Multi-Transduction Approach and Data Fusion for Enhanced Performance of Features Extraction in Chemical Sensing Applications

摘要

We investigate the sensing properties of the single-walled carbon nanotubes deposited onto cadmium arachidate buffer multilayer by acoustic and fiber optic sensors for VOCs detection, at room temperature. The carbon nanotubes are deposited by the Langmuir-Blodgett technique. The acoustic and optical sensors have been simultaneously exposed, in the same test cell, to individual VOCs of ethanol, methanol, iso-propanol, acetone, ethylacetate, toluene with different vapor pressures for comparing the sensitivity of the carbon nanotubes onto the several kinds of the sensors. Signals from different sensor types have been analyzed and processed by using pattern recognition techniques such as principal component analysis (PCA). The recognition and discrimination capabilities of the hybrid system are significantly improved when data fusion from acoustic and optical sensors is realized. To our knowledge, this is the first study reported of a combined hybrid integration of low- and high- frequency acoustic sensors with optical fiber sensors using nanostructured materials as single-walled carbon nanotubes for VOCs detection, at room temperature.