CFD-assisted safety design in a flue gas treatment plant retrofit

A retrofit project related to an Al2O3 calcination flue gas treatment line, consisting in the installation of a fabric filter downstream an existing electrostatic precipitator (ESP) was the object of this study. In this case, the replacement of the existing exhaust fan with a more performing one would have been potentially exposed the existing precipitator to suction conditions that did not fall within the process conditions considered when the ESP was designed. In other words, performing a detailed safety analysis focused on the interface between new and existing plants it was highlighted a possible risk of implosion that would not be easily found by following the most common classical approaches as the historical analysis or an HAZOP limited to the new equipments. This hazard would not have been easily reduced by installing reinforcements due to the difficulty (both technical and economical) to modify such existing equipment. The chosen approach was to face the problem analogously to the protection of a vessel from overpressure, that means to size a direct action PSV (counterweight actuated), which opens when a given differential pressure (with opposite sign with respect to the traditional PSV design) is reached. The risk thus determined was translated in terms of a worst case scenario, on the basis of which the design of safeguards has been carried out. In particular, this scenario was identified considering that all the pressure drops are concentrated upstream the equipment to protect, exposing the filter to maximum suction levels. The presence of a fabric filter bypass duct, allowing direct communication between fan and ESP, the complete obstruction of a duct upstream of it and the fan running at full rotational speed were the main hypotheses constituting the worst case scenario. Design choices were verified and validated through a computational fluid-dynamics analysis, evidencing that CFD can be a powerful and useful tool to address safety design issues, allowing to numerically test safeguards actions and consequently evaluating the impact that design choices would have for the purposes of risk reduction and mitigation.