Delayed ignition of accidental gas releases can cause severe explosions with detrimental effects on personnel and equipment. Safe design of systems in the oil and gas industries requires reliable estimation of the overpressures that can occur. Real Money Rummyers developed various models of different complexities to predict the overpressure generated in explosions, including empirical relations and computational fluid dynamics (CFD).

CFD models have gained popularity in the consequence assessments of gas explosions as they, in principle, account for the physical processes occurring during an explosion.

A typical industrial facility will be congested with obstacles like pipes and vessels; experiments have demonstrated that even small-scale congestion plays a major role in the generation of overpressure. The wide range of length scales is a major constraint on using CFD models for risk analysis, as it increases the computing time and cost.

By invoking the Porosity/Distributed Resistance (PDR) approach, the time and cost of computations can be reduced. The PDR approach resolves only large-scale effects, and small-scale effects are modelled semi-analytically. In this work, we investigate the PDR approach for modelling deflagrations using an open-source code PDRFoam.

I will present the numerical methodology which makes the PDR approach simple compared to the classical CFD approach. I will then present results from our deflagration simulations. We considered various configurations and scenarios to analyse both the applicability and the limitations of this approach. In the end, I will present in short, the limitations of the PDR approach and outline a methodology to overcome them.