Diseño y simulación de un Sistema de Tracking basado en redes neuronales para mantener la máxima eficiencia de paneles solares

This research work focuses on the use of artificial neural networks (ANN) as a promising tool to improve the efficiency and stability of solar photovoltaic systems. Although photovoltaic systems harness a clean and renewable energy source, they face challenges due to variations in solar radiation, temperature and environmental conditions. These factors cause fluctuations in the output current and voltage of the solar panels, affecting the power generated. To address this problem, it is necessary to implement control strategies that maximize power extraction from the photovoltaic field. The main focus of this work is the maximum power point (MPP), which represents the optimal power transfer point on the current-voltage (I-V) characteristic curve of a solar panel. The challenge lies in adapting to changing conditions and achieving accurate monitoring of the MPP to improve system efficiency. Although there are different proposed tracking algorithms, they have shown limitations in terms of tracking rates and steady-state oscillations. To overcome these deficiencies, the applications of ANNs in the design of control algorithms are explored. ANNs stand out for their high dynamic response and ability to adapt to non-linear conditions. However, obtaining accurate training data for the controller is one of the main challenges. In this study, important variables such as solar radiation, temperature and optimal voltage are considered as inputs to the controller.

• Autor Principal: Tapia Palma, Jessy Corina (author)
• Formato: masterThesis
• Idioma: spa
• Publicado: 2023
• Subjects: REDES NEURONALES ARTIFICIALES (RNA); EFICIENCIA; SISTEMAS SOLARES FOTOVOLTAICOS; PUNTO DE MÁXIMA POTENCIA (MPP); ELÉCTRICA
• Acceso en liña: http://repositorio.utc.edu.ec/handle/27000/10772