Search

Natural hazards can cause multiple and simultaneous releases of Hazardous materials over extended areas, damage or destroy safety barriers and systems, and down lifelines often needed for accident prevention and mitigation. These are also the ingredients for cascading disasters. For this reason, successfully controlling a Natech accident has often turned out to be a major challenge, if not impossible, where no prior preparedness planning had taken place.

In recent years, severe natural events raised the concern for the so-called NaTech (natural-technological) accident scenarios: technological accidents caused by the impact of a natural event on an industrial facility or infrastructure. Severe scenarios typical of the process industry, as fires, explosions, toxic releases, and water pollution were reported as the consequence of natural events in industrial areas. The historical analysis of accidental scenarios triggered by lightning shows that the impact of a lightning on an atmospheric storage tank might be the initiating event of a severe accident. The analysis of past accident evidences that several alternative damage mechanisms and accident scenarios may follow lightning impact. Although lightning hazard is well known and is usually considered in the risk analysis of chemical and process plants, well accepted quantitative procedures to assess the contribution of accidents triggered by lightning to industrial risk are still lacking. In particular, the approaches to the assessment of accident scenarios following lightning strike are mostly based on expert judgment. In the present study, a detailed methodology is presented for the assessment of quantified event trees following lightning impact on an atmospheric tank. Different damage mechanisms have been considered in order to assess the frequencies of loss of containment due to lightning strikes. The results were used in a case study to assess the overall risk due to lightning impact scenarios in typical lay-outs of tank farms of oil refineries.

The historical analysis of accidental scenarios triggered by lightning shows that the impact of a lightning on a process plant might be the initiating event of a severe accident. The analysis identifies the most common damages to the equipment as either immediate ignition of flammable atmospheres or structural damage with subsequent release. However, the available information on lighting damage to industrial equipment is fragmented and not very detailed. The aim of the present study is to define a failure model for reference equipment categories following lightning impact. Starting from the analysis of lightning damage, the possibility of equipment damage is assessed on the basis of lightning characteristics. The probability of failure is evaluated as the probability to find, in impacting lightning strikes, the minimum energy necessary to cause the loss of containment of the selected equipment. Probability distribution functions available for lightning characteristics, used together with the lightning capture frequency, allow the development of a failure probability model for reference equipment categories.

No abstract available.

A detailed analysis of past accidents evidenced that lightning strikes on storage or process vessels containing flammable materials may cause severe accidents at refineries, storage sites, processing sites and other industrial facilities where hazardous materials are stored or processed. Although the hazard due to lightning is well known and is usually considered in the safety analysis of chemical and process plants, well accepted quantitative procedures to assess the contribution of accidents triggered by lightning to industrial risk are still lacking. In particular, the approaches to the assessment of lightning strike probability, damage caused by lightning strike and intensity of loss of containment (LOC) caused by lightning are mostly based on expert judgment and not on up-to-date models that take into account the relevant progress made in the last years in the analysis of the lightning phenomenon. In the present paper an innovative procedure for the quantitative assessment of the industrial risk due to LOC of hazardous substances triggered by lightning is introduced. The study was focused on the development of models for lightning impact probability and for damage probability. In particular, a Monte Carlo approach was developed to calculate the probability of lightning impact on the basis of the geometry of the structure and of lightning intensity and frequency data.

There is evidence that the impact of natural hazards on process plants as well as on infrastructures for energy supply may cause severe accidents involving the release of significant quantities of hazardous substances. These so-called “Natech” disasters deserve particular attention as they can result in multiple and simultaneous releases of hazardous materials. A further complication comes from the different consequences that may be related to different types of natural events that result in important differences in the expected damage to process equipment. Currently, there is only limited data available on Natech disasters, their prevalence, their triggering mechanisms, and the vulnerability of the various types of industry exposed. The present contribution is devoted to exploring the characteristics of different categories of Natech disasters, in particular those triggered by floods and earthquakes. Available data on Natech accidents were retrieved from several industrial accident databases in Europe and in the US. The analysis of accident data gives an indication of the expected damage of process equipment caused by the impact of natural events. Information was obtained on the expected loss of containment (LOC) events. The scenarios following the LOC events were related to substance hazard and to the intensity of LOCs. The presence and the role of safety barriers was investigated as well. The results allowed the identification of the critical equipment items with respect to the different Natech events and the definition of reference accident scenarios likely to be triggered by natural events in industrial plants.