Characterisation of humidity dependence of a metal oxide semiconductor (MOS) sensor array using partial least squares (PLS)

The humidity dependence of a sensor array, which consists of 12 MOS sensors was studied. The results were used to develop calibration models using partial least squares. The model output appeared credible and accurate, having values for the correlation coefficient, root-mean-square error of calibration, and root-mean-square error of cross-validation of 0.99, 1.693 and 1.698, respectively. The model reduced moisture-related variance in the sensor array and therefore, enhanced sample classification and predictive accuracy of subsequent pattern recognition algorithms. Further improvement could be obtained by incorporating proper sensor drift reduction algorithms. This is an abstract of a paper presented at the 14th International Union of Air pollution Prevention and Environmental Protection Associations World Congress 2007 (Brisbane, Queensland 9/9-13/2007).