Research on Pressure Sensor Temperature Compensation by WNN Based on FA

摘要

For the sake of eliminating the temperature influence on the output of sensor, wavelet neural network(WNN) based on factor analysis is proposed, and the validity of the method is tested. It realizes screening and dimensionality reduction for primitive data through factor analysis(FA), decreases data redundancy regardless of related data, we make use of the nonlinear reflection ability and association learning ability of wavelet neural network, and proposes a model of pressure sensor temperature compensation. The result shows that the method is efficient for the problem of temperature drift, the stability of the pressure sensor is improved.