Effective and Robust Parameter Identification Procedure of a Two-Site Langmuir Kinetics Model for a Gas Sensor Process

Abstract

Gas sensors have received plenty of attention due to various applications, and the methods to model the kinetic processes and estimate the corresponding parameters play a critical role in characterizing the sensor response behavior. In this work, a two-site Langmuir kinetics model is applied to describe the adsorption/desorption response processes of a SnO2/reduced graphene oxide resistive gas sensor and the pertinent kinetic parameters are optimized based on the genetic algorithm (GA). For the robustness and fast convergence of the GA, the initial values and ranges of kinetic parameters are obtained step-by-step. This a priori knowledge is sufficient to guarantee reasonable parameter identification from experimental data. Moreover, the kinetics model and GA are integrated into graphical user interface software for subsequent application. Eventually, the exploration of improvements to experimental design is uncovered to increase the accuracy and reliability of the estimation.