Study On Assessment Methodology Of Ship Fire Survivability

Blast, flooding and fire are the three major damage factors of ship survivability. In the current ship survivability assessment, only the damage effects of blast and flooding are considered while damage effects of fire are ignored. However, fire is one of the most important threat to ships in both peacetime and wartime. To promote the ship survivability against fire, in the process of ship design, the fire survivability of the designing ship should be evaluated and judged whether it meets operation demands. Through ship survivability assessment, the defects of the ship design can be identified and corrected and the fire survivability of the designing ship can be promoted. Hence, it is necessary to study the assessment methodology of ship fire survivability.For lack of overall framework and assessment methods to assess ship survivability against fire, based on theories of ship survivability and fire risk, a new concept of ship fire survivability is promoted and the assessment methodology of ship fire survivability is studied in this paper. The main works of this paper are focused on the following aspects:A mission-oriented ship fire survivability assessment system is constructed. From the point of system science, some concepts about ship fire survivability are defined and the objective of ship fire survivability is analyzed. A mission-oriented ship fire survivability objective system is proposed which consists of four objective levels, including overall level, performance level, system level and component level. With probabilistic analysis, the assessment model of ship fire survivability is built. According to fire risk assessment theory, the assessment frame work and flow of ship fire survivability are designed. This assessment system not only follows the basic methodology of current ship survivability assessment, but also is combined with the fire risk assessment theory. The assessment framework and flow which is based on probability theory and aims at protecting ship mission can be applied both in peacetime and wartime.Methods coupling with fire scenario analysis are employed to evaluate compartment fire susceptibility. The current concept of susceptibility is extended to define fire susceptibility which consists of ignition possibility and fire scenario occurrence possibility. A compartment can be ignited by weapon hits or accidents. In the case of fire caused by weapon hits, the probability distribution form of igniting point is studied and the compartment ignition probabilistic model is built, In the case of fire caused by accident, ship compartment categories and corresponding ignition frequencies on unit area are studied and the compartment ignition frequency model is built. In order to analysis compartment fire scenario, three methods are employed. Fault tree is applied to analyze the successful probability of a pivotal event in event tree. Event tree is applied to analyze the occurrence probability of a fire scenario. For the progress of fire is described by a time sequence of events in event tree analysis, this method is disabled when the occurrence sequence of some events cannot be determined. Hence, a new method named Bayesian Network is introduced to analysis fire scenario.A method based on fire dynamic simulation is proposed to analyze equipment component vulnerability to fire. Firstly, fire damage indicators and their corresponding criteria are established. Secondly, zone model is chosen as the fire dynamic simulation tool. Thirdly, Monte Carlo simulation coupling with zone model is used to analysis component vulnerability to fire. In wartime, the weapon-induced opening of a compartment is uncertain which is worth researching for naval ship fire. Therefore, opening parameters about position and size are introduced. Based on the method mentioned before, the effect of the two uncertain opening parameters on component vulnerability to fire is analyzed in a hypothetic aircraft carrier hangar bay. In peacetime, the method is applied to analyze the component vulnerability to fire caused by accidents in a hypothetic machinery space. In this case, the effect of uncertain parameters about fire source and fire-fighting measure is discussed.A method to evaluate time-dependent fire survivability of multi-state system is proposed. Firstly, the fire damage process is treated as a time-continuous, state-discrete stochastic process and a theory of time-dependent multi-state damage is established. Secondly, the theory is applied to analyze the time-dependent multi-state component vulnerability to fire. Thirdly, universal generating function is employed to calculate system vulnerability based on its components’vulnerabilities. Fourthly, a time-dependent tree is introduced to analyze fire scenario occurrence probability with time. Finally, combining the above methods, a time-dependent fire survivability model of multi-state system is established based on risk assessment theory. The model can be applied both in wartime and peacetime and can be used to predict time-dependent fire risk of ship systems which is useful to damage control decision-making.