Standardization methods for handling instrument related signal shift in gas-sensor array measurement data Original Research Article

This paper describes two different approaches that attempt to solve the problem of signal shift between measurements acquired with gas-sensor array systems of identical construction. Both methods provide standardization models that can be used to compensate such instrument related signal shifts by postprocessing of the measurement data. The first approach is a straightforward univariate direct standardization method, based on linear regression, where unique shift compensation models are created for each sensor. The other approach is a multivariate method, based on partial least squares regression, which can be used to design shift compensation models for the whole gas-sensor array. Both methods effectively removed signal shift after being applied on measurement data acquired with five commercial instruments of identical configuration with quartz micro balance (QMB) sensor arrays.