文摘
|
由气体传感器阵列构建的电子鼻系统在易挥发性化学品的快速检测中起着重要的作用. 利用 12路气体传感器阵列采集易挥发性化学品蒸气的响应曲线, 通过研究具有一定抗干扰能力的具有现场实用意义的电子鼻识别算法, 提出了一种基于BP神经网络的算法来提取特征向量. 所需计算量小, 稳定性高, 受浓度影响小; 在此基础上进行神经网络的训练并对样本进行测试, 达到了90%的平均识别率, 验证了算法具有实用性 |
其他语种文摘
|
Electronic nose made up of gas sensors array now is playing an important role in the detection of volatile chemicals. Aimed at the research on practical electronic nose recognition algorithm which has anti-jamming ability and can be used in local detecting, response curves of 9 kinds of volatile gases of chemicals are acquired by the array made up of 12 gas sensors. To make it more practical, a new method of algorithm is proposed for feature extraction based on the BP neural network. It needs less calculation, has high stability and is not easily influenced by the gas concentration. On the basis of the network training and test on the samples, and the average recognition rate reaches 90 % ,and it approves that this algorithm is practical |
来源
|
传感器与微系统
,2011,30(1):29-30,34 【核心库】
|
关键词
|
易挥发性化学品
;
电子鼻
;
气体传感器
;
特征提取
;
BP神经网络
|
地址
|
中国科学院合肥物质科学研究院, 安徽省仿生感知与先进机器人重点实验室, 安徽, 合肥, 230031
|
语种
|
中文 |
文献类型
|
研究性论文 |
ISSN
|
1000-9787 |
学科
|
自动化技术、计算机技术 |
基金
|
国家自然科学基金
|
文献收藏号
|
CSCD:4121382
|