Publicou 4 edições por ano
ISSN Imprimir: 2152-2057
ISSN On-line: 2152-2073
Indexed in
EXPERIMENTAL AND NUMERICAL INVESTIGATIONS OF EFFECTIVE THERMAL CONDUCTIVITY OF LOW-DENSITY POLYETHYLENE FILLED WITH Ni AND NiO PARTICLES
RESUMO
In the present communication, we report our experimental results of Ni and NiO particles as a filler material to enhance the effective thermal conductivity (ETC) of low-density polyethylene (LDPE) composites. ETC of the present composites with varying volume fraction of fillers is measured using a 30-mm-long dual needle sensor (SH-1), which consists of two parallel needles spaced 6 mm apart. An artificial neural network (ANN) model is developed to predict ETC of these materials based on feedforward backpropagation (FFBP) networks with the training functions, i.e., Gradient descent (GD), Gradient descent with adaptive learning rate (GDA), Gradient descent with momentum (GDM), and Gradient descent with momentum and adaptive learning rate (GDX). The best outcome for the use of artificial neural network appertained to feedforward backpropagation network with different training and threshold functions, i.e., Tangent sigmoid (TANSIG) and Pure-linear (PURELIN) functions.
-
Misiura A. I., Mamunya Ye. P., Kulish M. P., Metal-Filled Epoxy Composites: Mechanical Properties and Electrical/Thermal Conductivity, Journal of Macromolecular Science, Part B, 59, 2, 2020. Crossref