Correlation of sensory analysis with a virtual sensor array data for odour diagnosis of fragrant fabrics

Abstract

A single-sensor gas diagnosis device made by temperature modulation of a chemo-resistive gas sensor which acts as a virtual array was employed to quantitatively measure the odour emission of fragrant fabrics. Repeatability of the virtual sensor array response was confirmed by the determination of correlation coefficient between sensor responses for 14 experiments in four heating steps. Applying principle component analysis on a device responses revealed that the system would be able to discriminate fragrant fabrics based on fabric types, odour types and intensity. To compare this method with sensory evaluation, quad analysis method was performed by 14 assessors to rank the odour intensity of samples. The comparison of weighted assessors results and virtual array responses showed that there is a good correlation between the two methods with a correlation coefficient of r=0. 96. Therefore, the proposed single-sensor gas diagnosis device as a virtual sensor array could be a suitable alternative tool to odour assessment of fragrant fabrics for sensory panels.