Users of the next release of Phast and Safeti will benefit from significant improvements and increased choice in the modelling of jet fires and the evaluation of associated radiation effects.

Jet flames occur following the ignition and combustion of flammable fluids issuing continuously from a pipe or orifice,
which burn close to the release plane. Releases that fuel jet fires could be accidental or intentional. An example of the latter are flames from flare systems of offshore oil and gas production facilities, which are primarily operated to provide a safe means of disposal of hydrocarbon gas under a variety of process conditions. Jet flames dissipate thermal radiation which, away from the flame’s visible boundaries, transmit heat energy that could be hazardous to life and property. Thus, in the evaluation of the hazard posed by jet flames, the accurate determination of the likelihood of flame impingement and the amount of radiant energy received by objects at a distance from the flame is of primary importance.

Previously, in the evaluation of hazards associated with jet flames, users of Phast and Safeti were limited to the choice of the ‘API’ and the ‘SHELL/JFSH-Cook’ ‘surface-emitter’ models. The comparison of the performance of the above models against recently developed ‘surface-emitter’ models and published field data (i.e., flame geometry and attendant radiation effects), underlined the need to improve on and extend the modelling capabilities and choice available to users.

From the result of work related to addressing the above needs, the next release of Phast and Safeti will benefit from significant improvements and increased choice in the modelling of jet fires and the evaluation of associated radiation effects.

Increased options
In addition to the previously available modelling options, you will be able to select any of the following sub-models under the ‘JFSH’ surface-emitter models:

• Chamberlain (1987) model (JFSHChamb): this model, originally suited to the modelling of jet flames resulting
  from pure-vapour releases, has been extended to provide better estimation of flame characteristics for two-phase and liquid releases.
• Johnson et al. (1994) model (JFSHJohnson): this model is only available for the modelling of horizontal releases.
• The modified Cook et al. (1990) twophase model (JFSH-Cook): this model provides significantly improved simulation of the characteristics of jet flames resulting from two-phase or liquid releases.
• The DNV recommended option: this option provides automated model selection. It picks from the available modelling options, a model that is expected to provide the best performance, in terms of accuracy, for the set of conditions you have specified.

Modelling and graphical improvements
Further features and capabilities that will be available in the ‘JFSH’ modelling suite for the next release of Phast and Safeti include:

• The modelling of cross-wind effects on flame shape and characteristics.
• The modelling of jet flames resulting from negatively inclined release sources.
• Modelling of flame buoyancy for jet flames resulting from horizontal releases [i.e. JFSH-Johnson model].
• Improved modelling of jet flames resulting from two-phase releases.
• Tests on simulated flame geometry to determine and report the likelihood of on-ground flame-impingement and its approximate location.
• Improved graphical plots (see Fig. 1) showing the effect of cross-winds on flame orientation (i.e. flame deflection) and simulated radiation ellipses.

Fig. 1 illustrates the effect of perpendicular cross-winds travelling at speeds of 1.5, 5 and 10m/s respectively on jet flames resulting from a horizontal release. The simulated results are based on the JFSHJohnson model. The orientations and sizes of the simulated radiation ellipses show the effect of cross-winds and cross-wind velocities on the assessment of risks posed by jet flames to their immediate surroundings.

Validation
Part of the model improvement work involved the verification and validation of simulated jet fire characteristics using the JFSH models against published data. The validation covered the simulation of logged data on jet flames resulting from vapour, two-phase and liquid jets.

Additionally, simulated results obtained from the combination of the jet fire and radiation models available in Phast and Safeti (i.e. JFSH-RADS) were compared against field data (see Fig. 2 and 3). From Fig. 2 and 3, the predicted incident radiation over a wide range of observer locations and orientations have been shown to compare well with field data (ca 12% average absolute deviation) and generally lie within ±40% of measurements. Details of the validation process and results obtained will be available in the jet fire theory documentation, which will be included in the Phast/Safeti installation CD.

Conclusion
In the next release of Phast and Safeti, the modelling of cross-wind effects on jet flames will be limited to the standalone jet fire model. However, the above improvements and increased choice provide the user with a valuable tool for the detailed and accurate analysis of risks associated with jet flames.

Author: Dr. Adeyemi Oke