1. Demand response of prosumers integrating storage system for optimizing grid-connected photovoltaics through time-pricing
- Author
-
Díaz-Bello, Dácil, Vargas-Salgado, Carlos, Alcázar-Ortega, Manuel, and Gómez-Navarro, Tomás
- Abstract
Complex, non-linear systems with diverse consumption, generation, and storage technologies require advanced energy management by considering several factors: resource availability, energy efficiency, economic viability, consumption costs, and installation feasibility. To address these challenges, this paper integrates artificial intelligence into a real grid-connected household with photovoltaic systems, consumption curves, and a potential storage system applying demand response. The first approach incorporates neural networks for generation prediction and genetic algorithms for demand optimization. The second approach integrates optimization and storage, significantly contributing to the feasibility study of storage implementation post-initial optimization; combining such technologies, the synergy between demand-side optimization and energy storage is evaluated by employing commonly used scientific tools, such as MATLAB for modeling and applying artificial intelligence and HOMER for economic analysis. The methodology and results show these approaches' achievements since few studies comprehensively address household power management, considering aspects to minimize costs and maximize efficiency. Results show minimum nRMSEs in the forecast of 8.80, together with regression values up to 0.99594. The study reports cost reductions of up to 69 % and 100 % in worst-case periods for systems without and with batteries, respectively; additionally, it reports increasing self-consumption of renewables by 76 % and 99 %, respectively.
- Published
- 2024
- Full Text
- View/download PDF