Inferring Photovoltaic System Specs via Net Load Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurate forecasting of photovoltaic power generation is hindered by inaccuracies in photovoltaic system configuration specifications and the high costs and complexities of collecting high-speed time series power production data from dispersed systems, especially under cloudy conditions, leading to unpredictable fleet performance.
Innovation Solution
A computer-implemented system infers operational specifications of photovoltaic power generation systems using net load data and measured solar resource data, simulating power output for various hypothetical configurations to minimize total squared error, enabling accurate forecasting and maintenance assessment without requiring complete configuration data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-speed time series power production data is collected from dispersed photovoltaic systems, then forecasting accuracy is improved, but data collection costs and system complexity increase significantly
Solution Approach 1:
The patent creates a virtual model (digital twin) of the photovoltaic system that replicates its operational behavior. Instead of collecting extensive real-world data, the virtual model is trained on limited historical data and then used to simulate and predict system performance under various conditions, effectively copying the system's characteristics into a computationally tractable form.
Solution Approach 2:
The patent introduces a virtual model as an intermediary between the physical photovoltaic system and the forecasting process. This virtual model acts as a mediator that translates limited measured data into comprehensive performance predictions, eliminating the need for direct complex data collection infrastructure while maintaining forecasting accuracy.
2Measurement precision
If complete photovoltaic system configuration specifications are obtained, then power generation forecasting accuracy is improved, but data collection costs and time increase
Solution Approach 1:
The virtual model creates a digital representation of the photovoltaic system configuration, capturing essential operational characteristics without requiring complete physical inspection or data collection. The model learns system parameters from limited operational data, effectively copying the relevant configuration information needed for accurate forecasting.
Solution Approach 2:
The patent performs preliminary training of the virtual model using available historical data before actual forecasting operations. This preliminary action allows the model to learn system characteristics in advance, so that during operational phases, accurate forecasting can be achieved without time-consuming data collection processes.
3Measurement precision
If extensive data collection infrastructure is deployed, then operational specifications accuracy is improved, but system complexity and cost increase
Solution Approach 1:
The patent replaces physical data collection infrastructure with a virtual model that replicates system behavior. The virtual model captures and processes operational specifications through computational means rather than physical sensors and data collection networks, significantly reducing infrastructure complexity while maintaining accuracy.
Solution Approach 2:
The patent substitutes mechanical and physical data collection systems with computational and algorithmic approaches. Instead of deploying extensive sensor networks and data collection hardware, the solution uses software-based virtual modeling and machine learning algorithms to achieve the same measurement precision with minimal physical infrastructure.
Data Source
AI summary
A system and method for net load-based inference of operational specifications of a photovoltaic power generation system with the aid of a digital computer are provided. Photovoltaic plant configuration specifications can be accurately inferred with net load data and measured solar resource data. Power generation data is simulated for a range of hypothetical photovoltaic system configurations based on a normalized solar power simulation model. Net load data is estimated based on one or more component loads. The set of key parameters corresponding to the net load estimate that minimizes total squared error represents the inferred specifications of the photovoltaic plant configuration.