Device Design Space Exploration of Thin Film Hydrogen Sensor Based on Macro-model Generated Using Machine Learning

Abstract

An efficient attempt has been performed towards device design optimization, using machine learning approach for exploration of design space of zinc oxide (ZnO) thin film Schottky diode based hydrogen sensor. We have adopted Least Square Support Vector Machine (LS-SVM) to build the regression model to predict the output behavior of ZnO thin film Schottky diode based hydrogen sensors. ATLAS package from SILVACO international has been used for generating data set, that is required to train the machine learning model. The hydrogen induced barrier height variations (Δϕb) at a wide range of temperature (300 K to 575 K) and wide range of ZnO thin film thickness (5 nm to 300 nm) have been calculated, which was used used for training the regression model. It has been observed that the proposed modeling scheme can serve a guide for fabrication of ZnO thin film based Schottky diode for hydrogen sensing applications.