Effect Of Distribution Of Sample Moisture, Amino Acid And Crude Protein On Prediction Model By Fourier Near-infrared

The effect of the distribution (Gaussian distribution and homogeneous distribution) of moisture content and crude protein and amino acids with controlling moisture content under 10-11% of soybean meal on the predicting results of the moisture content by Fourier-NIR was researched in this article. single factor experiment design was used. sample sets taking on Gaussian distribution and sample set with homogeneous distribution were designed, which were used to set up predicting model of moisture content, crude protein and amino acids with controlling moisture content under 10-11% respectively. The RMSECV of 3 Gaussian distribution model (a value: 0.67,0.99,1.17 ) and homogeneous distribution of moisture content were 0.151, 0.174, 0. 185 and 0. 186; The RMSECV of Gaussian distribution model of crude protein , aminosuccinic acid, threonine, serin, glumatic acid, glycine, alanine, valine, isoleucine, leucine, tyrosine, Phenylalanine, lysine, histidine, arginine and praline taking on Gaussian distribution model were 0.046,0.07,0.03,0.03,0.07,0.02,0.04,0.03,0.04,0.06,0.03,0.03,0.042,0.04,0.03,0.05 and 0.03, the RMSECV of homogeneous distribution model of crude protein , aminosuccinic acid, threonine, serin, glumatic acid, glycine, alanine, valine, isoleucine, leucine, tyrosine, Phenylalanine, lysine, histidine, arginine and praline were 0.47,0.08,0.03,0.04,0.08,0.03,0.05,0.03,0.04,0.07,0.04,0.05,0.03,0.05 and 0.03, Gaussian distribution model is better than homogeneous distribution model . When predicting the soybean meals with the average of moisture content was 11.47±1.28, the predicting results of 3 Gaussian distribution model (a value: 0.67,0.99,1.17 ) and homogeneous distribution were 12.23±1.29, 11.41±1.33, 11.48±1.24 and 11.47±1.15, homogeneous distribution is better than Gaussian distribution obviously. When predicting crude protein and amino acids with controlling moisture content under 10-11%) of soybean meals, homogeneous distribution is better than Gaussian distribution when predicting the high or low content ,but but when predicting the sample with the middle content, the reverse is true. Both Gaussian distribution and homogeneous distribution have their own advantages. When predicting the sample with high or low content, homogeneous distribution is more accurate than Gaussian distribution, but when predicting the sample with the middle content, the reverse is true.