Gas Flow Rate Estimation with Artificial Intelligence: Bridging Reality Through Computer Vision and Machine Learning

Abstract Flaring in the oil and gas industry is a critical process where excess gases are burned off through a flare stack. This practice is essential for safety reasons, such as pressure relief during unplanned overpressuring of plant equipment, and for managing gases that cannot be processed economically. However, flaring is also a significant source of greenhouse gas emissions, releasing harmful gases such as carbon dioxide and methane into the atmosphere. The environmental impact of these emissions makes it imperative to monitor and control flaring activities effectively. Despite the necessity of monitoring, the traditional methods involving flowmeters present significant challenges. These devices, which measure the rate of flow of the gas being flared, are often prohibitively expensive and complex to install, especially in remote or offshore locations. This high cost and complexity can hinder comprehensive monitoring efforts, leaving a gap in effective environmental management practices. Moreover, the practice among oil and gas operators to rotate available flowmeters across different flare stacks further complicates consistent monitoring. This rotation often results in minimal monitoring—sometimes only sufficient to meet the bare minimum of legal reporting requirements. Such practices underscore the need for more robust and continuous monitoring solutions. To address these challenges, in this paper we explore an innovative approach to estimate flaring emissions using a more accessible and cost-effective technology. By leveraging a simple system composed of a camera and an edge computer, this method uses visual data and advanced computing techniques to estimate the volume of gas flared. This approach not only reduces the economic burden associated with traditional flowmeters but also enhances the feasibility of continuous monitoring across various operational settings in the oil and gas domain. Through this paper, we aim to demonstrate the effectiveness of this system and discuss its potential implications for environmental monitoring and regulatory compliance in the industry.