Solar Irradiation Forecasting via Community Sky Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current solar prediction technologies face challenges in accurately predicting short-term and localized solar irradiation due to perturbations from clouds, dust, and aerosols, particularly in scenarios where high-resolution sky images are limited to small spatial distributions and short time horizons, making it difficult to provide reliable solar forecasting for individual consumers and off-grid users.
Innovation Solution
A method and system that utilize a community network of sky imaging devices to collect and share data, leveraging machine learning models to predict cloud positions and solar irradiation based on cloud shape and position, overcoming the circumsolar region challenge by combining images from multiple angles and locations, and enabling predictions up to 11 hours in advance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If satellite images are used to predict solar energy over a large area over long time horizons, then the spatial coverage and time horizon are improved, but the resolution and accuracy for short-term and specific location predictions deteriorate
Solution Approach 1:
The patent segments the prediction task into multiple components: using satellite images for long-term forecasting while using local terrestrial sky images for short-term high-accuracy predictions. This segmentation allows each method to optimize for its specific temporal scale, resolving the contradiction between long time horizon and high precision.
Solution Approach 2:
The patent applies local quality by using terrestrial sky imaging devices deployed at specific locations to capture local cloud conditions and solar irradiation data. This local measurement approach provides high-resolution, location-specific predictions for short-term horizons, while satellite data provides the broader context for long-term forecasting.
2Measurement precision
If high-resolution sky images are acquired for short time and small spatial distribution, then the measurement precision for localized predictions is improved, but the spatial coverage and time horizon are limited
Solution Approach 1:
The patent merges multiple data sources: satellite images providing broad spatial coverage with local terrestrial sky images providing high-resolution localized measurements. By combining these complementary datasets, the system achieves both wide spatial coverage and high measurement precision for specific locations.
Solution Approach 2:
The patent creates a multi-functional system where terrestrial sky imaging devices serve multiple purposes: capturing local cloud conditions, measuring solar irradiation, and providing data for both short-term high-precision predictions and contributing to broader spatial understanding through the networked approach.
3Reliability
If terrestrial sky imaging devices are deployed at multiple geographic locations, then the spatial coverage and prediction reliability are improved, but the device complexity and data processing requirements increase
Solution Approach 1:
The patent introduces a centralized processing system as an intermediary that receives data from multiple terrestrial imaging devices, performs cloud-based processing, and distributes predictions back to users. This intermediary layer simplifies the overall system architecture by centralizing complex computations while maintaining distributed data collection.
Solution Approach 2:
The system implements feedback mechanisms where predictions from the networked devices are continuously refined using new data from all locations. The centralized processing system analyzes data from multiple sources and provides updated predictions that feed back into the system, improving reliability while managing complexity through iterative refinement.
Data Source
AI summary
Solar irradiation may be predicted based on input terrestrial sky images comprising cloud images, the terrestrial sky images taken from a plurality of geographic locations by a plurality of devices; for example, wherein the terrestrial sky images are crowd sourced from the plurality of devices. A model may be generated that predicts solar irradiation in a geographic area based on the input terrestrial sky images and the geographic locations from where the terrestrial sky images were taken. A signal representing the solar irradiation predicted by the model is output.