极地水域船舶冰困风险演化机理与航行决策方法研究

极地水域海冰复杂多变，如何有效避免冰困是保障极地水域船舶航行安全的重要技术难题。目前，国内外对于冰困风险的研究刚刚起步，对船舶冰困风险的评价预测和相应的航行决策还停留在经验判断阶段。不确定性问题是极地水域船舶冰困风险研究的核心技术难题，主要体现为系统认知的不确定性、相关数据缺失的不确定性、测量误差的不确定性以及风险演化趋势的不确定性。本研究将针对上述问题，系统分析识别极地水域船舶冰困风险的影响因素，并通过结构方程模型阐明各影响因素间的动态响应关系；运用贝叶斯网络等不确定性分析方法，实现极地水域船舶冰困风险与风险影响因素之间耦合关系的参数表征，探索极地水域船舶冰困风险演化机理；建立极地水域船舶冰困风险的评价指标体系和衡准，运用证据推理和蒙特卡洛仿真进行风险控制方案的多目标分析，研发极地水域船舶航行的决策支持系统。

Due to the complex sea ice condition in the polar water, how to prevent and mitigate the risk of ship getting stuck in ice is a significant issue to navigational safety. However, there have been few studies on ship stuck risk in home and abroad, and the corresponding risk assessment and decision making techniques remain in preliminary stages. Moreover, uncertainties that induced by one-sided recognition, lack of data, measurement error or unpredictable tendency make it more challenging to estimate ship stuck risk in polar waters. According to the above problems, first of all, this study will identify the safety critical factors (SCFs) in terms of ship stuck risk and illustrate the interrelationships among SCFs through structure equation model. Then, evolution mechanism of ship stuck risk will be conducted using Bayesian network model. Finally, based on the proposed ship stuck risk criteria, this study will conduct multi-objective analysis on risk control options in combination with evidence reasoning and Monte Carlo simulation so as to develop a navigational risk management software for polar waters.