Study On Determination Of Solvent Retention Capacities Of Wheat Flour By Using Near Infrared Spectrum Reflection Technology

In terms of quality evaluation of wheat flour, solvent retention capacities(SRC) of Wheat Flour is developing rapidly in recent years. SRC of wheat flour could predict the physicochemical properties of flours through the establishment of correlation relationship between the physicochemical properties of flours and SRC of flour to 4 kinds of solvent including water, sucrose, sodium carbonate and lactic acid. But SRC method isn’t used well in wheat breeding and real-time monitoring of tailored flour production and guidance, because SRC method is limited by the amount of samples, measurement time and cost. Featured in simplicity, celerity, green, efficient and non-destructive analytical measurement, NIR(Near infrared spectroscopy) has been applied in the aspect of determination of grain quality.This research systematically introduced the progress of SRC of wheat flour, the principle of NIR analysis technology and application in grain quality detection. This research emphatically discussed the effect of spectral acquisition parameters such as scanning frequency of NIR, sample thickness and temperature on the spectral acquisition. The NIR quantitative analysis model of SRC of wheat flour WSRC, SUSRC, SCSRC, LASRC were established by partial least squares and by changing experimental conditions such as wavelength range, spectral pretretment method and main factor, with the principle that the higher calibration set correlation coefficient of model established and the validation set correlation coefficient of the sample, and the lower predicted root mean square error(SEC)of calibration set and the predicted root mean square error(SEP)of validation set.The calibration set correlation coefficient(Rc2)of the final NIR quantitative analysis model established for WSRC、SUSRC、SCSRC、LASRC of wheat flour were 0.9862、0.9586、0.9793、0.9223, respectively; the predicted root mean square error(SEC)of calibration set were 1.1975、2.7810、2.4887、6.8698, respectively; the validation set correlation coefficient(Rv2)were 0.9854、0.9480、0.9757、0.9026; the predicted root mean square error(SEP)of validation set were 1.2410、3.1577、2.6992、7.6570, respectively. This study provided a new technological method for the rapid detection of SRC of wheat flour.