基于遗传算法的渡船横渡航法寻优模型

目前,渡船横渡内河水域主要依赖船长的经验,依据周围水域的实时态势做出决策,无法对整条横渡航线进行整体把握。因此,所选择航法经常使渡船在航道中间,处于低速等待穿越时机的状态,横渡航行风险相对较高。建立安全评价模型,结合研究添加重要判据求得实际最近会遇距离,加入船舶尺度建立安全评价指标作为适应度函数阈值。应用遗传算法(Genetic Algdrithm, GA)建立基于遗传算法的内河横渡模型,综合考虑经济性、转向角度和转向点数量求取路径的适应值并执行遗传操作。试验选取长江某渡口真实交通流数据作为输入进行模型验证,导出迭代所得路径关键参数,经过专家对关键参数进行分析比对,所得航线的经济性、及时性和安全性均较为合理,构建的模型可应用于渡船横渡内河水域航法寻优中。

River Crossing Navigation Optimization for Ferry by Means of Genetic Algorithm

The process of sailing a ferry to cross a river normally entirely controlled by the captain according to the situation of the surrounding waters,even if his experience-based decision might not be the best for the whole crossing route.It is often seen that a ferry has to wait in the middle of the navigation channel for its opportunity to cross the river,which associates high risk.A reasonable plan based on the GA(Genetic Algorithm)may improve the operation.The proposed model can evaluate the safety factors(e.g.ship encounter distances,ship dimensions),the economy feasibility and the navigation operations(e.g.the steering angle and number of turning points)of crossing scheme options,and gives respective“fitness”indexes.The method is performed and proved by comparative analysis on one of the ferry services crossing the Yangtze River and should be introduced to the river crossing navigation optimization practice.