PV Power Forecasting Using Sky Images and Weather Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Predicting photovoltaic power generation accurately is challenging due to environmental factors like weather and climate, which significantly impact solar radiation levels.
Innovation Solution
A device and method using deep neural networks to analyze sky images and meteorological data to predict solar radiation and photovoltaic power generation amounts by training models with historical data, incorporating power plant specifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional meteorological data alone is used for prediction, then the prediction method is simple, but the prediction accuracy is insufficient due to environmental factors affecting solar radiation
Solution Approach 1:
The patent combines sky image data with meteorological data to create a comprehensive prediction system. The sky image analysis module processes visual information about cloud cover and sky conditions, while the meteorological data module handles numerical weather parameters. These two data streams are merged in the prediction model to achieve more accurate solar radiation forecasts than either data source could provide alone.
Solution Approach 2:
The patent introduces an intermediate processing layer that converts sky images into quantitative cloud cover indices and sky condition parameters. This intermediary transformation allows visual sky information to be integrated with numerical meteorological data in a unified prediction framework, bridging the gap between qualitative visual assessment and quantitative meteorological measurement.
2Measurement precision
If sky images are incorporated into the prediction system, then prediction accuracy improves, but data processing complexity increases
Solution Approach 1:
The patent segments the sky image analysis into distinct functional modules: cloud detection module, cloud cover quantification module, and sky condition classification module. This segmentation allows each aspect of sky image processing to be handled independently with specialized algorithms, reducing the overall complexity compared to attempting to analyze all sky image features simultaneously.
Solution Approach 2:
The patent uses image processing techniques to create simplified representations (copies) of the actual sky images. These processed images contain extracted features such as cloud cover percentages, sky brightness distributions, and cloud type classifications, which are then used in the prediction model instead of processing the full-resolution original images, reducing computational complexity.
3Reliability
If historical data is used for training models, then prediction reliability improves, but data storage and processing requirements increase
Solution Approach 1:
The patent performs preliminary processing of historical sky images and meteorological data during the training phase. Historical data is pre-labeled with corresponding solar radiation measurements, and sky images are pre-processed to extract relevant features. This preliminary action allows the model to be trained efficiently on compact feature representations rather than raw data, reducing storage requirements while maintaining prediction reliability.
Solution Approach 2:
The patent transforms historical sky images into parameterized representations such as cloud cover indices, sky brightness values, and cloud morphology parameters. This parameterization converts large-volume image data into compact numerical parameters that capture the essential information needed for prediction, significantly reducing data storage requirements while preserving the information necessary for reliable predictions.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy of predicting photovoltaic power generation by leveraging AI algorithms to analyze sky images and meteorological data, improving power generation planning and operational efficiency.
Implementation Method 1
Photovoltaic power generation converts light from the sun into electrical energy using the photovoltaic effect
Data Source
AI summary
A method of predicting a photovoltaic power generation amount according to an embodiment may include predicting a sky image of a next time by analyzing sky images captured from a past time to a present time and meteorological data from the past time to the present time, predicting a solar radiation amount of the next time by analyzing the sky image of the next time, a clear-sky solar radiation amount of the next time, and meteorological data of the next time, which are generated as a result of the prediction, and predicting a photovoltaic power generation amount of the next time by analyzing the solar radiation amount of the next time and the meteorological data of the next time, which are generated as a result of the prediction, and power plant specification data.