Abstract
A natural gas (NG) odorization system requires continuous monitoring as well as an optimal injection to satisfy the odorization guidelines, minimize over-odorization, and prevent hazardous gas leaks. NG consists of hydrocarbons such as methane, odorants such as tert-butyl mercaptan, and other sulphur-based VOCs such as hydrogen sulphide; therefore, selectivity is paramount for the reliable and accurate monitoring of odorants. In this study, we developed a portable device integrated with an array of five different sensors to detect a mixture of tert-butyl mercaptan and methyl ethyl sulphide for a concentration range of 1 ppm to 10 ppm. A machine learning model was developed to predict the presence and concentration of NG odorants from the sensor data. The best-performing sensors in the array achieved high sensitivity and selectivity indicators (measured using the Davies-Bouldin index) of 0.3667 (1⁄ppm) and 0.125, respectively. The sensor system achieved a classification accuracy of 98.75% between NG odorants and hydrogen sulphide, with an overall Mean Squared Error (MSE) and R2 error (for the regression model) of 0.50 and 95.16%. These results indicate that the developed portable device and the machine learning model have promising applications for the selective monitoring of NG odorants.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.