Solar Energy Disaggregation Using Irradiation and Net Power Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Distribution system operators (DSOs) face challenges in accurately estimating solar generation and consumption due to lack of separate measurements and unreliable knowledge of installed PV capacity, limiting efficient grid planning and decision-making.
Innovation Solution
A physics-based model is trained using net power and solar irradiation data to estimate aggregate solar generation without prior separate measurements or knowledge of installed PV capacity, employing a regression model that considers temperature dependency and linear relationships with irradiation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If supervised learning methods with labeled data are used for solar energy disaggregation, then measurement precision can be improved, but device complexity and cost increase due to requiring separate measurement equipment
Solution Approach 1:
The patent extracts and removes the need for complex separate measurement equipment by using only existing net power measurements and public irradiation data. The disaggregation problem is solved by taking out the requirement for specialized sensors and using a simplified approach with readily available data sources.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary that processes net power measurements and irradiation data to estimate solar generation. This intermediary computational layer replaces the need for direct physical measurement of solar generation, bridging the gap between available measurements and desired information.
2Ease of operation
If assumed knowledge of installed PV capacity is used for solar generation estimation, then ease of operation is improved, but measurement precision deteriorates due to inaccurate or incomplete reporting
Solution Approach 1:
The patent uses feedback mechanisms where the model learns from historical net power measurements and adjusts its estimates accordingly. The system continuously refines its predictions by comparing model outputs with actual measurements, improving precision while maintaining ease of operation without requiring manual PV capacity input.
Solution Approach 2:
The system performs self-service by automatically estimating solar generation using only net power measurements and public irradiation data. It does not require manual input of PV capacity information or external assistance, making it both easy to operate and precise by leveraging its own learning capabilities.
3Measurement precision
If supervised learning with ground truth data is used, then measurement precision improves, but loss of time and resources increase due to expensive and unrealistic data acquisition
Solution Approach 1:
The patent performs preliminary action by training the model offline using historical data, so that during operation it can quickly make predictions without requiring real-time ground truth data acquisition. The heavy computational work is done in advance, enabling fast deployment and operation.
Solution Approach 2:
The patent replaces expensive, hard-to-acquire ground truth data with cheap, readily available public irradiation data and existing net power measurements. This substitution uses inexpensive data sources that are easily accessible and can be obtained quickly without special equipment or permissions.
Data Source
AI summary
A computer-implemented method for solar energy disaggregation at a node of a distribution system includes obtaining first data including samples of averaged net power measured by a meter installed at the node and second data including samples of solar irradiation local to the node averaged over intervals temporally correlated with the first data. A physics-based model of the node is trained using the first and second data, to optimize a set of model parameters. The physics-based model is defined such that an average net power includes a composite of an average solar generation and an average power consumption at the node and the average solar generation is modeled as a function of an average solar irradiation. An aggregate solar generation at the node is estimated/predicted from actual irradiation data/irradiation forecast data local to the node using a model parameter optimized by the training.