Gas mixtures recognition using an array of amperometric gas sensors with drifting or faulty sensors

In this study, the possibility of using selected methods for diagnostics of performance of matrix composed with six amperometric electrochemical gas sensors is investigated. Measurements of sensor responses in selected concentrations of single toxic gases or gas mixtures were performed and were repeated over time to show sensor drift. Additionally, the studies on the sensors’ drift were performed. The drift has been recognized by Principal Component Analysis (PCA). The results show that the methods used in this work can be potentially applied to detect faulty or drifted sensor responses.