after physical boundary constraint correction, temporal continuity verification correction, and provision of auxiliary information, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power plant is generated. As an example, to provide supplementary information, if the absolute value of the rate of change of cloud occlusion intensity characteristics between adjacent moments in the future period is not less than 0.2 / minute, then it is determined that there is rapid occlusion change between adjacent moments, and the adjacent moments are marked as high fluctuation risk periods; if the image clarity of the ground cloud map is lower than the preset image threshold, it is determined that the image quality is poor; if the cloud cluster matching success rate of three consecutive frames in the ground cloud map is less than 70%, then it is determined that the uncertainty of state inference is high, and the prediction confidence is reduced by 5%; if the change trend of multi-source data at three consecutive prediction moments is consistent, it is determined to be a stable scene, and the prediction confidence is increased by 5%.

[0088] The beneficial effects of this disclosure are: This disclosure uses cloud obstruction propagation process as the core driving factor for prediction, and reveals the causal relationship between cloud obstruction and photovoltaic power mutation from a physical mechanism perspective, thereby improving the accuracy and response speed of ultra-short-term photovoltaic power prediction.

[0089] This disclosure enhances the adaptability to typical ultra-short-term scenarios such as rapid local cloud passage, rapid boundary advancement, and sudden shading of the solar neighborhood by extracting cloud boundary morphology and motion propagation characteristics; and effectively characterizes the rapid rise, rapid fall, and short-term violent oscillation process of photovoltaic power by predicting the cloud shading state in future periods.

[0090] This disclosure quantifies the propagation path and order of action of cloud shadows within a photovoltaic power station by introducing spatial structure data of the power station and combining it with the cloud shadow projection coverage status, so that the prediction results can reflect the spatial propagation characteristics of cloud shadows within the power station from a mechanistic perspective.

[0091] This disclosure establishes a mapping mechanism between cloud obstruction propagation, effective irradiance state, and power response, forming a clear intermediate physical transmission chain. This enhances the interpretability of the prediction process and facilitates independent optimization and expansion of different stages.

[0092] This disclosure improves the stability and robustness of photovoltaic power prediction by integrating multi-source data such as ground-based cloud images, historical photovoltaic power, irradiance meteorological data, solar position, and power plant structure, and supports the prediction process from multiple dimensions including image observation, operating status, environmental conditions, and spatial topology.

[0093] This disclosure introduces physical boundary constraints, temporal continuity checks, and provides auxiliary information before the results are output, so that the final prediction results can not only reflect the real fluctuation characteristics caused by cloud shading propagation, but also meet the physical laws and engineering application requirements of actual operation of photovoltaic power plants.

[0094] To implement the exemplary embodiments described above, this disclosure also proposes a photovoltaic power prediction device.

[0095] Figure 2 This is a schematic diagram of a photovoltaic power prediction device provided as an exemplary embodiment of the present disclosure. The device 200 can be implemented by software and / or hardware and is generally integrated into an electronic device. For example... Figure 2 As shown, the device 200 includes: an acquisition module 201, a construction module 202, a first generation module 203, a obtaining module 204, and a second generation module 205, wherein, The acquisition module 201 is configured to acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data and power station structure data of the area where the target photovoltaic power station is located; The construction module 202 is configured to construct a cloud shading state vector based on the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data; The first generation module 203 is configured to predict the evolution of cloud occlusion state in future periods based on cloud occlusion state vectors in historical periods, and generate a cloud occlusion propagation state sequence for the future periods. Module 204 is configured to determine the effective irradiance state and power response for the future period based on the historical photovoltaic power data, the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, the astronomical location data, and the power station structure data, and obtain the future power response sequence. The second generation module 205 is configured to perform physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output result of the target photovoltaic power station.

[0096] The apparatus provided in the exemplary embodiments of this disclosure includes historical photovoltaic power data, which includes station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power plant. The ground-based cloud image data includes a continuous sequence of sky images; The irradiance meteorological data shall include at least total irradiance, diffuse irradiance, direct irradiance, ambient temperature, component temperature, relative humidity, wind speed, and wind direction; The astronomical location data includes at least the solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance values ​​for the predicted time and future periods. The power station structure data includes the location coordinates of the photovoltaic array, the arrangement of the components, the tilt angle and orientation of the components, the row and column relationship of the array, the mapping relationship between the inverter and the array, and the spatial information of the station area; The acquisition module 201 is specifically used for: Based on five types of data channels, the historical photovoltaic power data, the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data are acquired; Based on the preset standardized data preprocessing, the historical photovoltaic power data, the ground cloud map data, the irradiance meteorological data, the astronomical location data and the power station structure data are preprocessed. The standardized data preprocessing includes data format unification, time alignment, spatial alignment and data quality labeling.

[0097] The apparatus provided in the exemplary embodiments of this disclosure, wherein the construction module 202 is specifically used for: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, cloud obstruction intensity characteristics, cloud boundary morphology characteristics, cloud movement and propagation characteristics, and cloud shadow projection coverage characteristics are extracted. Based on the cloud occlusion intensity characteristics, cloud boundary morphology characteristics, cloud motion propagation characteristics, and cloud shadow projection coverage characteristics, a cloud occlusion state vector is constructed.

[0098] The apparatus provided in the exemplary embodiments of this disclosure includes a first generation module 203 specifically configured to: Construct a historical state sequence based on the cloud occlusion state vectors of historical time periods; Based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed to determine the propagation trend in future periods; Based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is extrapolated step by step to obtain the future cloud occlusion state vector at the predicted time. Based on the solar position parameters for future time periods, the cloud shadow projection coverage features in the future cloud occlusion state vector are dynamically corrected to generate the cloud occlusion propagation state sequence for the future time periods.

[0099] The apparatus provided in the exemplary embodiments of this disclosure, wherein the obtaining module 204 is specifically used for: Based on the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, and the astronomical location data for the future period, the effective irradiance state for the future period is calculated. Based on the power station structure data, the spatial sub-regions of the photovoltaic power station are determined, and based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of the spatial sub-regions in the future time period is determined. Based on the effective irradiance state, the irradiated state, and the historical photovoltaic power data, the effective irradiance state and power response for the future time period are determined through comprehensive mapping or partition aggregation, thereby obtaining the future power response sequence for the future time period.

[0100] The apparatus provided in the exemplary embodiments of this disclosure includes physical boundary constraints such as nonnegativity constraints, rated capacity constraints, solar position constraints, and irradiance consistency constraints. The second generation module 205 is specifically used for: Based on the physical boundary constraints, the future power response sequence is modified according to the physical boundary constraints. Based on the preset power change range, the power change amplitude between adjacent prediction times is checked and corrected for temporal continuity. Based on the future power response sequence, a time-series verification and correction are performed on abnormal jumps that do not match the cloud obstruction propagation state sequence. Based on the physical boundary constraint correction and the temporal continuity verification correction, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station is generated.

[0101] The photovoltaic power prediction device provided in the exemplary embodiments of this disclosure can execute the photovoltaic power prediction method provided in any exemplary embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the method.

[0102] To implement the exemplary embodiments described above, this disclosure also proposes a computer program product, including a computer program / instructions that, when executed by a processor, implement the photovoltaic power prediction method in the exemplary embodiments described above.

[0103] Figure 3 A schematic diagram of the structure of an electronic device provided as an exemplary embodiment of the present disclosure.

[0104] The following is a detailed reference. Figure 3 The diagram illustrates a structural schematic suitable for implementing the electronic device 300 in the exemplary embodiments of this disclosure. The electronic device 300 in the exemplary embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the exemplary embodiments disclosed herein.

[0105] like Figure 3 As shown, the electronic device 300 may include a processor (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a memory 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processor 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0106] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0107] In particular, according to exemplary embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, exemplary embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such exemplary embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from memory 308, or installed from ROM 302. When the computer program is executed by processor 301, it performs the functions defined in the photovoltaic power prediction method of exemplary embodiments of the present disclosure.

[0108] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0109] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0110] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0111] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the aforementioned photovoltaic power prediction method.

[0112] Electronic devices can be programmed with computer program code in one or more programming languages ​​or combinations thereof to perform the operations of this disclosure. These programming languages ​​include, but are not limited to, object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0113] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various exemplary embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0114] The units described in the exemplary embodiments of this disclosure can be implemented in software or in hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0115] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0116] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0117] The above description is merely a preferred exemplary embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0118] Furthermore, although the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. Multitasking and parallel processing may be advantageous in certain environments. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual exemplary embodiments may also be implemented in combination in a single exemplary embodiment. Conversely, various features described in the context of a single exemplary embodiment may also be implemented individually or in any suitable sub-combination in multiple exemplary embodiments.

[0119] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A photovoltaic power prediction method, characterized in that, include: Acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data, and power station structure data of the area where the target photovoltaic power station is located; Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, feature extraction is performed to construct a cloud obscuring state vector; Based on the cloud occlusion state vector of historical time periods, the evolution process of cloud occlusion state in future time periods is predicted, and a cloud occlusion propagation state sequence for the future time periods is generated. Based on the historical photovoltaic power data, the cloud obstruction propagation state sequence, the future period's irradiance meteorological data, astronomical location data, and power plant structure data, the effective irradiance state and power response for the future period are determined, and a future power response sequence is obtained. The future power response sequence is corrected by physical boundary constraints and temporal continuity verification to generate the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station.

2. The method according to claim 1, characterized in that, The historical photovoltaic power data includes the station-level active power, array-level power, inverter-level power, and string-level power data of the target photovoltaic power station. The ground-based cloud image data includes a continuous sequence of sky images; The irradiance meteorological data shall include at least total irradiance, diffuse irradiance, direct irradiance, ambient temperature, component temperature, relative humidity, wind speed, and wind direction; The astronomical location data includes at least the solar altitude angle, solar azimuth angle, sunrise and sunset times, and theoretical clear-sky irradiance values ​​for the predicted time and future periods. The power station structure data includes the location coordinates of the photovoltaic array, the arrangement of the components, the tilt angle and orientation of the components, the row and column relationship of the array, the mapping relationship between the inverter and the array, and the spatial information of the station area; The acquisition and preprocessing of historical photovoltaic power data, ground-based cloud map data, irradiance meteorological data, astronomical location data, and power station structure data for the target photovoltaic power station area includes: Based on five types of data channels, the historical photovoltaic power data, the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data are acquired; Based on the preset standardized data preprocessing, the historical photovoltaic power data, the ground cloud map data, the irradiance meteorological data, the astronomical location data and the power station structure data are preprocessed. The standardized data preprocessing includes data format unification, time alignment, spatial alignment and data quality labeling.

3. The method according to claim 1, characterized in that, The cloud obscuration state vector is constructed by extracting features from the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data, including: Based on the ground-based cloud map data, the irradiance meteorological data, the astronomical location data, and the power station structure data, cloud obstruction intensity characteristics, cloud boundary morphology characteristics, cloud movement and propagation characteristics, and cloud shadow projection coverage characteristics are extracted. Based on the cloud occlusion intensity characteristics, cloud boundary morphology characteristics, cloud motion propagation characteristics, and cloud shadow projection coverage characteristics, a cloud occlusion state vector is constructed.

4. The method according to claim 1, characterized in that, The cloud occlusion state vector based on historical time periods is used to predict the evolution of cloud occlusion state in future time periods, generating a cloud occlusion propagation state sequence for the future time periods, including: Construct a historical state sequence based on the cloud occlusion state vectors of historical time periods; Based on the historical state sequence, the propagation trend of the current dominant cloud cluster or the main obstruction area is analyzed to determine the propagation trend in future periods; Based on the propagation trend, the cloud occlusion state vector at each predicted time in the future period is extrapolated step by step to obtain the future cloud occlusion state vector at the predicted time. Based on the solar position parameters for future time periods, the cloud shadow projection coverage features in the future cloud occlusion state vector are dynamically corrected to generate the cloud occlusion propagation state sequence for the future time periods.

5. The method according to claim 1, characterized in that, Based on the historical photovoltaic power data, the cloud cover propagation state sequence, the future period's irradiance meteorological data, astronomical location data, and power plant structure data, the effective irradiance state and power response for the future period are determined, resulting in a future power response sequence, including: Based on the cloud obstruction propagation state sequence, the irradiance meteorological data for the future period, and the astronomical location data for the future period, the effective irradiance state for the future period is calculated. Based on the power station structure data, the spatial sub-regions of the photovoltaic power station are determined, and based on the cloud shadow projection coverage characteristics in the cloud shading propagation state sequence, the illumination status of the spatial sub-regions in the future time period is determined. Based on the effective irradiance state, the irradiated state, and the historical photovoltaic power data, the effective irradiance state and power response for the future time period are determined through comprehensive mapping or partition aggregation, thereby obtaining the future power response sequence for the future time period.

6. The method according to claim 1, characterized in that, The physical boundary constraints include nonnegativity constraints, rated capacity constraints, solar position constraints, and irradiance consistency constraints. The process of performing physical boundary constraint correction and temporal continuity verification on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station includes: Based on the physical boundary constraints, the future power response sequence is modified according to the physical boundary constraints. Based on the preset power change range, the power change amplitude between adjacent prediction times is checked and corrected for temporal continuity. Based on the future power response sequence, a time-series verification and correction are performed on abnormal jumps that do not match the cloud obstruction propagation state sequence. Based on the physical boundary constraint correction and the temporal continuity verification correction, the ultra-short-term photovoltaic power prediction output of the target photovoltaic power station is generated.

7. A photovoltaic power prediction device, the device comprising: The acquisition module is configured to acquire and preprocess historical photovoltaic power data, ground cloud map data, irradiance meteorological data, astronomical location data, and power station structure data of the area where the target photovoltaic power station is located; The construction module is configured to extract features based on the ground-based cloud image data, the irradiance meteorological data, the astronomical location data, and the power station structure data to construct a cloud obscuring state vector; The first generation module is configured to predict the evolution of cloud occlusion state in future periods based on cloud occlusion state vectors in historical periods, and generate a cloud occlusion propagation state sequence for the future periods. The module is configured to determine the effective irradiance state and power response for the future period based on the historical photovoltaic power data, the cloud shading propagation state sequence, the irradiance meteorological data for the future period, the astronomical location data, and the power station structure data, and obtain the future power response sequence. The second generation module is configured to perform physical boundary constraint correction and temporal continuity verification correction on the future power response sequence to generate the ultra-short-term photovoltaic power prediction output result of the target photovoltaic power station.

8. An electronic device, characterized in that, include: Memory; processor; as well as Computer programs; The computer program is stored in the memory and configured to be executed by the processor to implement the steps of the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, It stores a computer program / instruction thereon, which, when executed by a processor, implements the steps of the method described in any one of claims 1-6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-6.

Images