Machine Learning Surrogate Dynamical System Model for Thermal Energy Storage (#1795)
Read ArticleDate of Conference
July 17-19, 2024
Published In
"Sustainable Engineering for a Diverse, Equitable, and Inclusive Future at the Service of Education, Research, and Industry for a Society 5.0."
Location of Conference
Costa Rica
Authors
Diaz Sierra, Erimar F
Cruz De Jesus, Felix M
Jimenez Ortiz, Jorge J
Lebron Santana, Andres J
Nuñez Rios, Jeremy J
Traverso, Luis Miguel
Abstract
Abstract– A thermal energy storage (TES) can serve as a mean of minimizing energy losses when there is fluctuation of energy demand. A coupled fluid and conduction thermal model are performed to obtain the history of the temperature profile over 500 timesteps simulations. Results files are generated in text format and imported as numerical arrays in Python programing. These results are used to train a deep learning algorithm based on convolutional and dense layers. Two of these architectures are presented here. Under this architecture, results can match validation data for a certain number of cases with relatively low errors