Elsevier

Structures

Volume 25, June 2020, Pages 566-577
Structures

An improved method for quantitative risk assessment of unconfined offshore installations subjected to gas explosions

https://doi.org/10.1016/j.istruc.2020.03.019Get rights and content

Abstract

Previous related research has focused on consequences analysis of confined rather than unconfined structures against explosion accidents. This paper introduces an improved method for quantitative risk assessment of unconfined offshore installations subjected to gas explosions. In the present study, a floating, production, storage, and offloading unit (FPSO) is given as an example to present the proposed method. Instead of the most unfavorable scenario, lots of random scenarios are selected by the probabilistic sampling approach. The method for determining the equivalent gas cloud position is illustrated in the conversion between dispersion and explosion scenarios. Maximum and average overpressures are obtained by computational fluid dynamics (CFD) simulation. Besides overpressure exceedance curves, the combination of overpressure and probability method is adopted based on the definition of risk. This work allows finer scenarios’ sampling results and reduces the computational costs.

Introduction

A series of major disasters such as fires and explosions have sounded the alarm for the development of the offshore oil industry worldwide. HSE [1] reported that fire and explosion are identified as major hazards causing serious casualties, property losses and marine pollution among >60 offshore accidents that happened in the past 40 years. Therefore, the safety of offshore installations operating in harsh environments has placed an urgent need for studying the evolution of major disasters and risk assessment of the whole system.

An extreme or accidental event in a structural system is always associated with multiple parameters that are random and probabilistic by nature. Therefore, a huge number of possible scenarios must be relevant to the real event. However, it is unrealistic to simulate all possible scenarios. As such a selection must be made. In the current industry practice, only a single scenario or at most a few scenarios are considered to represent the so-called “most unfavorable” event. Such scenarios obviously cannot represent the physics of reality. Therefore, it is essential to select a set of realistic scenarios that must represent all the possible events but with a limited number which is needed to minimize the computational costs.

A lot of efforts have been put into the prediction and selection of gas explosion scenarios in offshore installations [2], [3], [4], [5], [6]. API RP WSD [7] provides a selection process for screening out high-risk fire and explosion scenarios, but it still relies on expert’s experience and judgment to further refine those scenarios. UKOOA [8] put forward some guiding ideas for representative scenario selection and the calculating method of accident probabilities, which is valuable for putting scenario selection into practice. Paik and Czujko [9] proposed a quantitative procedure that randomly generates fire and explosion scenarios by input several random variables associated with individual probability distribution functions. This method takes advantage of digital technologies instead of relying on human experience and insights. Based on the probability statistical theory and stochastic sampling modeling technology, different variables are considered in this paper, which reduces computational costs and gets accurate results at the same time.

A framework for the quantitative risk assessment of explosion accidents requires both the probabilities and consequences evaluation. While considerable attention has been paid to consequences analysis [10], [11], [12], [13], literature combing the probabilities and consequences have emerged relatively slowly and in a more scattered way [14], [15]. Moreover, consequence simulations are mostly done with the help of commercial software [16], [17]. Software users need to rely on their engineering experience to accurately input every parameter and adjust these parameters to achieve the best effect, which makes the calculating results inevitably subjective. Quantitative assessment based on numerical simulation tools and field trials, such as Computational Fluid Dynamics (CFD), rather than qualitative assessment relying on engineering experience, has begun to be more extensively applied in offshore designs [18], [19], [20].

A gas explosion is an explosion resulting from mixing a gas, typically from a gas leak, with air in the presence of an ignition source. Isolated recent efforts have been made to investigate flammable gas leak, dispersion, explosion or risk assessment, but different target structures are used in the above studies [21], [22], [23]. The discontinuity of the whole accident simulation reduces the accuracy of the assessment results to some extent. Explosion risks are always related to three critical conditions, which are confinement, congestion, and ventilation [15]. Extensive literature [24], [25] has focused on the quantitative risk assessment of specific facilities subjected to gas explosions, but comparatively little research cared about the whole offshore installations [26], [27]. Various conditions in the surrounding environment, such as the distribution of facilities, significantly affect gas dispersion characteristics and subsequent consequences of explosions. Moreover, much more severe consequences may happen due to a larger volume of gas formed in the large space. It is necessary to consider the interaction of different parts of offshore installations and different stages of gas explosions when doing the quantitative risk assessment.

In this paper, an authentic FPSO is used as an example to present an improved method for quantitative risk assessment of unconfined offshore installations subjected to gas explosions. Eight affecting parameters are sampled to define the gas dispersion scenarios. Leak facility is first determined and then the environmental conditions. Sets of gas dispersion scenarios are obtained by the probabilistic sampling approach. The method of determining gas cloud position is given in the conversion of dispersion and explosion scenarios. The maximum and average overpressures are then obtained in CFD simulation. The overpressure-probability combination method is proposed based on the concept of risk and verified by overpressure exceedance curves.

Section snippets

General procedure for quantitative risk assessment of flammable gas explosions

Fig. 1 presents procedures for quantitative risk assessment (QRA) of flammable gas explosions. The procedure of QRA can be classified into 3 phases.

  • (1)

    Phase I: identification of scenario

Selecting explosion scenarios starts with hazard identification based on a large database and provides important insights that help to define the parameters that affect explosions and to characterize the probability density functions of the influential parameters. A few unfavorable scenarios do not represent the

Target structure

It is widely recognized that FPSO is a popular choice for oilfield development due to its huge advantage in low cost, wide suitability and huge oil storage, and discharge ability. Comparing with other forms of oil production platforms, FPSO shows various characteristics like high input, high risk, and high profit. Therefore, risk and reliability evaluation research of FPSO is very essential.

In this study, an authentic topside module of FPSO is selected as a target structure for an applied

Selection of scenarios

In gas explosions of structures and infrastructures, the characteristics of gas dispersion can be normally identified as a function of eight parameters, namely wind direction (X1), wind speed (X2), leak rate (X3), leak duration (X4), leak direction (X5), leak position X (X6), leak position Y (X7), leak position Z (X8). Wind direction, wind speed, leak rate can be collated by the historical database, while leak direction and leak position are difficult to determine because of their randomness.

Probability analysis

An explosion event occurs after a period of the gas leak when it reaches the explosive limit. Gas explosions cannot occur without ignition even in the event of flammable gas or oil leak. The frequency of gas explosions can be calculated as follows [9].Fexplosion=Fleak×Pignitionwhere Fleak represents the leak frequency and Pignition represents the ignition probability.

The leak frequency can be determined by Table 1. The ignition probability is positively correlated with the leak rate [30]. The

Risk evaluation

Book [33] gives methods for the determination of possible damage to people and assets resulting from releases of hazardous materials. The acceptable criterion for the overpressure of the personnel and structure can be summarized as Table 4, Table 5.

According to the damage degree caused by different overpressures on personnel and assets, probabilities of casualties and structural damage in four regions are presented in Table 6, Table 7. In general, the probabilities of death directly caused by

Conclusions

An improved method for quantitative risk assessment of unconfined offshore installations subjected to gas explosions is proposed in this paper.

  • (1)

    Based on the stochastic sampling method, different affecting parameters are considered in this paper. Leak facilities are determined firstly, then leak position can be obtained by combing the leak facility and leak direction. This method not only ensures that the leak points are located on the corresponding facilities but also reduces calculation costs.

  • (2)

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

One of the authors (Y. Xu) expresses her gratitude to the joint Ph.D. program (CSC NO.201806320085) of the China Scholarship Council for finical support.

References (33)

  • Veritas DN. Accident statistics for floating offshore units on the UK Continental Shelf 1980-2005. Health and Safety...
  • Spouge J. A Guide to Quantitative Risk Assessment for Offshore Installations. DNV Tech 1999. https://doi.org/ISBN I...
  • J.K. Paik et al.

    Assessment of hydrocarbon explosion and fire risks in offshore installations: Recent advances and future trends

    IES J Part A Civ Struct Eng

    (2011)
  • C. Azzi et al.

    Influence of scenario choices when performing CFD simulations for explosion risk analyses: Focus on dispersion

    J Loss Prev Process Ind

    (2016)
  • S.J. Kim et al.

    Methods for determining the optimal arrangement of water deluge systems on offshore installations

    Ocean Eng

    (2016)
  • S.J. Kim et al.

    A New Method for Structural Assessment of Topside Structure Subjected to Hydrocarbon Explosions

    Procedia Eng.

    (2017)
  • A.P.I. Recommended

    Practice for Planning, Designing and Constructing Fixed Offshore Platforms — Working

    Stress Design. Api Recomm Pract

    (2007)
  • Fire and explosion guidance. UKOOA/HSE...
  • J.K. Paik et al.

    Quantitative assessment of hydrocarbon explosion and fire risks in offshore installations

    Mar Struct

    (2011)
  • J.M. Sohn et al.

    Nonlinear structural consequence analysis of FPSO topside blast walls

    Ocean Eng

    (2013)
  • O.R. Hansen et al.

    Estimation of explosion loading on small and medium sized equipment from CFD simulations

    J Loss Prev Process Ind

    (2016)
  • K.Y. Kang et al.

    Dynamic response of structural models according to characteristics of gas explosion on topside platform

    Ocean Eng

    (2016)
  • Z.I. Syed et al.

    Non-linear Finite Element Analysis of Offshore Stainless Steel Blast Wall under High Impulsive Pressure Loads

    Procedia Eng.

    (2016)
  • Y. Jin et al.

    Probabilistic fire risk analysis and structural safety assessment of FPSO topside module

    Ocean Eng

    (2015)
  • Y. Huang et al.

    Multi-level explosion risk analysis (MLERA) for accidental gas explosion events in super-large FLNG facilities

    J Loss Prev Process Ind

    (2017)
  • G.H. Chen et al.

    Quantitative risk assessment of liquefied chlorine leakage accident via SAFETI

    Huanan Ligong Daxue Xuebao/Journal South China Univ Technol (Natural Sci

    (2006)
  • Cited by (0)

    View full text