在船舶航行过程中，两船近距离并行或交汇时，由于船体周围水流的相互干扰，会出现船间效应。此时船舶会出现不规则的摆动或转向偏差，增加碰撞事故发生的概率。因此，提出基于昆兹概率模型的海上船舶碰撞风险评估算法。通过推算出主船舶避免与目标船舶碰撞的偏航角和停船距离2个表征值，实时捕捉船舶在航行过程中的状态变化。碰撞风险表征值为依据，利用昆兹概率模型估算主船舶在特定航迹上与目标船舶碰撞概率，量化船间效应风险。并基于事件树和碰撞后续事件进行各事件发生概率、伤亡人数期望值、风险值及总风险值的计算公式，完成船舶碰撞风险评估。实验结果表明，该算法能够有效且准确地实现船舶碰撞风险评估，使评估值的最大误差不超过0.0002；通过对比实验凸显，该算法的可行性更优越，可为海上交通安全提供有力支持。
During the process of ship navigation, when two ships are parallel or intersect at close range, the ship to ship effect may occur due to the mutual interference of water flow around the hull. At this time, the ship may experience irregular swaying or steering deviation, increasing the probability of collision accidents. Therefore, a collision risk assessment algorithm for maritime vessels based on the Kunz probability model is proposed. By calculating the yaw angle and stopping distance of the main ship to avoid collision with the target ship, real-time capture of the ship's state changes during navigation can be achieved. Based on the collision risk characterization value, the Kunz probability model is used to estimate the probability of collision between the main ship and the target ship on a specific trajectory. And based on the event tree and subsequent collision events, calculate the probability of each event, expected number of casualties, risk value, and total risk value to complete the ship collision risk assessment. The experimental results show that the algorithm can effectively and accurately achieve ship collision risk assessment, with the maximum error of the assessment value not exceeding 0.0002; Through comparative experiments, it is highlighted that the feasibility of this algorithm is superior and can provide strong support for maritime traffic safety.
2025,47(12): 185-189 收稿日期：2025-1-20
DOI：10.3404/j.issn.1672-7649.2025.12.033
分类号：U675
基金项目：教育部产学合作协同育人项目（240704084094538）；江西省核地学数据科学与系统工程技术研究中心开放基金项目（JETRCNGDSS202002，JETRCNGDSS202204）
作者简介：梁也(1992-),男,博士,讲师,研究方向为数据挖掘和分析
参考文献：
[1] 董鹏, 张其霄, 史博元, 等. 基于云模型的战时海上运输风险评估研究[J]. 火力与指挥控制, 2022, 47(7): 157-164.
[2] 刘小飞, 李明杰, 喻佳. 无人机立体视觉识别船舶航行风险仿真[J]. 计算机仿真, 2024, 41(2): 29-33.
[3] 冯胤伟, 刘正江, 蒋子怡, 等. 基于关联规则挖掘和复杂网络理论的船舶碰撞事故影响因素分析[J]. 大连海事大学学报, 2023, 49(3): 31-44.
[4] YAN W , YI Z , HENGCHAO Z , et al. Assessment method based on AIS data combining the velocity obstacle method and pareto selection for the collision risk of inland ships[J]. Journal of Marine Science and Engineering, 2022, 10(11): 1723–1723.
[5] GIANG H H , THUAN V P, LUONG V T, et al. An evaluation model of the risk of collision between two target ships in a pollution management strategy in coastal waters[J]. International Journal of Knowledge-Based Development, 2022, 12(3-4): 195–216.
[6] 范中洲, 严啸, 李锦晓. 连续弯曲航道的船舶碰撞风险评估方法研究[J]. 安全与环境学报, 2023, 23(10): 3429-3437.
[7] 李新宏, 付雅倩, 刘亚洲, 等. 基于Copula-BN的海上船舶碰撞风险评估方法[J]. 中国安全科学学报, 2023, 33(9): 204-213.
[8] 柯宝贵, 孙碧娇, 徐凡凡, 等. 确定近海船载重力数据系统偏差的虚拟测线新方法[J]. 武汉大学学报(信息科学版), 2024, 49(11): 2071-2078.
[9] 王朝红, 曹穆, 许嘉沄, 等. 考虑蓄电池Run-time等效模型的船舶直流应急电网短路电流计算[J]. 武汉大学学报(工学版), 2024, 57(3): 348-355.
[10] 谭康力, 谢伟, 刘祖源, 等. 机械故障下双桨双舵船操纵运动数值模拟[J]. 华中科技大学学报(自然科学版), 2024, 52(4): 56-63.
[11] 林治浩, 梁海志, 鞠鹏, 等. 运维船与半潜式风机平台碰撞过程的物理建模仿真分析[J]. 船舶力学, 2024, 28(8): 1221-1229.
[12] 王煜, 王庸凯, 任玉清, 等. 基于FTA-BN的中国商渔船碰撞事故因素分析[J]. 大连海洋大学学报, 2024, 39(5): 864-873.
