基于LSTM-ResNet模型的定点有效波高预测

Prediction of significant wave height of a fixed point based on LSTM-ResNet model

基于北部湾单站位浮标采集数据,提出一种基于长短期记忆网络(LSTM)和残差网络(ResNet)相融合的网络模型,将研究结果运用到短时波高预测中,并将模型的数值预测结果与LSTM网络、反向传播(BP)网络和ResNet网络在短时波高预测中的数值计算结果进行对比分析。结果表明:该模型在短时波高预测中,预测结果偏差较小且实用性较高,能够在一定条件下提高有效波高短期预测数值的有效性。

Based on the data collected by single-station buoys in the Beibu Gulf, this paper proposes a network model using the fusion of Long Short-Term Memory(LSTM) and Residual Network(ResNet), and applies the research results to short-term wave height forecasting;Thereafter, the numerical prediction results of the model are compared with the numerical calculation results of LSTM network, Back Propagation(BP) network and ResNet in the prediction of short-term wave height. Finally, the research results show that the LSTM-ResNet model have the characteristics of small deviation and high practicability in predicting the short-term wave height,and could improve the effectiveness of the short-term prediction value of the significant wave height under certain conditions.