用基于神经网络的电子鼻评定卷烟香气质量

An Electronic Nose Based on Neural Network for Evaluating the Aroma Quality of Cigarettes

卷烟香气质量通常是靠人的嗅觉感受评定的 ,评定结果的准确性往往难以保证。针对这一问题 ,运用气敏传感器阵列技术、神经网络技术构造了一种电子鼻 ,用来评定卷烟的香气质量。实例表明 ,5个 TGS- 8系列的气敏传感器 (TGS812、TGS813、TGS82 2、TGS82 4和 TGS82 5 )组成的阵列及本文给出的神经网络学习算法是适用的 ,所构造的电子鼻可以实现卷烟香气质量的评定。该研究为卷烟质量的仪器化、客观化评定奠定了一些研究基础。

The aroma quality of cigarettes is important quota, but it has been evaluated by sense organ of people so far, the accuracy of evaluation results is hard to be guaranteed. In this paper, and electronic nose is construc- ted using gas sensor array (composed with TGS812, TGS813, TGS822, TGS824 and TGS825) and radial basis function (RBF) neural network in order to evaluate the aroma quality of cigarettes. A concrete application shows that the array and the RBF neural network learning algorithm given by this paper are suitable, the electronic nose can be used to evaluate the aroma quality of cigarettes. Some foundations for evaluating cigarettes quality with intelligence apparatus are also laid.