| There are a large number of petrochemical enterprises in the chemical industrial park,with a wide variety of inflammable,explosive and toxic hazardous substances.Under the influence of all kinds of natural disasters,it is easy to trigger industrial accidents such as fires and explosions.For such natural hazard triggering technological disasters,they are internationally defined as Natech events.Natural disasters and their secondary fire and explosion accidents may damage the dangerous equipment units in the park,resulting in domino effects and enlarging the consequences of accidents.Domino effects in chemical industrial parks caused by natural disasters happen from time to time and show a growing trend,indicating the deficiencies of chemical industry parks in dealing with the risk prevention and control caused by natural disasters.Therefore,the study of propagation probabilities and risk prevention &control of domino effects triggered by natural disasters is of great theoretical and engineering value for improving the level of safety management and accident prevention & control in chemical industrial parks.At present,the research on Natech events mainly considers the fire and explosion accidents which are derived from natural disasters,and the subsequent possible domino effects are seldom considered.However,most of the researches on domino effects only consider the primary accidents caused by equipment or management defects within the chemical industrial park,and seldom consider the simultaneous occurrence of multiple primary accidents triggered by natural disasters,and often ignore the possibility of multi-level propagation of domino effects.To this end,this paper takes the domino effects in chemical industrial parks triggered by typical natural disasters such as lightning,flood and seismic as the research object,and the risk prevention &control in chemical industrial parks as the purpose to carry out the following researches:(1)Study on the evolution mechanism of domino effects in chemical industrial parks triggered by natural disasters.The characteristics and mechanism of domino effect caused by local and regional natural disasters are analysised,and the occurrence conditions and structural characteristics of accidental chains of domino effects triggered by natural disasters are discussed based on the disaster chain theory and the thought of system science.The composition structure and the common law of evolution of the accidental chains are expressed mathematically,and the evolution analysis model of the accidental chains of domino effects induced by natural disasters was built to reveal the complexity law and motion nature of the evolution process of the accidental chain.The three-level quantitative analysis framework of the evolution process of domino effects triggered by natural disasters accident chain is constructed to provide theoretical basis for the subsequent chapters.(2)Study on the probabilistic prediction method of domino effects triggered by local natural disasters.According to the characteristics of domino effects in chemical industrial parks triggered by local natural disasters represented by lightning,the probabilistic prediction process of multi-level domino effects triggered by lightning is proposed for the first time.The causes and scenarios of tank fires triggered by lightning are analyzed by using event tree,and the probability model of tank fire triggered by lightning is proposed.Considering the synergistic effects and the multi-level domino effects among the domino escalation factors,the propagation chain diagram of domino effects triggered by lightning is constructed.Tthe Bayesian network is applied to calculate the probability of each accident chain,and the most dangerous primary equipment is identified by comparison with the probabilities of the domino effect at different levels.By setting up the failure states of different tanks,the probabilities of events are updated under a given situation,and the most susceptible target equipment with respect to the domino effect are identified,which provides data support for quantitative risk assessment and prevention & control in Chapter 6.(3)Study on chemical equipment vulnerability assessment model under the effect of regional natural disasters.The limit state equations which can reflect the relationship between flood intensity parameters and the resistance parameters of atmospheric vertical tank is established.The databases of flood intensity parameters and tank characteristic parameters are respectively established by using Latin hypercube random sampling method.By means of logistic regression method,a vulnerability assessment model which can consider three failure modes of sliding,floating and buckling simultaneously and is applicable to the flood with different intensity parameters and the tanks with different characteristic parameters is constructed to overcome the limitations of previous studies which are limited to specific tanks.On the basis of vulnerability assessment model,the influences of flood submergence depth,flood flow velocity,tank diameter,tank height and filling level on vulnerability of the tank are analyzed by fragility curves.The fragility magic cube that can simultaneously represents the influence of flood depth,flood velocity and filling level is constructed to analyze the critical disaster causing conditions and critical filling levels of the tank.Finally,the seismic vulnerability assessment models of chemical equipment and the flood vulnerability assessment models of horizontal storage tanks which can be used for quantitative risk assessment are summarized,which provides data support for accidental chain prediction in Chapter 5 and risk prevention & control in Chapter 6.(4)Study on the probabilistic prediction method of domino effects triggered by regional natural disasters.According to the characteristics of regional natural disasters,such as floods and earthquakes,which may cause simultaneous failure of multiple equipment units,the failure state combination and probability evaluation model of all possible equipment caused by regional natural disasters are established.All possible failure state combinations and probability assessment models for equipment units damaged by regional natural disasters are established,and a probabilistic prediction method for the domino effects with multiple levels and multiple primary units triggered by natural disasters is proposed.The proposed method establishes the directed graph of propagation chains of domino effects triggered by regional natural disasters based on the failure state transition matrix and state transition probability matrix,and overcome the shortcoming of the present work on domino effects triggered by natural disasters.Finally,the most likely primary accidental scenario,the most likely extended link and the equipment unit with the greatest impact on domino effects are identified through a case study,and the effects of natural disaster and multi-level domino effect on failure probability of equipment are analyzed,which provides data support for quantitative risk assessment and prevention & control in Chapter 6.(5)Study on quantitative risk assessment and prevention & control system of domino effects triggered by natural disasters.On the basis of the probabilistic prediction of domino effects triggered by natural disasters,the individual risk and social risk assessment model for domino effects triggered by local natural disasters and regional natural disasters are proposed,and overcome the limitation that the present risk assessment method without considering the impact of natural disasters and domino effects simultaneously.According to the rules of propagation accidental chians,the model of chain-cutting disaster mitigation is proposed,and the full life cycle prevention & control system is studied considering the influence of natural disasters and multi-level domino effects simultaneously.The individual risks and social risks of natural disasters and multi-level domino effects are compared and analyzed through the case study,to reveal the expanded role of natural disasters and domino effects on the risks of chemical industrial parks,and to quantify the effects of accident prevention & control with the proposed risk assessment models,so as to provide theoretical basis for optimizing prevention& control strategies. |