Application of Principal Component-Artificial Neural Networks in Near Infrared Spectroscopy Quantitative Analysis
The principal components of near infrared spectroscopy were calculated by nonlinear iterative partial least squares (NIPALS). The best number of principal components was determined by cross-validation method. Thus, limited principal components that free from noise and orthogonal each other were obtained. After standardization, these principal components were used as input nodes of back propagation artificial neural networks (B-P ANN). ANN was used to build nonlinear model. In the method, the data of whole spectra can be fully utilized, the best principal components free from noise and nonlinear model obtained, the iterative time of B-P ANN shorted strongly, and better calibration model can be obtained. The method has been applied to quantitatively determine the starch of barely. The calibration and prediction correlation coefficients are 0.982 and 0.945; the relative standard deviations are 1.81% and 2.80%, respectively.
KeywordsHide Layer Partial Less Square Input Node Near Infrared Spectroscopy Iterative Time
Unable to display preview. Download preview PDF.
- 7.Ji, H.Y., Yan, Y.L.: The Quantitative Analysis for Cereal’s Components by Artificial Neural Networks. Chemical Journal of Chinese Universities 14(5), 618–620 (1993)Google Scholar