Introduction and objectives

The seismic risk of industrial facilities is significantly higher compared to ordinary structures, since the damage of an industrial component may lead to a series of failures and trigger the release of hazardous chemical substances. Typical examples are the extensive and severe damages on industrial facilities due to the Northridge (1994), Kobe (1995), Kocaeli (1999), Chi-Chi (1999) earthquakes and, most recently, the great Japan earthquake (2011). The latter, apart from damages in nuclear power plants, resulted in significant damages in major Japanese refineries (e.g. Ishihara) (Figure 1, Figure 2). Therefore, safeguarding the integrity of industrial facilities against earthquakes is vital for maintaining the unhindered flow of essential products and energy resources, and also for preventing serious consequences on the economy, the population and the environment. 

To ensure an appropriate level of safety, commensurable to the importance of industrial facilities, their seismic design requires the use of state-of-the-art Performance-Based Earthquake Engineering (PBEE) methods that incorporate all possible sources of uncertainty (Hoffmeister & Sedlacek 2003). Although the PBEE concept has been adopted in recent codes and guidelines (ATC-40, ASCE-41) for common civil engineering structures (e.g. buildings, bridges), its application to industrial equipment structures (i.e. liquid storage tanks, pressure vessels, piping systems) is an open issue. So far, structural design of industrial structures under seismic loading follows the American Standards (API 650, ASME B&PV VIII, ASME B31), or the European norms (EN 14015, 13445, 13480) which are based on the “allowable-stress design” concept. Therefore, these standards do not allow for inelastic structural behaviour, neither consider uncertainties and are thus lagging far behind contemporary seismic design philosophy, as appears in modern seismic design standards (i.e. EN 1998). Past research efforts, which can be considered as initial attempts to determine the seismic risk of industrial equipment structures (mostly atmospheric liquid storage tanks), led to the recent EN 1998-4 provisions for seismic design of “silos, tanks and pipelines”. However, these provisions are mainly of informative nature, and are far from constituting a general performance-based framework for such special structural systems. In the framework of European research projects on this subject, i.e LESSLOSS project (Calvi & Pinho 2004), seismic loss assessment methodologies have been developed and applied on several structures, mainly buildings and bridges, and only one atmospheric tank. Most recently, the INDUSE project (Pappa et al. 2011, Varelis et al. 2011, Diamanti et al. 2011) evaluated the current seismic design practice of industrial structures, through a series of experimental testing on piping components.

The main objective of the RASOR project is to develop a systematic PBEE methodology for the seismic risk mitigation of industrial equipment structures, focused on the most important structures of a typical industrial facility: liquid storage tanks, industrial pressure vessels, industrial piping systems and their supports. This objective will be achieved through a multi-disciplinary approach which combines Civil and Mechanical Engineering with Earthquake Engineering, Engineering Seismology, Computational and Stochastic Mechanics, in a effort to produce an integrated seismic risk analysis framework, tailored to the specific characteristics and particularities of industrial installation structures, such as their geometry, high pressure and temperature, operational requirements, material aging and corrosive effects. Of particular importance are the increased safety requirements due the explosive or toxic content of such facilities. Other critical parameters to be considered are the following: (i) The presence of contained liquid with free surface (in tanks and vessels), which causes liquid-structure interaction and generates additional hydrodynamic loads. (ii) The presence of high pressure and the ensuing danger of fracture and loss of content, mainly in pressure vessels and piping.(iii) The use of steel plates of relatively small thickness, which may lead to local buckling or “wrinkling”. (iv) The sensitivity to initial imperfections that may reduce significantly the capacity of liquid storage tanks against elephant’s foot or sloshing buckling. (v) The large variation of pressure and temperature operational loads in pressure vessels and piping components may lead to accumulation of plastic strains, material degradation or ratcheting, and eventually failure due to fracture, when combined with strong cyclic loading due to earthquake. (vi) The consideration of the surface waves of the ground motion, particularly when they have similar frequency to industrial equipment structures with large eigenperiods compared to civil engineering structures, is very important. (viii) The effect of asynchronous ground motion on structures with large footprints (long piping systems or very broad tanks).