Gas discrimination in an air-conditioned system

In this work, an identification system based on an array of semiconductor tin dioxide gas sensors has been developed. This system has provided a good success rate in the discrimination of carbon dioxide, forane R134a or their mixtures without a sensor dedicated to carbon dioxide. After a characterization of the five-sensor array, pretreatment was tested on the collected data to select the most representative parameters. Then the information contained in the resulting look-up table was analyzed with PCA, but no significant result was observed. Discriminant factorial analysis was then used and has shown a better separation of the different clusters. Next, unknown data were taken to validate the classification. The results show that a reliable system can be designed using nondedicated chemical semiconductor gas sensors.