Artificial Neural Networks Methods Applied to Conductometric Microhotplate Data for the Identification of the Type and Relative Concentration of Chemical Warfare Agents

Published

April 1, 2003

Author(s)

Zvi Boger, Douglas C. Meier, Richard E. Cavicchi, Stephen Semancik

Abstract

Response data from microhotplate (MHP) sensor arrays were measured for various chemical warfare (CW) agents at several concentrations. Efficient large-scale artificial neural networks (ANN) modeling has been evaluated as a method for the classification and concentration prediction of the CW agents based of the MHP data. Four MHP sensor elements, two pairs of SnO2 and two pairs of TiO2, were operated in a pulsed, ramped temperature mode to generate the data used. The CW agents and related compounds tested were tabun (GA), sarin (GB), sulfur mustard (HD), and chloroethyl-ethyl-sulfide (CEES), in four concentration levels in dry air, between several nano-mols/mol (ppb) to several micro-mols/mol (ppm). Recursive ANN pruning and re-training techniques were used for the identification of the more relevant inputs, out of the original 80 inputs (different sensor elements and temperatures). ANN models with 6 to 15 inputs produced good classification between the different CW agents. Other ANN models, trained for each agent, gave good prediction values for the concentration of the CW agents.

Proceedings Title

Neural Networks, International Joint Conference | | Neural Networks | Pergamon Press

Volume

2

Conference Dates

July 20-24, 2003

Conference Title

Proceedings of the International Joint Conference on Neural Networks 2003

Pub Type

Conferences

Keywords

chemical warfare agent, classification, gas sensor, microhotplate, microsensor, neural networks, tin oxide, titanium oxide

Citation

Boger, Z. , Meier, D. , Cavicchi, R. and Semancik, S. (2003), Artificial Neural Networks Methods Applied to Conductometric Microhotplate Data for the Identification of the Type and Relative Concentration of Chemical Warfare Agents, Neural Networks, International Joint Conference | | Neural Networks | Pergamon Press (Accessed April 25, 2024)