【摘要】 本研究应用近红外光谱快速检测技术,对生产线上的豆粕样品进行扫描,建立近红外光谱检测预测模型,分别采用近红外光谱检测及湿化学方法检测相同样品后,大豆粕水分、蛋白、脂肪含量的SEP分别为0.111、0.238和0.109,准确度与湿化学方法相近,对近红外检测技术引入的点位进行了摸索,最终将近红外检测点设定在蒸脱机后及成品粕处。使用近红外检测技术持续扫描生产线豆粕样品,及时反馈回工艺控制,产品质量更为稳定,生产控制的精确度和准确度较好。更多还原

【Abstract】 The near-infrared spectroscopy was used to rapidly detect the soybean meal on the production line. Establish forecast model of near infrared spectra of near infrared spectrum detection and wet chemical methods were used respectively to detect the same samples, moisture, protein and fat content model of SEP is 0.111, 0.238 and 0.109 respectively, the accuracy of similar with wet chemical methods. Groping the point of near infrared detection technology, finally, we put the near infrared detection technology points after DT and packaging. Using the near infrared detection technology scan the soybean meal samples in the production line and timely feedback to the process control. The product quality is more stable and the accuracy of production control are better.更多还原