Research on wave height prediction method based on deep learning and CatBoost
LU Peng
NIAN Shengquan
ZOU Guoliang
WANG Zhenhua
ZHENG Zongsheng
Abstract：Based on deep learning and CatBoost technology,a hybrid model(LACM)was proposed to predict wave heights in the Gulf of Mexico,Bay of Fundy and Gulf of Alaska,which could then be ap-plied to other sea areas.Firstly,the wave data obtained from the National Data Buoy center(NDBC)were preprocessed.Secondly,the LSTM neural network model and CatBoost integrated learning model were constructed.The predicted results were reconstructed.Compared with the LSTM,support vector regression(SVR),CatBoost and other methods,the experimental results showed that the mean abso-lute error(MAE),root mean square error(RMSE)and mean absolute percentage error(MAPE)of the LACM model were the lowest,and the prediction result was the best and the fitting effect was the closest,and it had a certain robustness.
Keywords：AMLSTMCatBoostwave height predictionreconsitution
Publication Date：2024-10-28
Online Publishing Date：2026-05-22(First online date of this platform, not the publication date of the document)
Pages：7( 28-34 )