“Special risk” industrial plants can be highly vulnerable when subjected to natural phenomena such as earthquakes, flooding and explosions. In this study we focused our attention on the performance of a liquefied natural gas (LNG) terminal subjected to extreme earthquakes; an LNG terminal consists of a series of process facilities connected by pipelines of various sizes carrying hazardous chemical components. Although tanks, pipes, elbows, bolted flanges have been a major concern in terms of seismic design, generally, they have not been analyzed with modern performance-based procedures. Thus, our work has been devoted to the assessment, within the performance-based earthquake engineering (PBEE) framework, of the seismic performance of tanks, pipes, elbows and bolted flange joints by means of seismic fragility functions. Particular attention was paid to component resistance and to loss of containment (LoC) due to leakage.
A representative case study of an LNG terminal has been selected and tank, support structures and pipework, including elbows and flanges were analyzed with a detailed 3D finite elements model. Preliminary analyses identified elbow and bolted flange joints as critical components. A mechanical model, based on experimental data, defines leakage limit states for bolted flange joints. A significant effort was also devoted to identification of a leakage limit state for piping elbows, and we found the level of the hoop plastic strain to be an indicator.
On a first stage for the probabilistic seismic demand analysis (PSDA) we applied the Cloud method, due to its advantages in terms of consistency in the seismic input and of computational savings. More precisely, we studied the behaviour of critical components using a set of 36 ground motions, selected from a database of historic earthquake accelerations. The results of seismic analysis show that bolted flange joints remain significantly below their leakage threshold while elbows at the top of the LNG tank are likely to show LoC. Fragility functions show that elbows located on the tank platform are relatively unsafe against earthquakes.
On a second stage, in order to detect more complex failure scenarios, we analysed an LNG plant comprising a more complex piping substructure with respect to the original FE model. Moreover, we have treated the LoC of elbows from a probabilistic point of view by means of a Gaussian probability density function associated to a hoop strain limit state. The results provided by the aforementioned refined piping system allow for an improved and more accurate seismic risk assesment of the LNG plant.