Estudio para incorporación de inteligencia artificial en circuitos de transformadores de distribución para análisis de demanda, enlace y análisis de información

Abstract

This research focuses on the development of a thermal image prediction model that classifies images according to their degree of thermal degradation severity. The significance of this classification has been highlighted for the early detection of problems and decision-making in several fields. The methodology used in this study was based on collecting a dataset of thermal images with different degrees of thermal degradation. Image processing techniques were applied, and a deep learning algorithm was used to train and evaluate the prediction model. The results obtained showed that the developed prediction model was capable of classifying thermal images according to their degree of thermal degradation severity with an acceptable accuracy. A substantial evaluation was observed between the thermal characteristics of the images and the degree of thermal degradation. In conclusion, this study demonstrated the feasibility of developing a thermal image prediction model for classifying the severity of thermal degradation. However, some limitations were identified, including the need for a more diverse dataset and the exploration of more advanced image processing techniques. It is recommended to continue research and improve the prediction model, which includes collecting a more extensive and diverse dataset. Furthermore, it is suggested to explore more advanced image processing techniques and consider the integration of other relevant variables to improve the accuracy of the model.