Detection and Estimation of Ethanol Concentration in a Disturbed Environment Utilizing Random Forest

Gas detection and estimation play key roles in ensuring environmental safety, industrial efficiency, and public health. As industries continue to evolve, the need for accurate and reliable methods to detect and quantify gas concentrations becomes increasingly critical. This study investigates the detection and estimation of Ethanol (C<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</inf>H<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</inf>O) or (EtOH) concentration in a disturbed environment using Random Forest (RF). The experimental setup involves a system with precise control over the dilution of polluting gases, including EtOH, Acetone (C<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf>H<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</inf>O), and Carbon Monoxide (CO). The Metal Oxide (MOX) sensor, situated in a 3D-printed test chamber within a Faraday cage, records data encompassing sensor responses, heater states, and gas concentrations. The proposed approach leverages the robustness of RF for both classification and regression tasks, addressing uncertainties and disturbances inherent in the environment. The results demonstrate that the RF model achieves a precision of 94%, indicating its efficacy in accurately detecting and estimating EtOH concentration amidst various environmental disturbances.