PV Configuration Inference From Production Data for Fleet Forecasting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing photovoltaic fleet forecasting methods face inaccuracies due to incomplete or unavailable system specifications, computational complexity, and high costs associated with direct data collection, leading to unpredictable power output and inefficiencies in grid operations.
Innovation Solution
A method to infer photovoltaic system configuration specifications using historical measured production data and solar resource data, employing a statistical approach to simulate a range of configurations and identify the optimal system configuration through comparison with actual data, allowing for high-speed fleet production data generation and correction of missing specifications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If brute force approach is used to test all possible combinations of system specification parameters, then complete search space coverage is achieved, but computational time and resources become prohibitive
Solution Approach 1:
The patent segments the combinatorial search space by identifying and fixing parameters that have minimal impact on production output (such as inverter size, loss factor, and other parameters) separate from the critical geometric parameters (azimuth, tilt angle, obstruction elevation angles). This segmentation allows the algorithm to focus computational effort on the most influential parameters while treating others as secondary, dramatically reducing the effective search space from 12 quadrillion combinations to a manageable subset that still achieves 1° specification accuracy.
Solution Approach 2:
The patent applies preliminary action by using measured photovoltaic production data to constrain and narrow the search space before performing the full combinatorial search. By comparing simulated production data against actual measured data, the algorithm pre-identifies plausible parameter ranges and eliminates impossible combinations in advance, reducing the computational burden before the main optimization routine begins.
2Ease of manufacture
If user-supplied system specifications are accepted, then simplicity is maintained, but accuracy of production forecasting deteriorates
Solution Approach 1:
The patent implements feedback by using measured photovoltaic production data to validate and refine the system specification parameters. The algorithm simulates production output using candidate parameters, compares these simulations against actual measured production data, and iteratively adjusts parameters to minimize the difference between simulated and measured output. This feedback loop ensures that the final inferred specifications accurately reflect the real system while maintaining the simplicity of using readily available production data as the input basis.
Solution Approach 2:
The system performs self-service by automatically inferring its own configuration parameters from its operational production data. Rather than requiring external specialists to manually survey and measure physical system characteristics, the algorithm uses the system's own measured production output to self-determine its azimuth, tilt angle, and obstruction parameters, eliminating the need for complex external data collection while achieving high accuracy.
3Loss of information
If measured photovoltaic production data is used to infer system specifications, then data availability is improved, but computational complexity increases
Solution Approach 1:
The patent applies parameter changes by transforming the problem from directly solving for physical dimensions to solving for production parameters that can be inferred from operational data. By changing the parameter space from physical measurements (panel dimensions, exact component locations) to production-relevant parameters (azimuth, tilt angle, obstruction elevation angles), the algorithm makes the problem tractable using standard optimization techniques rather than requiring complex inverse modeling.
4Productivity
If high-speed fleet production data generation is implemented, then productivity is improved, but support for multiple time resolutions and fleet configurations increases system complexity
Solution Approach 1:
The patent achieves universality by designing an inference algorithm that can handle multiple time resolutions and various fleet configurations through a single unified framework. The same core optimization routine works whether analyzing hourly, daily, or monthly production data, and whether inferring parameters for a single system or a fleet of systems with different configurations. This multi-functional approach maintains high productivity while avoiding the need for separate specialized algorithms for each scenario.
Data Source
AI summary
A photovoltaic system's configuration specification can be inferred by an evaluative process that searches through a space of candidate values for the variables in the specification. Each variable is selected in a specific ordering that narrows the field of candidate values. A constant horizon is assumed to account for diffuse irradiance insensitive to specific obstruction locations relative to the photovoltaic system's geographic location. Initial values for the azimuth angle, constant horizon obstruction elevation angle, and tilt angle are determined, followed by final values for these variables. The effects of direct obstructions that block direct irradiance in the areas where the actual horizon and the range of sun path values overlap relative to the geographic location are evaluated to find the exact obstruction elevation angle over a range of azimuth bins or directions. The photovoltaic temperature response coefficient and the inverter rating or power curve of the photovoltaic system are determined.