An intelligent pressure sensor using neural networks

In this paper, we propose a scheme of an intelligent capacitive pressure se nsor (CPS) using an artificial neural network (ANN). A switched-capacitor c ircuit (SCC) converts the change in capacitance of the pressure-sensor into an equivalent voltage. The effect of change in environmental conditions on the CPS and subsequently upon the output of the SCC is nonlinear in nature . Especially, change In ambient temperature causes response characteristics of the CPS to become highly nonlinear, and complex signal processing may b e required to obtain correct readout. The proposed ANN-based scheme incorporates intelligence into the sensor, It is revealed from the simulation studies that this CPS model can provide co rrect pressure readout within +/-1% error (full scale) over a range of temp erature variations from -20 degrees C to 70 degrees C. Two ANN schemes, dir ect modeling and inverse modeling of a CPS, are reported, The former modeli ng technique enables an estimate of the nonlinear sensor characteristics, w hereas the latter technique estimates the applied pressure which is used fo r direct digital readout. When there Is a change in ambient temperature, th e ANN automatically compensates for this change based on the distributive i nformation stored in its weights.