Neural network-based intelligent sensor operating in harsh environment

An artificial neural network (NN)-based smart interface for sensors operating in harsh environments is proposed to compensate for nonlinear response characteristics and the complex nonlinear dependency of the sensor characteristics on the environmental parameters. To show the potential of the proposed NN-based framework, we provide results of a smart capacitive pressure sensor (CPS) operating under a wide temperature range of 0 to 250⁰C. Three forms of nonlinear dependency of the sensor characteristics on temperature were considered. Through extensive simulated experiments, we have shown that the multilayer perceptron (MLP)-based CPS model can provide pressure readout with a maximum full-scale (FS) error of only ±1.0% over this temperature range. A novel scheme for estimation of the ambient temperature from the sensor characteristics, by using a second MLP, is proposed. A microcontroller unit (MCU)-based implementation scheme is also provided.