摘要
为了实现高粱水分指标含量的快速分析,并使分析结果能够及时有效地反馈到生产车间,以指导车间日常生产。采用近红外光谱技术对高粱样品进行光谱扫描,并结合传统分析方法,最终建立了高粱水分指标的快速分析模型。结果表明,所建模型的线性关系较好,R^(2)为0.9558,且模型内部参数RMSEC和RMSECV值均较小,分别为0.280和0.319,模型的质量较好;另取验证样品对模型的预测能力进行验证,通过与传统分析方法的分析结果进行比较,模型预测的平均相对误差小于2.0%,说明模型具有较好的预测能力,可用于实际生产分析检测,提高样品分析效率。
In order to quickly detect the water content of the sorghum, and the analysis results can be timely fed back to the production workshops, which can guide the daily production. Near infrared spectroscopy(NIR) combined with conventional analytical method was used to treat the NIR maps of the samples, and establish the rapid analysis model of water index in the sorghum. The results showed that the linear relations(R^(2)) of the model was relatively good, and the value was 0.9558, the model parameters of mean square error(RMSEC) and cross validation mean square error(RMSECV) were relatively small, and the values respectively were 0.280 and 0.319, it indicated that the quality of the model was relatively good. Moreover, another samples were taken to verify the prediction ability of the model, and the results were compared with those of conventional analytical method, it showed that the average relative error of the model was less than 2.0%, which indicated that the model had relatively good prediction ability and can be used for daily production analysis and improve analysis efficiency.
作者
苏鹏飞
刘丽丽
闫宗科
张攀峰
张武岗
SU Pengfei;LIU Lili;YAN Zongke;ZHANG Panfeng;ZHANG Wugang(Xifeng Distillery Co.,Ltd.,Baoji 721400,Shaanxi,China)
出处
《酿酒》
CAS
2021年第1期87-90,共4页
Liquor Making
