PSI - Issue 44

Gianluca Quinci et al. / Procedia Structural Integrity 44 (2023) 251–258

252

2

Gianluca Quinci et al. / Structural Integrity Procedia 00 (2022) 000–000

1. Introduction Past destructive earthquakes in China (Sichuan, 2008 and Yushu, 2010), Japan (Tohoku, 2011) and Italy (Emilia, 2012) have emphasized social and political attention on seismic risk of process plants. Both observations of damage following major/moderate seismic events (Brunesi et al. (2012)) and numerical/experimental studies (Bursi et al. (2015)) have highlighted that petrochemical facilities are particularly and even disproportionately vulnerable to earthquakes. Besides, several researchers investigated numerous industrial accidents that have resulted in severe loss of life and injuries, damage to natural and built environment as well as significant economic losses. As a matter of facts due to earthquake damage, it emerges that non-structural components (NSCs) and support systems interactions account for the majority of direct property losses (Filiatrault (2005)). In this respect, to collect and document the body of available knowledge related to the seismic performance of NSCs, the National Institute of Standards and Technology (NIST) invested on a year-long study summarized in NIST (2018). From these studies, it clearly emerges: (i) a lack of definition of performance objectives for NSCs in a performance-based design methodologies perspective; (ii) a gap in performing fully comprehensive testing campaigns and investigations on coupling effects between primary structures and NSCs; (iii) a need to improve and enforce code requirements along with the development of reliable NSCs seismic demands models. While for civil-engineering structures probabilities of exceedance of peak ground acceleration (PGA) vs limit states are clearly identified, little information can be found in literature for hazardous industrial equipment. Along this vein, a first attempt in the field of petrochemical plants can be found in Bursi et al. (2015), where typical refinery piping systems were analyzed referring to Operating Basis Earthquakes (OBEs) and Safe Shutdown Earthquakes (SSEs). The former corresponds to a probability of exceedance of 10% in 50 years and is defined as the condition under which plants remain fully functional without undue risk to health and safety of public. Increased return periods can be achieved through additional PGA multipliers. The latter is related to a lower probability of exceedance under which certain relevant structures, systems and important components must be designed to remain operational and allow a safe closure. For the sake of completeness, the OBE has been replaced by the design basis earthquake (DBE) in the last version of ASCE 4-16 (2016); from now on, we’ll use this nomenclature. The importance of well-established relationships between limit states and reached damage in an industrial or nuclear system is clearly demonstrated by the consequences of failures of components in several accidents in industrial plants, as already illustrated in Brunesi (2012). It emerged on one hand, a limited knowledge of the seismic behaviour of both critical units, components and support structures; on the other hand, the lack of information about their interactions. In fact, release of hazardous materials from secondary elements, i.e. pipes, tanks, etc., even not directly considered by current standards, could generate uncontrollable disasters, dramatically exacerbated by domino effects (Necci et al. 2015). Therefore, researchers, (Alessandri et al. 2018), focused on the relation quantification between loss of containment (LoC) and seismic damage, while Paolacci et al. (2021) have investigated the behaviour of a flange joint under a cycle loads; moreover, since current design practice for piping systems contain little information about seismic component details (Vathi et al. 2017), Reza et al. (2014) suggested innovative solutions based on non standard bolted flange joints (BFJs). Consequently, the present paper aims to define limit states and Engineering Demand Parameters (EDP), for the NSCs installed on a primary steel frame structure based on a shaking table test campaign and Finite Element Model (FEM) analysis, Nardin et al. (2022) and Butenweg et al. (2021). Furthermore, a new algorithm for the ground motion selection proposed by some authors of this paper, (Giannini et al 2022) has been used because of two main advantages: (i) it is no longer necessary to refer to a specific , but only requiring spectrum compatibility conditions of the ground motion records with mean and mean plus standard deviation UHS, (ii) allows the selection of pairs of spectrum-compatible natural records, facilitating the seismic assessment of three-dimensional structures without artificial scaling operations. Finally, the fragility analysis of NSC (i.e. vertical tank) for the case study is carried out and the mean annual frequency to exceed the specific limit state is evaluated. Two different sets of accelerograms are herein considered for the risk assessment, in order to validate the advantages of the adopted ground motion selection method, that have already been demonstrated for the risk analysis of primary structures in (Giannini et al. 2022).