针对当前方法存在的风险态势综合得分计算误差大、风险态势预测精度低等问题，提出大数据驱动的液化天然气（Liquefied Natural Gas，LNG）动力船风险态势预测方法。根据LNG动力船风险态势预测需求，确定风险数据来源，并进行标准化处理，选取最适合风险指标构建风险指标体系，采用组合赋权法确定风险指标权重系数，联合风险指标实际数值计算风险态势综合得分，并制定风险态势等级划分规则，实现了LNG动力船风险态势的有效预测。实验结果表明，设计方法的LNG动力船风险态势预测资源消耗量小，风险态势综合得分计算误差最小值为0.2%，船风险态势的预测结果与实际结果相同，具有较高的实际应用价值。
In response to the problems of large calculation errors and low prediction accuracy of risk situation comprehensive scores in existing methods, a big data-driven Liquefied Natural Gas (LNG) powered ship risk situation prediction method is proposed. Based on the risk situation prediction requirements of LNG powered ships, the sources of risk data were determined and standardized. The most suitable risk indicators were selected to construct a risk indicator system. The combination weighting method was used to determine the weight coefficients of risk indicators, and the actual values of risk indicators were combined to calculate the comprehensive score of risk situation. The risk situation classification rules were formulated to achieve effective prediction of the risk situation of LNG powered ships. The experimental results show that the resource consumption for predicting the risk situation of LNG powered ships using the design method is small, and the minimum error in calculating the comprehensive score of the risk situation is 0.2%. The predicted results of the ship's risk situation are consistent with the actual results, and have high practical application value.
2025,47(15): 168-172 收稿日期：2025-3-14
DOI：10.3404/j.issn.1672-7649.2025.15.028
分类号：U674.925
基金项目：山东省基金面上项目（ZR2024MF024）
作者简介：孙守强(1978-),男,硕士,教授,研究方向为信息安全、智能软件、商务智能、数据挖掘与分析及大数据可视化等
参考文献：
[1] 郑盛文, 周游. LNG动力船舶船厂内加注方案[J]. 船海工程, 2024, 53(S1): 69-72.
ZHENG S W, ZHOU Y. Travel around LNG powered ship shipyard refueling plan[J]. Ship and Sea Engineering, 2024, 53(S1): 69-72.
[2] 王学峰, 苏高辉, 谢旭阳, 等. 基于do算子的船舶风险干预及推理方法[J]. 科学技术与工程, 2023, 23(12): 5332-5339.
WANG X F, SU G H, XIE X Y, et al. Ship risk intervention and inference method based on do operator[J]. Science, Technology and Engineering, 2023, 23(12): 5332-5339.
[3] 冯海商, 陈厚忠. 基于动态贝叶斯网络的客滚船航行风险预测[J]. 武汉理工大学学报, 2023, 45(9): 96-102.
FENG H S, CHEN H Z. Risk prediction of passenger roll on/roll off ship navigation based on dynamic Bayesian network [J]. Journal of Wuhan University of Technology, 2023, 45(9): 96-102.
[4] 胡甚平, 邹春, 吴建军, 等. 多重防护机制下LNG动力船风险态势分析方法[J]. 安全与环境学报, 2024, 24(8): 2895-2903.
HU S P, ZOU C, WU J J, et al. Risk situation analysis method for LNG powered ships under multiple protection mechanisms[J]. Journal of Safety and Environment, 2024, 24(8): 2895-2903.
[5] 张元元, 张春昌, 孙虎, 等. 基于动态贝叶斯网络的甲醇船对船加注风险评估[J]. 上海海事大学学报, 2024, 45(4): 54-61.
ZHANG Y Y, ZHANG C C, SUN H, et al. Risk assessment of methanol ship to ship refueling based on dynamic Bayesian network[J]. Journal of Shanghai Maritime University, 2024, 45(4): 54-61.
[6] 耿艳芬, 郭梦雅, 陈会, 等. 基于水流特性的桥区船舶通航风险分析[J]. 东南大学学报(自然科学版), 2024, 54(3): 717-723.
GENG Y F, GUO M Y, CHEN H, et al. Risk analysis of ship navigation in bridge areas based on water flow characteristics[J]. Journal of Southeast University (Natural Science Edition), 2024, 54(3): 717-723.
[7] 章露达, 朱凯, 冯甲鑫, 等. 搭接式海上风机安装船的水动力分析[J]. 中国海洋大学学报(自然科学版), 2025, 55(1): 166-174.
ZHANG L D, ZHU K, FENG J X, et al. Hydrodynamic analysis of overlapping offshore wind turbine installation ship[J]. Journal of Ocean University of China (Natural Science Edition), 2025, 55(1): 166-174.
[8] 闫纪红, 姬思阳. 大数据驱动的车间数字孪生模型构建方法[J]. 机械工程学报, 2023, 59(12): 62-77.
YAN J H, JI S Y. Building method of workshop digital twin model driven by big data [J]. Journal of Mechanical Engineering, 2023, 59(12): 62-77.
