酒精发酵液关键指标近红外预测模型的建立与评价

Construction and Evaluation of Near-infrared Prediction Model for Main Products in Ethanol Fermentation

随着过程控制技术与新一代信息技术的深度融合,食品发酵工业的数字化升级正在加速。近红外光谱是一种重要的过程分析技术(PAT),有助于生产过程参数的及时优化。为了建立酒精发酵液中关键指标含量的近红外预测模型,收集了酒精工业生产装置发酵60 h发酵周期内的252份样品,使用高效液相色谱检测方法获得发酵液各指标的化学值,用在线近红外光谱仪采集样品的光谱信息,通过对原始光谱进行预处理,采用偏最小二乘法(PLS)对酒精发酵液中10个关键指标进行预测建模。结果表明:利用在线近红外检测方法可以获得较好的预测结果,能够满足酒精发酵液中关键指标的实时检测,对于生产技术的提升具有重要意义。

With the development and combination of online detection technique and new generation information technology,the digitalization of traditional fermentation industry is upgrading quickly.Near-infrared spectroscopy is an promising process analysis technology(PAT),it is useful for online optimization of fermentation process parameters.In order to build a near-infrared prediction model for the main contents in ethanol fermentation products,in which 252 samples were collected during a 60-hour fermentation cycle in a production facility.The chemical values of various contents of the fermentation broth were obtained using high performance liquid chromatography method.The spectral information of the samples was collected using near-infrared spectroscopy.After preprocessing the original spectra data,the partial least squares(PLS)method was applied to model ten main products in the ethanol fermentation,The results indicated that good prediction results were obtained by using online near-infrared detection model,and satisfied the real-time detection of key indicators in ethanol fermentation process,which was of great significance for the improvement of production management.