论文部分内容阅读
为模拟生物化学传感体系,提出了可用于识别有机官能团的传感器阵列,用作人工气味识别系统.该阵列由八个压电晶体传感器组成,每个传感器涂以具有广谱响应性能的不同吸附活性材料.阵列对常见小分子有机溶剂混合蒸气的响应频移数据采用逐步判别分析(SDA)处理,选出五个供信能力最佳的判别变量,以此构成的阵列用于小分子有机溶剂混合蒸气中醇羟基、羰基与其它官能团的识别;并采用主成分分析(PCA)法降维投影,在二维空间含相同官能团的物质聚为一类;阵列可用于酒类、软饮料的识别.
To simulate biochemical sensing systems, a sensor array that can be used to identify organic functional groups is proposed as an artificial odor recognition system.The array consists of eight piezoelectric crystal sensors, each coated with a different adsorption with a broad spectrum of response properties Active materials.The array response frequency shift data of common small-molecule organic solvent mixed vapors were determined by stepwise discriminant analysis (SDA), and five discriminant variables with the best supposing ability were selected. The array was used for small molecule organic solvents Alcohol and hydroxyl groups in mixed vapors and other functional groups. The principal component analysis (PCA) method is used to reduce the dimension of the projection, and the substances containing the same functional group in two-dimensional space are clustered together. The array can be used to identify wine and soft drinks.