Intellectual temperature compensation and correction method of capacitor pressure sensor

A novel method of constructing functional link artificial neural networks (FLANN) with support vector regression (SVR) was presented and applied to capacitor pressure sensor (CPS) correction in this paper. In the methods, SVR-FLANN was used as an inverse model, by which the sensorpsilas nonlinear characteristic was mapped. Thus the sensorpsilas temperature compensation and correction of nonlinear characteristic were realized synchronously. A generic FLANN had been developed to solve the same problem for comparison. The experiment results show, proposed method has the characteristics of unique results, simple structure and global minimum, so it is more suitable for sensorpsilas correction.