An artificial neural network-based smart capacitive pressure sensor

Abstract

A smart capacitive pressure sensor (CPS) using a multi-layer artificial neural network is proposed in this paper. A switched capacitor circuit (SCC) converts change in capacitance of the CPS due to applied pressure into a proportional voltage. The nonlinear characteristics of the CPS make the SCC output nonlinear. Further, due to dependence of the CPS characteristics on ambient temperature, the SCC output becomes quite complex for obtaining correct digital output of the applied pressure, especially when the ambient temperature varies with time and/or place. To circumvent this difficulty, an ANN is employed to model the sensor. By training the ANN model suitably, the digital readout of the applied pressure can be obtained which is independent of ambient temperature. A new idea for collecting temperature information from the sensor characteristics themselves, and automatic feeding of this information into the ANN-based CPS model is proposed. From the simulation results it is verified that the ANN model can give correct readout of the applied pressure within ±1% error (FS) over a wide range of temperature variation starting from −20°C to 70°C. This modeling technique of the CPS provides greater flexibility and accuracy in a changing environment.