Game-planning Emergencies: Modeling Technique Simulates Effects of Explosions, Chemical Releases

Published Sept. 26, 2018

By Ed Rutkowski

For industrial operations, emergency preparedness often involves imagining worst-case scenarios in intricate detail. For example, what if a plant starts leaking a dangerous chemical? Where is the vapor cloud likely to travel given prevailing wind conditions? In the crucial first moments following an accidental release, what should be done to protect workers and surrounding communities?

To help answer these questions, some industrial operations are turning to a modeling technique called Computational Fluid Dynamics, or CFD. On Sept. 24, at IOHA’s 11th International Scientific Conference in Washington, D.C., presenters from an industrial hygiene consulting firm introduced attendees to CFD and demonstrated some of its applications.

A CFD model is a “mesh” of calculations that “wraps around” a virtual rendering of a three-dimensional space, said presenter Cassidy Strode, an industrial hygienist and CFD specialist at the consultancy Chemistry & Industrial Hygiene in Wheatridge, Colo. The calculations are performed on a grid; a single room could be divided into millions of specific locations. One model Strode worked on for a client calculated 9.3 million distinct volumes, he said.

Using CFD requires specialized software programs, some of which can be very expensive. The software can accommodate a multitude of variables. “The more complex the situation, the more data points you need” to develop a useful model, Strode said.

Among the more complicated applications of CFD are modeling the dispersion of contaminants from an explosion. Strode described working on a CFD model for a wastewater treatment plant whose emergency planners needed to consider the potential effects of an accidental release of chlorine gas. Chlorine is an effective disinfectant that many wastewater treatment facilities use to remove impurities, but an uncontrolled release can cause severe health effects, including death by asphyxiation. Through CFD, Strode modeled the client’s worst-case scenario: the explosion of a one-ton chlorine gas cylinder. The plant was located in a valley; the CFD model demonstrated the likely coverage and direction of the deadly plume as it spread across the valley and through a tunnel connecting the plant to a small town. Differences in temperature and wind speed, which could drastically change the plume’s direction, were among the variables emergency planners could take into account when preparing for the unthinkable.

Other uses of CFD are less dramatic but no less practical and have potential applications in several industries. Strode’s co-presenter, Daniel Hall, the head of engineering at Chemistry & Industrial Hygiene, said that an offshore oil platform concerned about production of diesel exhaust could use CFD to model the likely effects of redesigning the platform’s exhaust stacks, while a site that fabricates semiconductors could use CFD to understand the likely effects of an arsine leak.

For all its promise, CFD, like all tools, has limitations. CFD software is exceedingly complex and requires expertise to use correctly; for that reason, the software is not often recommended within the industrial hygiene community, Hall said. He also cautioned that industrial hygienists need to keep in mind that modeling is not an exact representation of the real world. “One of the things we have to remember is that all models are wrong,” Hall said, paraphrasing a well-known aphorism from the field of statistics. “But some of them are useful.”

Ed Rutkowski is editor in chief of The Synergist.