云端智能微型近红外光谱仪的酒醅定量分析模型开发

Development of Quantitative Analysis Models of Fermented Grains Based on Intelligent Micro Near Infrared Spectrometer and Cloud-Based Data Processing System

采用国产微型近红外光谱仪及配套的端云一体光谱智能建模系统,针对白酒酿造过程中出入窖酒醅进行定量分析模型开发,包含出窖酒醅水分、酸度、淀粉、残糖指标,入窖酒醅水分、酸度、淀粉指标,通过使用特制的酒醅压紧工装,以及不同的预处理算法组合,采用偏最小二乘法开发完成了上述共7种成分的近红外光谱定量分析模型并进行验证。

Using the micro near infrared spectrometer and the cloud-based spectrum data processing system made in China,we devel-oped the quantitative analysis models of the fermented grains.Adopting a specially designed sampling device,the combination of dif-ferent pre-processing methods,and the partial least squares method,the quantitative analysis models of the moisture,acidity,starch and residual sugar of the pit-exit fermented grains and the moisture,acidity and starch of the pit-entry fermented grains were devel-oped and validated.