Odour bags from 10 sewage treatment works were used to assess the application of using an electronic nose to monitor sewage related odours. Canonical correlation (a data reduction technique) was used to compare the multivariate data generated by the NOSE (Neotronics Olfactory Sensing Equipment) with independent sensory panel analysis. Comparisons between the NOSE output and threshold odour numbers (TON) showed that no universal relationship exists between the NOSE and TON for sewage odours from a range of locations within different works. However, when similar groups of data from the same works (ie. biofilters) were compared, a linear relationship can be obtained between the NOSE output and corresponding TON values. Furthermore, this relationship could be extended to similar odours samples from different works, when only low TON values (ie. less than 4000 ou/m(3)) were analysed. (C) 1998 Published by Elsevier Science Ltd. All rights reserved. [References: 13]
展开▼
机译：使用了来自10个污水处理厂的除臭袋，以评估使用电子鼻监测污水相关气味的应用。使用规范相关性（一种数据缩减技术）将由NOSE（电子嗅觉传感设备）产生的多元数据与独立的感官面板分析进行比较。对NOSE排放量和阈值臭味数（TON）的比较表明，在不同工程范围内，NOSE和TON对于污水臭味之间不存在普遍关系。但是，当比较来自相同作品（即生物过滤器）的相似数据组时，可以在NOSE输出和相应的TON值之间获得线性关系。此外，当仅分析较低的TON值（即小于4000 ou / m（3））时，这种关系可以扩展到来自不同作品的类似气味样品。 （C）1998由Elsevier Science Ltd.出版。保留所有权利。 [参考：13]
展开▼
