Error Processing Of Quartz Resonator In Humidity Measurement Based On Bp Neural Network

In this paper, a quartz resonator used in Quartz crystal microbalance (QCM) is used as the humidity sensitive element to directly measure the relative humidity, and a model of frequency-relative humidity(F-RH) based on BP neural network is established to eliminate the hysteresis and repeatability errors. The model is trained and tested by the data samples obtained from the humidity measurement experiment. The frequency-relative humidity(F-RH) model based on BP neural network can be used to calculate the relative humidity in the measured environment. This method can eliminate the hysteresis and repeatability errors of the quartz resonator in a certain extent, and make the humidity measurement result of the quartz resonator more accurate.