为优化船舶航行路线，减少因交通事故导致的延误和拥堵，提升海上运输效率和效益，研究基于历史数据挖掘的海上船舶交通事故预测方法。从海事机构获取海上船舶交通事故历史数据后，采用数据挖掘方法中的一维局阿尼自编码器对海上船舶交通事故历史数据展开挖掘，得到海上船舶交通事故特征，再建立灰色SCGM(1,1)C模型，将海上船舶交通事故特征输入到该模型中，并运用当前预测状态中间值作为修正产生，对灰色SCGM(1,1)C模型预测结果进行修正后，得到海上船舶交通事故预测结果。实验表明，该方法具备较强的海上船舶交通事故历史数据挖掘能力，灰色SCGM(1,1)C模型输出的海上船舶交通事故预测结果DBI数值较高，预测海上船舶交通事故能力较好。
Optimize ship navigation routes, reduce delays and congestion caused by traffic accidents, improve maritime transportation efficiency and efficiency, and study a prediction method for maritime ship traffic accidents based on historical data mining. After obtaining historical data of maritime vessel traffic accidents from maritime institutions, the one-dimensional local autoencoder in data mining methods is used to mine the historical data of maritime vessel traffic accidents, obtain the characteristics of maritime vessel traffic accidents, and then establish the grey SCGM (1,1)C model. The characteristics of maritime vessel traffic accidents are inputted into the model, and the intermediate value of the current prediction state is used as a correction. After correcting the prediction results of the grey SCGM (1,1)C model, the prediction results of maritime vessel traffic accidents are obtained. The experiment shows that this method has strong ability to mine historical data of maritime vessel traffic accidents. The grey SCGM (1,1)C model outputs a high DBI value for predicting maritime vessel traffic accidents, indicating a good ability to predict maritime vessel traffic accidents.
2024,46(14): 174-177 收稿日期：2024-01-05
DOI：10.3404/j.issn.1672-7649.2024.14.030
分类号：TP391
作者简介：张哲(1975-),男,硕士,副教授,研究方向为船舶航运技术、海事安全与防污染、海运物流及航运教育
参考文献：
[1] 彭文闯, 郭晟楠, 万怀宇, 等. 面向交通事故预测的时空多模态点过程[J]. 计算机应用研究, 2023, 40(8): 2340-2345.
PENG Wenchuang, GUO Shengnan, WAN Huaiyu, et al. Multimodal spatial-temporal point processes for traffic accident event prediction[J]. Application Research of Computers, 2023, 40(8): 2340-2345.
[2] 刘志, 王锦梦, 孔祥杰. 基于时空图卷积网络的交通事故预测研究[J]. 浙江工业大学学报, 2022, 50(2): 128-135,155.
LIU Zhi, WANG Jinmeng, KONG Xiangjie. Research on traffic accident prediction based on spatial-temporal graph convolutional network[J]. Journal of Zhejiang University of Technology, 2022, 50(2): 128-135,155.
[3] 严丽平, 郭成源. 基于边缘计算的城市交通事故风险深度预测[J]. 计算机仿真, 2022, 39(12): 226-229,329.
YAN Liping, GUO Chengyuan. Urban traffic accident risk depth prediction based on edge computing[J]. Computer Simulation, 2022, 39(12): 226-229,329.
[4] 张庆年, 张瑨, 杨杰, 等. 基于聚类分析的水上交通事故影响因素研究[J]. 重庆交通大学学报(自然科学版), 2022, 41(6): 1-7.
ZHANG Qingnian, ZHANG Jin, YANG Jie, et al. Influencing factors of waterway traffic accidents based on clustering analysis[J]. Journal of Chongqing Jiaotong University (Natural Sciences), 2022, 41(6): 1-7.
[5] 王婧娟, 陈庆奎. 一种时空周期性注意力网络的交通流量预测模型[J]. 小型微型计算机系统, 2022, 43(11): 2321-2327.
WANG Jingjuan, CHEN Qingkui. Traffic flow prediction model based on the spatial-temporal periodic attention network[J]. Journal of Chinese Computer Systems, 2022, 43(11): 2321-2327.
