气体传感器阵列温度漂移抑制的研究

半导体气体传感器存在漂移问题,温度变化对漂移的影响尤为明显.在气体传感器阵列中,可以加入温度、湿度等传感器,监测其工作环境.实验系统采用恒温箱设定一组温度,制备气体样本20例(两种浓度样本各10例),采集传感器对样本的响应;通过人工神经网络来识别样本;当有误判发生时,在原网络中引入温度传感器的响应值,消除了误判,在一定程度上抑制了漂移,改善了网络性能,验证了该温度漂移抑制方法的可行性.

Research on the Temperature Drift Suppressing of Gas Sensor Array

Semiconductor gas sensors have a problem of drift, which is obviously affected by the change of the circumstance temperature. Temperature sensor, and humidity sensor could be used in gas sensor array to monitor working circumstance of the gas sensors. A thermostat was employed to set a series of temperatures in the experiment system, 20 gas samples (ten for each concentration) were prepared and the response of sensors to these gas samples was collected. Artificial neural networks were used to recognize samples. The response of temperature sensor was introduced into the original networks and eliminated the misjudgments. Drift was suppressed to some extent and the performance of networks was also improved, which eventually verified the validity of the method for temperature drift suppressing.