Multicomponent Modeling in Consequence Analysis

The effects and consequence analysis applied to an industrial process risk assessment aims to predict the possible damages due to an accident by the estimate of vulnerability of an accidental typology. The consequence modeling is useful for the elaboration of emergency plans, to establish the relative safety of a plant control room, to locate flares and vents, to prepare facility sitting studies, to optimize plant layout and to perform quantitative risk analysis. Among the challenges of a consequence study, efforts are made for the development of multicomponent modeling since most process fluids are composed by a mixture of substances. Multicomponent models are harder to develop and usually require more computational resources than the pure component consequence modeling. The intention of this paper is to present and analyze models applied by commercial software available to deal with the properties of multicomponent process streams, as well as their link with each consequence calculation. The availability of thermodynamics equations of state for predicting the post-release fluid physical properties are mainly focused on oil and gas industry application where the benefit of non polar or slightly polar flows is observed. Some simple case studies are developed in order to discuss the comparison between the pseudo-component approach and multicomponent modeling and to analyze the application of equations of state to this stage of a risk analysis. The simulation results deviation may vary according to the process conditions and the models used for the vulnerability prediction. The usage of pseudo-component approach generate results that might under or overpredict the expected vulnerable area. A good agreement is found between the equations of state applied for non polar mixtures.