Abstract
Major accidents such as Buncefield, Texas City and, more recently, at the Amuay refinery in Punto Fijo Venezuela, have focused attention on assessing the effects of Vapour Cloud Explosions (VCEs) on assets such as buildings and process equipment, as well as people. Ensuring the safe location and design of occupied buildings is high on the agenda of key stakeholders. In recent years much research has been undertaken to improve the accuracy and understanding of the mathematical modelling of VCEs. This has resulted in the development of simplified empirical models based on Computational Fluid Dynamics (CFD) calculations and the results of experiments. The two most well-known models to emerge from this research are the TNO Multi Energy Method (ME) and the Baker- Strehlow-Tang Model (BST). Since the physical phenomena which drive explosions are complex, the nature of the most accurate predictive models available is also necessarily complex. CFD based models are generally considered to give the best approximations of these phenomena. However, such models are largely considered impractical for day-to-day work due to the significant resource and time requirements needed to build, run and analyse such complex models. In most consequence calculations performed in the context of a QRA (Quantitative Risk Analysis) or worst case analysis, this complexity is unnecessary. Additionally, the resource demands of this approach are such that applying it for a plant wide assessment is generally not feasible. The “simplified” models mentioned above are designed to solve this problem by allowing a reasonable level of characterisation with less geometric and physical complexity, placing emphasis on the key parameters that are known to have a significant influence on the consequences in terms of overpressure and impulse. This paper describes a more detailed implementation of the two most widely used vapour cloud explosion models mentioned above, namely the ME and BST models, in Phast, the well-known general purpose consequence analysis software package developed by DNV GL Software. The method provides a more detailed analysis of VCE threats without recourse to over simplification which can be detrimental to the accuracy of the results obtained from a risk management perspective. Example results from a hypothetical occupied building siting study are presented and the value of using the tool discussed.
