Use of spectrograph NIRS 6500 for prediction of wheat flour quality characteristics
2000
Pekarkova, J. | Hruskova, M. (Vysoka Skola Chemicko-technologicka, Prague (Czech Republic)) | Novotny, F. | Jurecka, D. | Moravkova, E.
Quality of 64 spring variety wheat flour samples from the Central Institute for Supervising and Testing in Agriculture tests prepared under standard laboratory conditions was assessed by estimation of some milling (flour humidity, yield of flour, ash content) and baking characteristics (indirect falling number, gluten and protein content, Zeleny sedimentation value, swelling number and farinograph hydration, stability, weakness and development and direct-loaf volume by baking test RMT). On spectrograph NIRSystems 6500 were measured NIR spectra of all samples. Calibration equations with cross validation for all characteristics above were computed by NIR Software ISI Presents Winisi II. using mPLS a PLS methods. The quality of prediction was evaluated by correlation coefficients between laboratory measured values and those predicted from cross validation. Statistically significant dependence between predicted and laboratory measured values (with probability higher than 99 %) was observed in all mentioned characteristics except falling number. The order of characteristics on dependence of correlation coefficient value is as follows: protein content, flour humidity, hydration of flour, gluten content, farinograph stability of flour; swelling number, yield of flour and Zeleny sedimentation value; time of dough development, ash content and dough weakness; loaf volume (RMT). Differences of flour samples from common Czech commercial white flours were probably due to wheat grain material and milling technology. It is necessary to verify the obtained calibration equations for the prediction of common commercial flour characteristics.
Показать больше [+] Меньше [-]Ключевые слова АГРОВОК
Библиографическая информация
Эту запись предоставил Library of Antonin Svehla