两种数据融合算法对扩散硅压力传感器的温度补偿

针对扩散硅压力传感器在实际应用中对温度存在交叉灵敏度的问题,文章采用改进的多维回归分析法和BP神经网络法两种实用的智能化数据融合方法对压力传感器输出进行处理,以消除非目标参量(温度)对传感器的影响;研究结果表明这两种方法均能有效地抑制交叉灵敏度,减小灵敏度温漂和零点温漂;由于这两种融合算法具有各自的特点,因此可应用于不同要求的智能传感器系统中。

Temperature Compensation of Pressure Sensor Based on TWO Data Fusion Arithmetic

For pressure sensor,the problem of intercross sensitivity always exists in the application.So two applied methods of intelligent data fusion,which are multi-dimension regress analysis and BP neural network,are used to eliminate affection caused by non-objective parameter(temperature).The results show that these two methods both can restrain intercross sensitivity effectively,depress the pressure fluctuation of sensor,reduce coefficient of temperature accuracy coefficient.With their own characteristics,they can be used in different smart sensor systems.