摘要
为提升船舶抛锚作业的安全性和智能化水平,针对传统方法数据维度不足、环境适应性差的问题,提出一种基于多源异构数据融合的船舶抛锚智能检测与风险预警方法。通过设计时空加权关联算法和分层融合架构,构建卷积神经网络-长短时记忆(Convolutional Neural Networks-Long Short-Term Memory,CNN-LSTM)融合检测模型和逻辑回归(Logistic Regression,LR)风险预警模型。结合仿真与实船测试,验证模型在不同场景下的识别精度和预警性能。研究表明,所提方法可实现船舶抛锚状态的精准识别和高效预警,具有良好的工程适应性和实际应用价值。
To enhance the safety and intelligence of ship anchoring operations,this study proposes an intelligent detection and risk warning method based on multi-source heterogeneous data fusion,addressing the limitations of traditional methods in data diversity and environmental adaptability.A spatiotemporal weighted matching algorithm and layered fusion architecture support the construction of a Convolutional Neural Networks-Long Short-Term Memory(CNN-LSTM)detection model and an Logistic Regression(LR)based risk prediction model.Simulation and field tests confirm the method’s accuracy,robustness,and practical applicability in diverse scenarios.
作者
薛勇军
张善明
余芳迪
黄进
丁清源
XUE Yongjun;ZHANG Shanming;YU Fangdi;HUANG Jin;DING Qingyuan(Zhoushan City Water Supply Co.,Ltd.,Zhoushan,Zhejiang 316000,China)
出处
《智能物联技术》
2025年第4期65-68,共4页
Technology of Io T& AI
关键词
多源数据融合
船舶抛锚
智能检测
风险预警
multi-source data fusion
ship anchoring
intelligent detection
risk warning
