摘要
为了改进传统基于人工感官的白酒品质评价,利用z NoseTM电子鼻结合感官评价检测同一企业不同品质等级的白酒样品。分析电子鼻指纹图谱各特征峰,发现指纹图谱中获得的"峰面积总和"参数,与感官评价对白酒质量等级的排序一致,可以成为确定品质级别的参数。分别对指纹图谱采用典型判别分析和主成分分析方法,结果表明:典型判别分析对6种不同等级酒样的区分度更大,准确区分率可达100%。因此,文中提出的基于电子鼻的白酒品质检测方法研究,能快速、准确表达出白酒的品质及风格特点,在白酒生产中通过气味在线检测来改进工艺。
In order to improve quality inspection based on conventional sensory evaluation,the liquor samples with different quality levels in the same company were analyzed with z Nose？ electronic nose and the sensory evaluation. The structures of the characteristic peaks in the electronic nose fingerprints were analyzed. ‘Sum of PA＇parameters described the quality level accurately and were arranged in order according to the sensory evaluation. The results showed that ‘Sum of PA＇parameter could accurately describe the liquor quality level. Principal Component Analysis（ PCA） and Canonical Discriminant Analysis（ CDA） were adopted for data analyses. CDA had a greater distinction compared to PDA,and its identification rate reached 100%. The model of quality inspection of liquor based on electronic nose can quickly and accurately express the quality level and characteristic style of liquor. It is feasible for quality-online inspection to improve liquor production.
出处
《食品与发酵工业》
CAS
CSCD
北大核心
2015年第4期160-164,共5页
Food and Fermentation Industries
基金
国家自然基金项目(编号:21206051)
江苏省产学研联合创新资金(BY20130155-22)
