摘要
为了提高智能型红外气体传感器的性能,通过采集数据分析了数据特点。在仪器的研制中采用数字滤波和神经网络的数据处理方法以克服干扰、提高稳定性和准确性。证明数字滤波方法适用于减小红外气体传感器数据的随机误差,神经网络方法可以拟合出平滑的浓度-比值曲线;应用这些数据处理方法可以充分发挥智能型红外气体传感器的优异性能。
To improve the performance of intelligent infrared gas sensor, through collecting data, the features of data are analyzed. In development of the instrument, the methods of digital filtering and neural network for data processing are adopted for anti-nterference and enhancing stability and accuracy. It is proved that digital filtering method is suitable for decreasing random error of infrared gas sensor, while neural network method is able to fit smooth curves of concentration vs. ratio, By using these methods of data processing, the superior performance of intelligent infrared gas sensor can be fully demonstrated.
出处
《自动化仪表》
CAS
2007年第4期57-60,共4页
Process Automation Instrumentation
关键词
红外气体传感器
数据处理
数字滤波
BP网络
Infrared gas sensor
Data processing
Digital filter
Back-propagation network
