摘要
针对野外文物古迹环境的多变性、传感器气敏性失效以及腐蚀性气体定性不精确等实际情况,结合物联网技术和模式识别技术,在综合考虑准确度和项目实际要求的情况下,设计了一种基于声表面波(SAW)传感器阵列的模式识别算法,并通过MATLAB对其进行仿真和验证。结果证明,将传感器阵列输出的数据输入到有6个隐层的神经单元的单隐层BP神经网络中进行训练,预测效果最好,对腐蚀性气体的识别率达到了95%左右,提高了野外微气象环境下腐蚀性气体的监测水平。
Aiming at the physical truth of the variety condition of the cultural relics,gas sensor failure and qualitative imprecision of the corrosive gases,based on the Internet of Things technology and pattern recognition technology,in the case of considering the accuracy and the actual requirements of the project,apattern recognition algorithm based on SAW sensor array is designed in this paper,Meanwhile,the simulation and verification are carried out based on MATLAB.It is shown that the best prediction results are obtained when the sensor array output is used as the actual input of the neural network,then the training and testing of this single layer BP neural network with six neurons in hidden layer are performed,The recognition rate of the corrosive gas is up to about 95%,which improves the monitoring level of corrosive gas in the field micro meteorological environment.
出处
《压电与声光》
CSCD
北大核心
2017年第4期549-552,共4页
Piezoelectrics & Acoustooptics
基金
重庆市教育委员会基金资助项目(KJ1500433)
重庆邮电大学自然科学基金资助项目(A2012-97)
2014年重邮文峰创新创业基金资助项目
