Processing Data from Catalytic Sensors for Recognition of Hydrogen in Mixtures of Combustible Gases

In this article the possibility of determining the hydrogen concentration in a multicomponent gas mixture using the principal component analysis is investigated. Source data were obtained by a system, consisting of 8 sensors, each of which measured its own response values. It was found that, the values of the principal components form linear dependences of concentration, which are proportional to each other. At the same time, a different hydrogen concentration, pure or in a multicomponent mixture, is uniquely determined. The results showed that the principal component analysis allows both visually distinguishing sensor responses at different concentrations, and using additional mathematical operations to obtain the concentration value.