What is it about?
This article investigates the application of Artificial Neural Networks (ANNs) to model Phase-Change Materials (PCMs) heat capacity using data from Differential Scanning Calorimetry (DSC) tests and experimentations. Coefficients of determination of 0.99 and 0.66 are respectively obtained using two (DSC test) and four (experimentations) independent variables to simulate the dependent variable, i.e., PCM heat capacity. The independent variables include the PCM temperature and heat transfer characteristics such as the heating/cooling rate, heating/cooling duration, and the previous state (temperature and heat capacity). These results show the ability of ANNs for PCM modeling if meaningful independent variables are used.
Featured Image
Why is it important?
It provides a new method for modeling phase change materials using available experimental data without developing a physical model through neural networks.
Perspectives
This paper proposes a new method for modeling phase change materials using experimental data through neural networks in order to accelerate the simulations and to increase the accuracy.
Benoit Delcroix
Ecole Polytechnique de Montreal
Read the Original
This page is a summary of:
MODELING PHASE-CHANGE MATERIALS HEAT CAPACITY USING ARTIFICIAL NEURAL NETWORKS
, Heat Transfer Research, January 2018, Begell House,
DOI: 10.1615/heattransres.2018020080.
You can read the full text:
Contributors
The following have contributed to this page
