Analyzing the Applicability of ML Powered Microwave Sensor for UAV Based CH4 Sensing

Abstract

In solid waste management, the Methane (CH4) gas generation needs to be monitored and assessed for the environmental protection and prevent the CH4 gas atmospheric dispersion over the dwelling regions near dump yards. Traditional methods using gas sensors measures only gas concentration at different height and never modelled or predict the gas circulation in the atmosphere. Moreover, gas circulation near every dump yard needs to be modelled based on microclimatic conditions for human safety and environmental protection. In this paper an Unmanned Aerial Vehicle (UAV) based, dispersion of CH4 gas is monitored and assessed for the atmospheric circulation of CH4. The CH4 dispersion and atmospheric circulation are monitored through X-Band Bi-Static microwave transceiver sensor and Machine Learning (ML) Algorithm-Multi Linear Regression (MLR). CH4 concentration is assessed through the RADAR’s reflected signals from CH4 gas, whereas RADAR is fixed in the UAV. The RADAR is mounted in UAV to sense concentration of CH4 dispersion at different heights for municipal dump yard located at Thiruninravur, Tamil Nadu, India. A three-days; six-hours of field results, shows the proposed UAV-RADAR based prediction of the CH4, performs better in the range of 250-800 ppm CH4 concentration. The proposed UAV-RADAR system in tracing CH4 dispersion in lower atmosphere is validated through analytical methods and modelled based on microclimatic conditions.