On the safety of LNG-fuelled ships

Rights statement
Awarding institution
  • University of Strathclyde
Date of award
  • 2018
Thesis identifier
  • T14915
Person Identifier (Local)
  • 201560052
Qualification Level
Qualification Name
Department, School or Faculty
  • With an increasing pressure to use cleaner fuels in shipping, LNG has become a realistic marine fuel of choice by many ship-owners. Although it will undoubtedly contribute towards cleaner shipping, there is a general concern about the safety of its use. Such fears are based on the fact that an accidental release of this flammable mixture may lead to potential incidents, in particular, of fire and explosion. In this context, ensuring the safety is one of the most crucial tasks for the industry in adopting LNG as a marine fuel.;It is hardly surprising, therefore, that several rules, regulations, standards and guidelines have been produced at both international and local level. Nevertheless, due largely to the brevity of their history, the existing regulations and class rules appear to have some limitations and shortcomings, in particular, they lack explicit requirements in quantitative terms.;It is true that there are rules and guidance for design and operation of LNG process systems in chemical industry with a relatively long history, and they can to some extent be used to safeguard the use of LNG as a fuel, but the arrival of LNG-fuelled ships has made it essential and urgent that their safety should be investigated in detail.;In an effort to examine the existing regulations more intimately a conceptual design exercise to retrofit a 300,000 DWT very large ore carrier to use LNG as one of the marine fuels was carried out in accordance with current regulations. Although it was felt that the most hazardous areas of such a ship are with the fuel preparation room with high-pressure fuel supply system and the bunkering systems, the current regulations were not of much help during this process.;In order to obtain meaningful insights into their threats, the safety of these critical areas were systematically evaluated using quantitative risk assessment models that have been proven to be an effective and efficient tool for evaluating risks in oil/gas process systems.;It became obvious from the outset that an efficient computation tool is essential for the investigation. Consequently, the conventional quantitative risk assessment methods were modified and augmented to best suit the current problem and implemented in a computer program called Integrated Quantitative Risk Assessment (IQRA).;One of the main strengths of IQRA is that it can carry out the whole process of risk analysis from frequency calculation to evaluation of consequence in a seamless process, unlike existing software which require tedious post-processing to obtain the risk values. It has more than proven its worth during the course of this project.;LNG bunkering, at the two weakest ends (on-board LNG receiving point and LNG supplying point) were examined first. Usually such weak points are made relatively safe by imposing a 'safety exclusion zone' where non-essential personnel are not allowed entry during the process in question is in progress. However, there is no realistic regulation or guidance to help determining the extent of the zones. One of the few existing guidelines based on population-independent analysis requires an exclusion zone far too extensive and impossible to implement.;A realistic method of determining the exclusion zones based on statistical quantitative risk assessment was developed and demonstrated in this thesis. An interesting fact discovered through this study was the fact that short duration bunkering at large scale tends to require smaller exclusion zones than smaller scale bunker for longer duration.;The next area to examine was the fuel preparation room with a high-pressure fuel gas supply system. In this study, an event tree analysis was used to identify events that may cause vapour cloud explosion. General frequency data from various sources was used, and fault tree analysis was applied. Consequence was analysed in terms of the impact from explosion using a CFD and an FEA program. It was found that the structure of some sections of the fuel preparation room must be strengthened in addition to the scantling required by the classification rules.;Additionally, the safety of a medium-sized floating regasification unit was investigated by applying a hierarchical system modelling method newly put forward in this thesis. It demonstrated the excellence of the hierarchical system modelling in assessing the overall risk of complex process systems. The results were compared to the recommendations produced by a benchmark HAZOP study and selective quantitative risk assessment method.;This project established and demonstrated the way that the quantitative risk assessment techniques can be used to investigate the safety of a variety of novel marine LNG plants. A series of studies have proven that the framework and methodology developed in this thesis provides structured guidelines to conduct quantitative risk assessment associated with LNG-fuelled ships as well as LNG process systems regarding fire and explosion.;The facility for parametric and sensitivity analyses have also been shown to be an excellent tool in gaining insight into the nature and characteristics of risks that are inherent in oil/gas process systems while improving understanding of contributing factors to risks and methods of mitigating them as well as points to note during quantitative risk assessment. It is thought that such a tool will be a very useful aid in future rule development.;Last but not least, the safety of people and ships in using and/or processing LNG is paramount. If it is agreed that enhancing the current regulations is an urgent task, it is expected that the results of the studies carried out in this thesis can make a significant contribution to decision-making and further regulatory framework for port authorities and rule-makers.
Advisor / supervisor
  • Lee, Byung Suk
Resource Type
Date Created
  • 2018
Former identifier
  • 9912616793402996