Neural network-based waveguide acoustic gas detector

Abstract

At present, gas chromatography is a universal technique used for analysis. The technical features of gas chromatographs are determined by the properties of gas detectors. One of the most recent and perspective gas detectors is the waveguide acoustic detector, in which chromatogram represents the mass concentration of the gas to be detected. With MATLAB computational language and iteration algorithm, a neural network-based waveguide detector is proposed, to predict the frequency and mass concentration of the unknown gas (sample). Experimental data has been chosen to create the database of the neural network-based detector. The proposed model has been tested and validated numerically with results.