Thesis Desarrollo y evaluación de un modelo de optimización determinista no lineal para la planificación y diseño de plantas híbridas de hidrógeno verde
Loading...
Date
2025-12-05
Authors
Journal Title
Journal ISSN
Volume Title
Program
Ingeniería Civil Eléctrica
Departament
Campus
Campus Casa Central Valparaíso
Abstract
La creciente penetración de energías renovables en Chile, junto con el interés estratégico en el hidrógeno verde, impulsa la necesidad de desarrollar modelos de diseño capaces de capturar tanto la complejidad operativa como las restricciones fisicoquímicas reales de los sistemas de producción. En el presente informe se evalúa el diseño y dimensionamiento de una planta híbrida de generación y almacenamiento para la producción de hidrógeno verde, utilizando dos modelos de optimización determinista: uno lineal y otro no lineal, con el objetivo de maximizar la producción total de hidrógeno verde y, en consecuencia, de amoníaco verde. El modelo lineal ofrece una representación simplificada con eficiencia constante, mientras que el modelo no lineal incorpora formulaciones basadas en variaciones reales de eficiencia, lo que permite capturar con mayor precisión el comportamiento y la operación del sistema. La información utilizada para el cálculo de resultados fue proporcionada por la empresa MAKO Inversiones SpA. Los resultados muestran diferencias significativas entre ambos enfoques. El modelo lineal tiende a sobreestimar la eficiencia del electrolizador, el desempeño técnico general y los indicadores económicos, proyectando una mayor producción anual de hidrógeno y amoníaco, así como menores costos energéticos. En contraste, el modelo no lineal reduce la potencia óptima de los equipos, incorpora pérdidas energéticas y evidencia un mayor consumo eléctrico, lo que se traduce en menores ingresos, una disminución del VAN en un 26,2 % y de la TIR en un 10,6 %, junto con un aumento del LCOA en un 8,6 %. Estos resultados reflejan que la inclusión de relaciones no lineales y pérdidas internas permite capturar de manera más precisa las ineficiencias presentes en la operación real del sistema(...).
The increasing penetration of renewable energy sources in Chile, besides the country’s strategic interest in the development of green hydrogen, has increased the need for design models capable of representing the operational complexity and the physicochemical restrictions to production systems. This report evaluates the design and sizing of a hybrid generation and storage plant intended for green hydrogen production, using two deterministic optimization models: one linear and one nonlinear. The objective is to maximize total green hydrogen production and, consequently, green ammonia output. While the linear model relies on a simplified representation with constant efficiencies, the nonlinear model incorporates formulations that reflect real efficiency variations, enabling a more accurate characterization of system behavior and operational performance. The data used for the analysis were provided by MAKO INVERSIONES SpA. The results reveal significant differences between the two approaches. The linear model tends to overestimate the electrolyzer efficiency, overall technical performance, and economic indicators, projecting higher annual production of hydrogen and ammonia, along with lower energy costs. In contrast, the nonlinear model reduces the optimal equipment capacity, incorporates internal energy losses, and reflects higher electricity consumption. This leads to lower revenues, a 26.2% decrease in NPV, a 10.6% reduction in IRR, and an 8.6% increase in LCOA. These outcomes demonstrate that including nonlinear relationships and internal losses captures, with greater accuracy, the inefficiencies present in real system operation(...).
The increasing penetration of renewable energy sources in Chile, besides the country’s strategic interest in the development of green hydrogen, has increased the need for design models capable of representing the operational complexity and the physicochemical restrictions to production systems. This report evaluates the design and sizing of a hybrid generation and storage plant intended for green hydrogen production, using two deterministic optimization models: one linear and one nonlinear. The objective is to maximize total green hydrogen production and, consequently, green ammonia output. While the linear model relies on a simplified representation with constant efficiencies, the nonlinear model incorporates formulations that reflect real efficiency variations, enabling a more accurate characterization of system behavior and operational performance. The data used for the analysis were provided by MAKO INVERSIONES SpA. The results reveal significant differences between the two approaches. The linear model tends to overestimate the electrolyzer efficiency, overall technical performance, and economic indicators, projecting higher annual production of hydrogen and ammonia, along with lower energy costs. In contrast, the nonlinear model reduces the optimal equipment capacity, incorporates internal energy losses, and reflects higher electricity consumption. This leads to lower revenues, a 26.2% decrease in NPV, a 10.6% reduction in IRR, and an 8.6% increase in LCOA. These outcomes demonstrate that including nonlinear relationships and internal losses captures, with greater accuracy, the inefficiencies present in real system operation(...).
Description
Keywords
Hidrógeno verde, Planta híbrida, Electrolizador PEM, Optimización no lineal, Planificación energética, Síntesis de amoníaco, Baterías BESS
