The commercially available metal-oxide TGS sensors are widely used in many applications due to the fact that they are inexpensive and considered to be reliable. However, they are partially selective and their responses are influenced by various factors, e.g. temperature or humidity level. Therefore, it is important to design a proper analysis system of the sensor responses. In this paper, the results of examinations of eight commercial TGS sensors combined in an array and measured over a period of a few months for the purpose of prediction of nitrogen dioxide concentration are presented. The measurements were performed at different relative humidity levels. PLS regression was employed as a method of quantitative analysis of the obtained sensor responses. The results of NO2 concentration prediction based on static and dynamic responses of sensors are compared. It is demonstrated that it is possible to predict the nitrogen dioxide concentration despite the influence of humidity.

Electrocatalytic gas sensors belong to the family of electrochemical solid state sensors. Their responses are acquired in the form of I-V plots as a result of application of cyclic voltammetry technique. In order to obtain information about the type of measured gas the multivariate data analysis and pattern classification techniques can be employed. However, there is a lack of information in literature about application of such techniques in case of standalone chemical sensors which are able to recognize more than one volatile compound. In this article we present the results of application of these techniques to the determination from a single electrocatalytic gas sensor of single concentrations of nitrogen dioxide, ammonia, sulfur dioxide and hydrogen sulfide. Two types of classifiers were evaluated, i.e. linear Partial Least Squares Discriminant Analysis (PLS-DA) and nonlinear Support Vector Machine (SVM). The efficiency of using PLS-DA and SVM methods are shown on both the raw voltammetric sensor responses and pre-processed responses using normalization and auto-scaling