Design of Experiments for optimizing and selectivity study of mems sensor array based electronic nose for hazardous gases measurement

Abstract

The present work highlights the use of design of experiments for selection and selectivity study of the MEMS based gas sensor array employed in electronic nose systems with application towards measurement of air pollution. Sensor array are crucial part of electronic nose systems as well as IoT systems. A plankett-burman design was developed for mixture of gas samples consisting of Hydrogen Sulphide, Methane, Carbon Monoxide, Ammonia and Ethanol. The 12 experimental sets were then exposed to sensor array to study the effect of individual as well as mixture of gases on the sensor array response. Pareto charts and ANOVA results were studied to infer the effect of various gases (single and mixture) on the response of the sensor. Further Principal Component Analysis (PCA) was applied on the dataset to study the selective of the developed sensor array.