University of North Florida. School of Engineering
Multi-objective optimization strategy for component sizing in solar-hydrogen microgrids using an advanced hybrid genetic algorithm
Abstract
dc:description.abstractThis thesis presents the development of a genetic algorithm (GA) optimization framework for the design and component sizing of hybrid solar-hydrogen microgrids. The framework addresses a critical gap in research and existing commercial tools by unifying performance maximization and cost minimization objectives across both grid-tied and islanded configurations. Integrating solar photovoltaics, electrolyzers, hydrogen storage, fuel cells, and batteries, the GA employs adaptive weighting and dynamic boundary constraints to balance technical feasibility with economic efficiency. To ensure real-world viability, the algorithm relies on a novel Daylight Sun Factor (DSF) for localized solar assessment and was rigorously validated against multi-year, high-fidelity irradiance datasets. Furthermore, the computational sizing logic was empirically verified using the PicoGrid, a physical pilot-scale microgrid at the JEA Sustainable Solutions Lab. Physical testing confirmed the algorithm's ability to engineer systems capable of maintaining sustained grid stability during severe low-insolation weather events. Comparative benchmarking against HOMER Pro demonstrated that the GA achieves comparable or superior system autonomy with more efficient, balanced energy storage sizing, successfully preventing the capital inflation often caused by component oversizing. Ultimately, this research delivers a robust, physics-informed computational tool that advances the autonomous design of resilient renewable energy systems.
Degree
thesis:*- Name thesis:degree_name
- Master of Science in Mechanical Engineering (MSME)
- Discipline thesis:degree_discipline
- Engineering
- Grantor
- University of North Florida. School of Engineering
- Date dc:date.issued
- Spring
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Jones, Dylan
- Jones, Dylan, 1998-
- Advisors dc:contributor.advisor
-
- Fletcher, James
- Bevill, Grant
- Oshman, Christopher
- Contributors dc:contributor
-
- Dr. James Fletcher
- Dr. Grant Bevill
- Dr. Christopher Oshman
Subjects
dc:subject × 13- Thesis
- University of North Florida
- UNF
- Dissertations, Academic -- UNF -- Master of Science in Mechanical Engineering
- Dissertations, Academic -- UNF -- Engineering
- Optimization
- Genetic Algorithm
- Microgrid
- Electrical and Electronics
- Energy Systems
- Other Mechanical Engineering
- Power and Energy
- Microgrids (Smart power grids) -- Technological innovations; Renewable energy sources -- Mathematical models -- Testing; Hydrogen as fuel -- Testing; Solar energy -- Industrial applications -- Testing; Genetic algorithms -- Industrial applications; Mathematical optimization -- Testing; Energy storage -- Management -- Technological innovations
Rights
dc:rightsIdentifiers
dc:identifier.*- Repository record dc:identifier
- https://digitalcommons.unf.edu/etd/1402