Correcting multivariate calibration model for near infrared spectral analysis without using standard samples

摘要: Model transfer is essential for practical applications of near infrared (NIR) spectroscopy because differences may exist between the spectra measured using different instruments. For correcting calibration models in cases where standard samples are unavailable, a method proposed based on relationship prediction error and spectral difference In this method, partial least squares (PLS) model, named primary was constructed with instrument, model then used to predict from secondary instrument. Error arise predicted obviously caused by two Assuming that there linear error, correction PLS can be built difference. The obtained correct model. Two NIR datasets, pharmaceutical tablets tobacco leaf samples, instruments were test performance method. results show that, both instrument satisfactorily corrected.