Application of artificial neural networks to prediction of chemical composition of electrodeposited Ni-Mo thin films

Allahyarzadeh, M. H ; Sharif University of Technology | 2012

603 Viewed
  1. Type of Document: Article
  2. DOI: 10.1149/05052.0063ecst
  3. Publisher: 2012
  4. Abstract:
  5. Present research represents the application of artificial neural networks to predict the chemical composition of electrodeposited Ni-Mo thin films. Artificial neural networks commonly are utilized as a prediction tools so that these networks could approximately find kind of logic relationships between inputs and target; they fitted appropriate coefficient and weighting factors to the inputs which are proportional to their importance. In order to evaluate the model developed, experimental results were compared with the predicted ones. However, more data are required to train more reliable prediction models, presents study revealed an acceptable error less than 1% between predicted values and experimental data. The Ni-Mo thin films also were conducted from citrate-ammonia deposition bath. Microstructure, morphology and chemical composition of deposited films was studied using scanning electron microscopy (SEM) and energy-dispersive X-ray (EDX) analysis
  6. Keywords:
  7. Chemical compositions ; Deposited films ; Deposition bath ; Energy dispersive x-ray ; Prediction model ; Prediction tools ; Weighting factors ; Electrodeposition ; Forecasting ; Neural networks ; Nickel ; Scanning electron microscopy ; Thin films ; Deposition
  8. Source: ECS Transactions ; Volume 50, Issue 52 , Oct , 2012 , Pages 63-71 ; 19385862 (ISSN)
  9. URL: http://ecst.ecsdl.org/content/50/52/63