Abstract:Reference data are indispensable to build near-infrared spectroscopy (NIR)calibration models. In the present paper, the effects of the accuracy of reference data on NIR calibration models and its prediction results were studied through two routine applications based on partial least square regression methods. The results indicate that the best NIR calibration statistics and the most accurate prediction results were aligned with the most accurate reference data. However, based on statistical analysis of numerous calibration samples, it is possible for NIR calibration models to obtain more accurate prediction results than the laboratory reference data used in the calibration sets. It is better to make less search for high accurate reference data and instead to introduce more calibration samples to improve the ruggedness of the calibration models.
Key words:Near-infrared spectroscopy;Chemometrics;Partial least square;Calibration model;Gasoline;Octane number
褚小立,袁洪福,陆婉珍. 基础数据准确性对近红外光谱分析结果的影响[J]. 光谱学与光谱分析, 2005, 25(06): 886-889.
CHU Xiao-li,YUAN Hong-fu,LU Wan-zhen. Effects of the Accuracy of Reference Data on NIR Prediction Results . SPECTROSCOPY AND SPECTRAL ANALYSIS, 2005, 25(06): 886-889.
[1] Gabor John Kemeny. Handbook of Near-Infrared Analysis. New York:Marcel Dekker Inc.,2001. [2] XU Guang-tong,YUAN Hong-fu,LU Wan-zhen(徐广通,袁洪福,陆婉珍). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2000,20(2):134. [3] HUANG Lan,DING Hai-shu,WANG Guang-zhi(黄 岚,丁海曙,王广志). Spectroscopy and Spectral Analysis(光谱学与光谱分析),2002,22(3):387. [4] Kennard R W,Stone L A. Technometrics,1969,11(3):137.