Search

Liquid storage tanks play a vital role in the modern chemical process industry (CPI). The strong ground motion caused by large-scale earthquakes may easily impose severe structural damage on liquid storage tanks, leading to a series of catastrophic cascaded events. The seismic damage estimation of liquid storage tanks is a challenging problem, as the fluid-structure interaction exhibits extremely complicated and non-stationary response behavior. This study develops a novel data-driven methodology to estimate the seismic damage state probability distribution of liquid storage tanks in the contexts of label ambiguity and data imbalance. With the support of the advanced deep learning framework, synthetic oversampling methods, and label enhancement techniques, a hybrid deep belief network-based label distribution learning system (HDBN-LDLS) is proposed for probability distribution learning. The proposed HDBN-LDLS is evaluated on the widely used ALA database. Simulation results indicate that HDBN-LDLS can achieve a balanced estimation for all damage states while maintaining sufficient robustness to cope with label ambiguity. The reliability of the obtained data-driven model is validated by a damaged tank in the 2006 Silakhor earthquake. For practical applications, a more natural way to estimate a seismic damaged tank is to assign a membership degree to each possible damage state. The proposed methodology can quickly obtain the seismic damage state probability curves of a specific liquid storage tank, which can be used to support quantitative risk assessment and seismic design. [ABSTRACT FROM AUTHOR]

• Agent-based modeling methodology for Natech event in a large-scale scenario is proposed. • Time- dependent state transition mechanism of multi-agent is analyzed. • The temporal pattern of the evolution is discussed and the effects of the Natech hazard and all types of safety barrier are quantitative analyzed. In chemical clusters, technological accidents triggered by natural events (Natech events) can lead to the failure of large number of units and cause catastrophic consequences. Previous studies on Natech events give primary concerns on the static analysis at small-scale region with few thoughts the impact of various safety barriers. From the perspective of the large-scale region and temporal characteristics, an agent-based modeling methodology is proposed to present the dynamic evolution process. All the units associated with the evolution are divided into natural hazard agent, hazard-affected facility agent, safety barrier agent, and escalation factor agent to reduce the computational complexity. The time-dependent state transition mechanism of each agent is analyzed to represent the evolution process, simultaneously, the temporal effect of natural hazards, safety barrier and multi escalation factors is considered by utilization of time steps. Moreover, Monte Carlo simulation is adopted to quantify the uncertainties accompanying an accident evolution. The proposed methodology is demonstrated on a complex chemical system in the large-scale region. The results of simulation show that the number of failed hazard-affected facilities in the Natech events scenario will cyclically increase and keep constant, with strong temporal characteristic. The intensity of natural hazards and the probability of failure of safety barriers make a big impact on the temporal characteristic. The results are valuable to support natural hazard prevention in chemical cluster. [ABSTRACT FROM AUTHOR]