基于动态贝叶斯网的双燃料船舶适航性评估

Seaworthiness Evaluation of Dual-fuel Ship Based on Dynamic Bayesian Network

随着柴油-LNG双燃料船舶新型技术的兴起,迫切需要科学有效的方法对其进行适航性评估与安全监管。本文首先根据双燃料船舶的营运特点,从船舶动力丧失概率的角度分析和确定了适航性评估的内容、程序和建模方法。然后针对船舶使用双燃料和纯柴油两种工况下的失效模式特点,建立了相应不同结构的可靠性框图。最后基于可靠性框图建立了动态贝叶斯网络并计算了各节点的条件概率,解决了双燃料船舶适航性分析中的动态性,小子样、相关失效等难点。实例计算表明双燃料工况下由于共因失效冲击船舶动力丧失概率明显增大,验证了评估模型的正确性。

With the rise of the diesel-LNG dual fuel ship as a new technology, it is imper- ative that using the scientific method into seaworthiness evaluation and safety monitoring. Firstly, according to dual-fuel ship characteristics, seaworthiness evaluation content, proce- dure and its effect factors are analyzed in terms of probability of power loss, seaworthiness e- valuation model is constructed. Then through the analysis of the failure mode of diesel-LNG ship, different Reliability Block Diagram of ship both in diesel-LNG section and in diesel sec- tion are proposed. Finally, the methods of model construction and conditional probabilities de- termining in corresponding Dynamic Bayesian Networks are given. The study solves the prob- lems which includes dynamic, small sample, depended failure of seaworthy analysis in diesel- LNG ship. The computed results indicates the probability of ship power loss increase signifi- cantly in diesel-LNG section due to the common cause failure. The validity and advantages of proposed method is demonstrated by the example.