基于多BP子网络的电子鼻信息融合技术  

Data fusion technique in electronic nose based on multi BP sub-network

作　　者：王米娜[1] 杨建华[1] 侯宏[2] 王晗[1] 

机构地区：[1]西北工业大学自动控制系,陕西西安710072 [2]西北工业大学应用物理系,陕西西安710072

出　　处：《传感器技术》2003年第11期75-77,80,共4页Journal of Transducer Technology

基　　金：国家自然基金(69902006);西北工业大学'英才计划'资助

摘　　要：针对电子鼻技术的应用中,利用常规BP(back-propagation)网络难以对复杂的混合气味进行识别这一问题,介绍了一种采用多BP网络联合进行电子鼻多传感器数据融合处理的新方法。利用这一方法并结合气体传感器阵列在不同温度条件下的响应信号,对不同品牌的葡萄酒样本进行定性识别,仿真验证结果表明该方法的识别准确率达到100%。In respect that it is usually difficult to identify the complicated odor successfully using general BP network in electronic nose (Enose),a new method to use multi BP subnetwork data fusion technique is presented for identifying the complicated odor in E-nose. Some samples of different brand wines are qualitatively identified by this method and in combination with the response signal of gas sensor array under different temperature.Result shows that preciseness rate of recognition reaches 100%.

关 键 词：BP神经网络 电子鼻 气体传感器阵列 信息融合 

分 类 号：TP212.2[自动化与计算机技术—检测技术与自动化装置]