Evaluation of the effective forcespinning parameters controlling polyvinyl alcohol nanofibers diameter using artificial neural network

Authors:
Publisher website
Google Scholar
Abstract
In this research, the polyvinyl alcohol (PVA) nanofibers through forcespinning process were successfully produced and the effective parameters for predicting nanofibers diameter using artificial neural network (ANN) were investigated. The various parameters of forcespinning process including rotational speed, orifice, distance to the collector, and polymer concentration were designed to produce PVA nanofibers. Scanning electron microscopy (SEM) showed that the produced fibers diameter was in the range of 0.56–1.9 μm. The neural network with four input factors, three hidden layers with 5, 10, 1 nodes in each layers, respectively, and one output layer had the best performance in the testing sets. Moreover, the mean squared error (MSE) and linear regression (R) between observed and predicted nanofibers diameter were about 0.1077 and 0.9387, respectively, demonstrating a suitable performance for the prediction of nanofibers diameter using the selected neural network model.
Funding Information
  • Student's Scientific Research Center, Tehran University of Medical Sciences, Tehran, Iran (93-04-61-27454)

This publication has 24 references indexed in Scilit: