In dispersion modeling evaluations, flares are typically treated as point sources with generic values. The EPA (SCREEN3, AERSCREEN) and various states (e.g., Texas, North Dakota) have outlined a method for estimating an equivalent stack diameter based on the flare’s heat content and standard assumptions of temperature, gas exit velocity, virtual stack height, and radiation heat losses. However, these assumptions are not realistic for the many types of utility, air-assist, steam-assist, high/low pressure gas assist, ground, and sonic flares used in various industries. Additionally, newer flares minimize radiation and are required to be smokeless by most regulatory agencies.
That means the traditional way of modeling flares is outdated and inadequate for most cases, and the traditional input parameters only predict the plume rise for a single flare. In cases with multiple flares in close proximity to each other, the additional buoyancy (and added plume rise) from the nearby flares is not accounted for.
CPP conducts wind tunnel and Computational Fluid Dynamics (CFD) modeling studies that can provide more accurate, site-specific plume rise effects for single or multiple flares. For fire related sources, such as flares, the FDS (Fire Dynamics Simulator) LES (large eddy simulation) model is the most appropriate and cost effective. The same approach can be used when characterizing hot (buoyant) power plant emissions from adjacent stack plumes that may merge together.