Abstract
In this current paper, a sensor array based on WO3 gas sensors has been used to detect O-3. Comparing to the classical analysis methods, metal oxide sensors present the advantages of low cost, short response time and versatility.
However, the use of these sensors is limited due to their lack of selectivity. Representative variable as gas response and time response have been extracted from gas response base. Considering concentration range from 50 ppb to 400 ppb, database is established for four gas sensor. Then, the variables are grouped in database and tested with principal component analysis (PCA) to evaluate the contribution to the classification to enhance the sensors selectivity.