Desarrollo de un software simulador basado en modelos matemáticos y machine learning para analizar y optimizar redes híbridas de energía renovable en centros educativos rurales del cantón Riobamba.

This research aimed to develop simulation software based on mathematical models and machine learning techniques to analyze and optimize hybrid renewable energy networks in rural schools in the canton Riobamba. The identified problem relates to the limited access to reliable energy in rural areas and the lack of accessible tools for planning self-sufficient energy systems. The simulator integrates differential equations and numerical methods to model energy generation, storage, and consumption; it also incorporates prediction algorithms such as Random Forest, Gradient Boosting, XGBoost, and neural networks to estimate solar/wind production and electricity consumption. An optimization module using linear programming and genetic algorithms was also developed, aimed at minimizing costs and maximizing efficiency. The results showed that the modeled solar-wind hybrid grids achieved energy self-sufficiency between 68% and 82%, with savings of up to 40% compared to traditional electricity consumption. The Random Forest model achieved the best predictive performance, with an R² of 0.936, demonstrating robustness to noisy climate data in high-altitude environments. The simulator is a useful tool for energy planning in rural areas, contributing to sustainability, cost reduction, and environmental impact mitigation.

• المؤلف الرئيسي: Tenenaula Cunduri, María Rebeca (author)
• التنسيق: masterThesis
• اللغة: spa
• منشور في: 2026
• الموضوعات: Random forest; Matemática computacional; Machine learning; Eficiencia energética
• الوصول للمادة أونلاين: http://dspace.unach.edu.ec/handle/51000/16460