电子鼻传感器筛选的组合优化法研究  

Combination Optimization Method for Screening Sensor Array of Electronic Nose

作　　者：赵镭[1] 史波林[2] 汪厚银[1] 李志[1] 

机构地区：[1]中国标准化研究院食品与农业标准化研究所,北京100088 [2]上海大学通信与信息工程学院,上海200072

出　　处：《食品科学》2009年第20期367-370,共4页Food Science

基　　金："十一五"国家科技支撑计划项目(2006BAK0405);中国博士后科学基金资助项目(20080440614)

摘　　要：以电子鼻传感器对不同等级龙井茶样品的响应区分为研究对象,探索建立一种传感器阵列优化的方法——组合优化法。该方法的核心是以方差分析的组内均方值和P值或F值对单一传感器在该测试体系下的响应性能进行初筛和分组,再对分组后的传感器进行排列组合,对各种组合后的传感器信号进行主成分分析(PCA),并以判别指数(DI)作为衡量PCA分析结果及各传感器优化组合后的性能指标,最终确定适用于目标的优化传感器阵列。A method designated as Combination Optimization Method was established for sorting sensor array of electronic nose in discriminating different grades of Longjing tea. The method was involved in following 4 steps. First, the suitable sensor for electronic nose was selected according to the mean square value and P or F value of the testing response in ANOVA. Second, the selected sensors with the same level of P or F value were placed into the same group. Third, testing results of all possible combinations for different sensor groups were subjected to principal components analysis （PCA）. Finally, the optimized combination for sensor arrays was selected by discriminating index （DI） of PCA. This method will provide theoretical references for the optimization of sensor arrays in intelligent sensory application system.

关 键 词：电子鼻 传感器 筛选 组合优化法 

分 类 号：TS207.7[轻工技术与工程—食品科学]