基于马尔科夫链的恶劣天气船舶行为预测

准确预测恶劣天气时船舶行为是港口水域应急调度和高效安全管理的关键,提出一种基于马尔科夫链的恶劣天气船舶行为预测模型.根据分析船舶行为实际变化规律,得出船舶行为是由船舶在不同空间进行状态转移而形成,然后结合船舶行为分布数据采用经验风险最小化策略求解状态转移矩阵,最后通过初始状态分布和状态转移矩阵预测恶劣天气时船舶行为.实例验证和分析结果表明该预测模型能科学准确地预测恶劣天气时船舶行为,可为港口应急调度与高效管理提供有效参考.

Prediction of Ship Behaviors in Heavy Weather Based on Markov Chain

Predicting the ship behaviors accurately in heavy weather is the essential to emergency dispatch and efficient safety management of port waters.Proposed a prediction method of ship behaviors in heavy weather based on Markov chain.According to the analysis of the actual changes in ship behaviors,it is concluded that the ship behaviors is formed by the status transition of the ship in different spaces.Then the empirical risk minimization strategy is used to solve the state transition matrix based on the ship behaviors distribution data.Finally,the initial status distribution and state transition matrix are used to predict the ship behaviors in heavy weather.The example and analysis results show that the prediction model can scientifically and accurately predict the ship behaviors in heavy weather,and can provide an effective reference for port emergency dispatch and efficient management.