基于自回归推广模型的海风预测方法分析  

Application of Auto-Regressive Model in Predicting Sea Wind

作　　者：张超群[1] 张帆[2] 罗伟强 周磊[2,3] ZHANG Chaoqun;ZHANG Fan;LUO Weiqiang;ZHOU Lei(State Key Laboratory of Maritime Technology and Safety,Shanghai Ship and Shipping Research Institute Co.,Ltd.,Shanghai 200135,China;Sanya Nanhai Innovation and Development Base,Harbin Engineering University,Sanya 572024,China;College of Mechanical and Electrical Engineering,Harbin Engineering University,Harbin 150001,China)

机构地区：[1]上海船舶运输科学研究所有限公司水路交通控制全国重点实验室,上海200135 [2]哈尔滨工程大学三亚南海创新发展基地,海南三亚572024 [3]哈尔滨工程大学机电工程学院,哈尔滨150001

出　　处：《上海船舶运输科学研究所学报》2024年第5期16-21,共6页Journal of Shanghai Ship and Shipping Research Institute

摘　　要：为准确预测海上风速和风向数据,提升船舶海上航行的安全性,针对现有海风预测方法存在的预测精度不高的问题,提出一种基于自回归(Auto-Regressive,AR)推广模型的海风预测方法。将风速和风向看作风矢量的2部分,在确定2组数据的依赖性之后,分别采用差分整合移动平均自回归(Auto-Regressive Integrated Moving Average,ARIMA)模型和向量自回归(Vector Auto-Regressive,VAR)模型这2种AR推广模型对风矢量进行预测。试验结果表明,VAR模型的预测结果中有34.17%的数据落在误差允许范围内,而ARIMA模型的预测结果中有61.25%的数据落在误差允许范围内,该方法可供海上风速和风向预测参考。In order to improve the accuracy of prediction of wind speed and direction at sea the application of Auto-Regressive(AR)models is studied.The autocorrelation of the wind data time series is examined before AR processing.The Prediction of wind vector is conducted with Auto-Regressive Integrated Moving Average(ARIMA)model and the Vector Auto-Regressive(VAR)model separately and the results are examined.The experimental results show that 34.17%of the VAR model prediction results fall within the error range,while this number for ARIMA model reaches 61.25%.

关 键 词：风向风速预测 差分整合移动平均自回归(ARIMA)模型 向量自回归(VAR)模型 

分 类 号：U675.5[交通运输工程—船舶及航道工程]