摘要
压阻型扩散硅压力传感器在测试压力时,容易受到环境温度的影响。为了消除温度所带来的影响,需要对压力传感器进行温度补偿。神经网络技术中的BP神经网络算法可以在压力试验中对压力传感器进行温度补偿。此方法将压力传感器和温度传感器所采集到的电压信号进行数据融合,削弱了温度对压力传感器所产生的干扰,补偿后比补偿前得到压力传感器灵敏度温度系数和满量程时相对误差都分别提高了2个数量级。
Silicon piezoresistive pressure sensor is affected by the environmental temperature. In order to eliminate the influence of temperature, pressure sensor needs temperature compensation. In this paper, BP neural network algorithm is used to do the temperature compensation in pressure measurement. This method fuses the voltage signals selected by the pressure sensor and the temperature sensor, which weakens the interference of the temperature on the pressure sensor. Compared with the data before compensation, the data after compensation of the temperature sensitivity coefficient and the relative error in full range have been raised by two orders of magnitude respectively.
出处
《工程与试验》
2015年第1期66-69,79,共5页
Engineering and Test
关键词
压力传感器
温度补偿
BP神经网络
灵敏度温度系数
满量程时相对误差
pressure sensor
temperature compensation
BP neural network
temperature sensitivity coefficient
relative error in full range
