Identifying Chemical Species in a Plasma Using a Neural Network

Authors:
Publisher website
Google Scholar
Abstract
This paper addresses the problem of identifying the chemical species present in a plasma using optical emission spectra. Two neural networks are initially explored, a multilayer perceptron and a linear associative memory which is trained using the Penrose pseudoinverse matrix. However, examination of the spectra leads to a proposal of a more appropriate architecture in which individual neurons are trained on individual species.