Application Of Near-infrared Spectroscopy In Detecting The Qualities Of Flour Products

Near infrared spectroscopy (NIR) is one of the fastest growing and most widely used green analytical technologies in the world. Being capable of making nondestructive, rapid, high eficient and convenient, NIRS has been widely applied in food industry for quantity and quality analysis.In this paper, samples of instand noodle and dried noodle were scanned at the NIR region 4000cm-1-10000cm-1. It also explores conditions of making noodle powder, and finds the optimal sampling contents,tightness and granularity to scan the spectra. According to near infrared Spectral characters, quantitive models for the detemlination of chemical indecies in instand noodle and dried noodle,texture analysis indices in dried noodle were developed using partial least squares technique(PLS). At the same time the models were evalucated through the internal and extemal test. The results were as follows:(1)6 qualities models of instant noodle were established. The correlation coefficient of calibration (Rc) of moisture,fat,peroxide value(POV),acid value(AV),protein,ash models were 0.99,0.97,0.85,0.70,0.85,0.64, respectively; and Correlation coefficient of validation (Rv) were 0.99,0.95,0.83,0.69,0.86,0.58, respectively. The correlation coefficient of external test were 0.99(H2O),0.97(Fat),0.88(POV),0.95(AV),0.94(Protein),0.68(Ash). NIR analysis technique could be adopted to correctly measure the content of moisture and fat, and to predict POV,AV and Protein content roughly. However, it could hardly be employed on measurement of ash content in practice.(2)5 qualities models of dried noodle were established. Rc of moisture,protein,ash,acidity,fatty acid value (FAV)models were 0.96,0.97,0.81,0.85,0.79, respectively; and Rv were 0.96,0.96,0.79,0.85,0.74, respectively. The correlation coefficient of external test were0.99(H2O),0.99(Protein),0.71(Ash),0.90(acidity),0.76(FAV). NIR analysis technique could be adopted to correctly measure the content of moisture and protein, and to predict ash content and acidity roughly. However, it could hardly be employed on measurement of fatty acid value in practice.(3)5 texture analysis indices models of dried noodle were established. Rc of tensile stress,flexural strength stress,chewiness,wring stress and hardness models were 0.83,0.83,0.89,073,0.72, respectively; and Rv were 0.87,0.81,0.89,0.71,0.73, respectively. Their correlation coefficient of external test were0.86,0.81,0.70,0.70,0.82. NIR analysis technique could be adopted to measure tensile stress and flexural strength stress, and to predict chewiness roughly. However, it could hardly be employed on measurement of wringness stress and hardness in practice.