考虑时间相关性的风速和波浪高度短期预测  

Short-term forecast of wind speed and wave height considering time correlation

作　　者：姚骥 汪雪良[1,2] 武文华 顾学康 张欣玉[1,2] YAO Ji;WANG Xue-iang;WU Wen-hua;GU Xue-kang;ZHANG Xin-yu(China Ship Science Research Center,Wuxi 214082,China;Taihu Laboratory of Deepsea Technological Science,Wuxi 214082,China;Faculty of Vehicle Engineering and Mechanics,Dalian University of Technology,Dalian 116024,China)

机构地区：[1]中国船舶科学研究中心,江苏无锡214082 [2]深海技术科学太湖实验室,江苏无锡214082 [3]大连理工大学运载工程与力学学部,辽宁大连116024

出　　处：《船舶力学》2024年第6期832-842,共11页Journal of Ship Mechanics

基　　金：国家重点研发计划项目(2021YFC2802300);江苏省卓越博士后计划项目(2023ZB629)。

摘　　要：风与波浪高度是海洋工程设计的重要参数之一,其准确预测具有重要的工程意义。针对波浪高度复杂多变、难以预测的挑战,本文考虑时间相关性,利用长短时记忆(Long-Short-Term Memory,LSTM)神经网络方法建立风和波浪高度的短期预测模型。首先,对波浪高度和风速等海洋环境原始时间序列进行分析,并选取不同时间间隔对浪高和风速极值进行取值;其次,对风速与波浪高度开展相关性分析,并基于LSTM方法,构建风和波浪高度单步预测模型。在时间间隔为0.5 h时,波浪的预测误差为0.12 m;进而构建风和波浪联合预测模型,预测风速下的波高预测误差仅为0.12m,与实测风速下的波浪预测误差基本一致。As wind speed and wave height are the main loading parameters in offshore facility operations,their accurate prediction is of great importance.In order to solve the problem of wind speed and wave height prediction with complex and changeable characteristics,a wave height forecast model was established based on prototype monitoring data and Long-Short-Term Memory(LSTM)neural network.Firstly,the correlation analysis of wind speed and wave height was carried out based on prototype monitoring data.Then,a one-step-ahead wind speed forecast model and wave height forecast method were established based on LSTM neural network.Different prediction models with different time intervals(t=0.5 h,1 h,3 h)were built to verify the accuracy.Finally,a joint prediction model based on two forecast models was obtained with a prediction error of only 0.12 m at the time interval of 0.5 h.

关 键 词：相关性分析 长短时记忆神经网络 风速预测 波浪预测 联合预测模型 

分 类 号：P751[交通运输工程—港口、海岸及近海工程]