3.3. The integrate quantitative model of AV for all the six kinds of oils
The individual models obtained above can only be applied to predicting the AV of a given kind of oil which belongs to the same kind of oil used for modeling, so it is necessary to ascertain the sort of an oil sample prior to prediction, consequently its applicability may be limited. In order to expand the application of this method, an integrate model using all the spectral data of the six kinds of oils was tried to establish. The 384 samples of the six kinds of oils were split into a calibration set (n = 288, in which 48 samples of each kind of oil were included) and a prediction set (n = 96, in which 16 samples of each kind of oil were included) with the principle of uniform distribution of AV. The modeling process also included three steps: wavelength selection, spectral-pretreatment and outlier exclusion, as done in the development of individual models above
3.3. The integrate quantitative model of AV for all the six kinds of oilsThe individual models obtained above can only be applied to predicting the AV of a given kind of oil which belongs to the same kind of oil used for modeling, so it is necessary to ascertain the sort of an oil sample prior to prediction, consequently its applicability may be limited. In order to expand the application of this method, an integrate model using all the spectral data of the six kinds of oils was tried to establish. The 384 samples of the six kinds of oils were split into a calibration set (n = 288, in which 48 samples of each kind of oil were included) and a prediction set (n = 96, in which 16 samples of each kind of oil were included) with the principle of uniform distribution of AV. The modeling process also included three steps: wavelength selection, spectral-pretreatment and outlier exclusion, as done in the development of individual models above
การแปล กรุณารอสักครู่..