Taking a Software Approach to Improve Accuracy of Commercially Available Methane Gas Sensors

Abstract:

In this paper we look at increasing the accuracy of commercially available MQ gas sensors using software. By utilizing Arduino or Arduino like micro-controllers this project aims to make an inexpensive portable and accurate methane sensor. MQ sensors detect multiple gasses and do not distinguish between individual gases so we decided to use multiple sensors to mitigate this and obtain a more accurate reading. We collected output from the MQ sensors individually at known concentrations of methane and other gases the MQ sensors detect this varies depending on the sensor. From this we are able to get multiple curves to determine specific concentration of methane. We then were able to develop a calibration curve using the different curves we obtained from each individual MQ sensor. This research could potentially lead to low cost sensors being widely implemented in a wide variety of fields and studies to better improve our knowledge of methane producers.