高分子湿度传感器湿滞和温度补偿神经网路算法研究

Research on High Humidity Sensor Hysteresis and Temperature Compensation Neural Network Algorithm

高分子湿度传感器的感湿特性存在湿滞环,并且传感器对温度比较敏感,需要进行湿度和温度补偿。本文构建了两个BP神经网路分别映射传感器感湿特性的升程和降成,并同时将环境温度参数作为神经网络输入的一部分,实现传感器湿滞映射和温度补偿。利用MATLAB神经网络工具箱对网络进行了离线训练和验证,结果表明该网络能够较理想地映射传感器的湿滞环,并能够达到温度补偿作用。

There is moisture hysteresis in the polymer humidity sensitive characteristics of humidity sensors, and the sensor is sensitive to temperature, humidity and temperature compensation required. Two BP Neural Networks were constructed in this research. One was designed to map the output voltage when humidity increased, and the other to map the output voltage when humidity reduced. As input variables, the parameters of environmental humility and temperatur were adopted in constructing the BP model, in order to accomplish humility and temperature compensation at the same time. The BP Neural Network model was trained with the neural network toolbox of MATLAB, and the result of training showed that the model was very good at humility and temperature composation.