Solar district energy systems with a seasonal energy storage: Advanced data-driven metaheuristic optimization

Identificador:  imarina:9462622

Handle: https://hdl.handle.net/20.500.11797/imarina9462622

Autors:  Kotegov, Ruslan; Abokersh, Mohamed; Mateu, Carles; Shobo, Adedamola; Boer, Dieter; Valles, Manel

Resum:
Optimizing Solar District Energy Systems (SDES) requires balancing economic feasibility, environmental impact, and computational efficiency. These systems integrate renewable technologies such as solar thermal collectors, photovoltaic (PV) panels, domestic hot water tanks, and seasonal thermal energy storage to meet the heating, electricity, and hot water needs of communities. However, designing cost-effective and sustainable configurations remains challenging due to the system's complexity and competing objectives. To address this, we propose a robust optimization framework that couples TRNSYS simulations with a Python-based control structure, enabling adaptive decision-making and an accurate performance assessment. Applied to a real SDES case study in Falset, Spain, the methodology identifies system configurations that balance economic and environmental goals. Compared to the fossil-based baseline, the most sustainable solution achieves a 33 % reduction in environmental impact and a 68 % decrease in cost, while the most economical solution lowers environmental impact by 11 % and cuts cost by 88 %. Several scenarios achieve full economic self-sufficiency, with electricity revenues exceeding operating expenses. Although initial investments increase by a factor of 25-32 due to renewable deployment, the optimization ensures strategic allocation to maximize longterm performance and returns. This hybrid methodology addresses adaptability challenges in energy system design, offering a practical and effective decision-support tool for planners, engineers, and policymakers. It facilitates a comprehensive trade-off analysis between cost and sustainability, unlocking cost-effective pathways for low-carbon urban energy transitions. The proposed methodology improves upon conventional optimization approaches by maintaining simulation accuracy, reducing data requirements, and enhancing adaptability to system changes.