摘要
为全面控制船舶航行轨迹,保持良好的航向应用条件,提出大数据网络下的船舶轨迹异常故障检测优化技术。从上限边界数值确定、下限边界数值确定2个角度,完成大数据网络下的船舶轨迹异常范围确定。在此基础上,通过轨迹故障类型划分、节点故障检测属性关系确定、偏导优化系数计算3个步骤,完成大数据网络下船舶轨迹异常故障检测技术的优化操作。模拟实验结果表明,与基础故障检测技术相比,应用优化技术手段后,船舶航行轨迹的时间复杂度得到适当降低,单一节点处的轨迹密度提升明显,船舶航行应用条件得到有效保障。
In order to fully control the ship’s trajectory and maintain good heading application conditions,an optimization technology for abnormal fault detection of ship’s trajectory based on large data network is proposed.The abnormal range of ship trajectory under large data network is determined from the two angles of upper bound and lower bound.On this basis,the optimal operation of ship trajectory abnormal fault detection technology under large data network is completed through three steps:the classification of trajectory fault types,the determination of node fault detection attribute relationship and the calculation of partial derivative optimization coefficient.The simulation results show that,compared with the basic fault detection technology,the time complexity of ship trajectory is appropriately reduced,the trajectory density at a single node is obviously increased,and the application conditions of ship navigation are effectively guaranteed.
作者
刘志方
LIU Zhi-fang(Guangzhou Nanyang Polytechnic College,Guangzhou 510925,China)
出处
《舰船科学技术》
北大核心
2019年第10期34-36,共3页
Ship Science and Technology
关键词
大数据网络
轨迹故障
异常检测
边界数值
故障类型
检测属性
偏导优化
large data network
trajectory fault
anomaly detection
boundary value
fault type
detection attribute
bias optimization
作者简介
刘志方(1971–),男,硕士,讲师,研究方向为计算机应用技术。
