Algorithmic Correction of MOS Gas Sensor for Ambient Temperature and Relative Humidity Fluctuations

Abstract

In ambient conditions, uncontrolled fluctuations of ambient temperature and relative humidity significantly reduce the accuracy of metal oxide semiconductor-based gas sensors. The main innovation of the work is an algorithmic self-correction method that employs a response from an auxiliary temperature sensor and a multivariable polynomial regression model to minimize the effect of such environmental fluctuations. Tin oxide (SnO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> ) MOS gas sensors were chosen for a sample study due to their fast response times and high resolution. Using the developed correction method, the study analyzes sensor accuracy while measuring a typical pollutant gas (carbon monoxide) over a temperature range from −10 to 40°C at relative humidity intervals of 35%, 65%, and 95%. On average, the corrected sensors measured a 40 ppm CO gas sample with 11 times more accuracy when compared to uncorrected sensors placed in the same variable ambient conditions. Corroborating ANOVA tests demonstrated that the prescribed correction algorithm reduced measurement variability with statistical significance. In the presence of CO2, CH4, ethanol, LPG, or smoke interference gas samples, sensors applying the correction method and temperature-based selectivity distinguished and measured CO concentrations with a maximum error of only ±2.0 ppm. Thus, the proposed technical solution is attractive for highly variable environments and field applications where low-cost drift stabilization is necessary.