Mid-IR UAV-based sensing platform with deep learning to Identify and Quantify Gaseous Emission in Gas Flares

Oil and natural gas contribute 30% of the total methane (CH4) emission in the United States, CH4 is a powerful greenhouse gas that traps 87 times more radiation than carbon dioxide (CO2). Methane is often intentionally released from oil and gas wells through venting or flaring natural gas. Methane emissions occur during all phases of drilling and production, and sometimes after a well has ceased production, improperly plugged wells and incomplete combustion can emit large volumes of methane and other hazardous air pollution. The mitigation of emissions of methane and other gases through the accurate monitoring/measurement of gas flares can play a significant role in setting appropriate and protective limits to protect health and meet U.S. “net-zero” carbon economy goals. In this proposal Omega Optics together with The University of Texas at Austin proposes the development of a portable lab-on-chip Mid-IR spectrometer system and integrates it on a small mobile platform such as unmanned air vehicles (UAVs) and other airborne systems for routine measurement of non-combusted methane (CH4), carbon monoxide (CO), and nitrous oxide (N2O) released to the atmosphere in the gas flare. Moreover, our proposed spectrometer mounted on UAVs will be developed to work in autonomous mode using an adaptive sampling approach to determine the safe location away from the gas flare with the highest emission gradient. In addition, artificial intelligence/machine learning (AI/ML) functions will be facilitated using the butterfly-style photonic-electronic-neural-chip (BPNC) to analyze the sensing dataset collected. Our proposed technical approach for investigating non-combustion gas emission is based on lab-on-chip absorption spectroscopy in the mid-IR regime. Owing to the unique molecular vibration signature and large absorption cross-section of the compounds and gases in the mid-IR wavelengths, spectroscopy in this regime is considered a promising technique in sensing applications. During the phase-I part of the program, our target will be broadly focused on two major objectives. In the first objective, a feasibility demonstration of the on-chip absorption spectrometer on a silicon-on-sapphire (SoS) platform will be performed with targeted detection sensitivity <50 ppb. The second objective will be to implement an adaptive sampling approach for automated UAVs motion planning to accumulate dense data sampling from the gas flare. The integrated outcomes of these two objectives will pave the way to implement a highly efficient in-situ gas detection platform important for numerous applications, especially where human intervention is problematic. As per the global gas sensor market size & share report, in 2020, the global gas sensor market size was estimated at around USD 2.33-billion, and it is expected to reach USD 4.54 billion by 2028 with a compound annual growth rate of 8.7% from 2020 to 2028. As per the future requirement of the gas sensing applications, the proposed sensing platform has distinct benefits over the existing sensing system like high compactness, high sensitivity & specificity, real-time autonomous detection, AI/ML enhanced prediction and analysis capabilities in a cost-effective platform will be beneficial in every sector where chemical and biosensing is required.