文献类型：期刊文献

英文题名：Accurate tidal prediction based on empirical mode decomposition-enhanced multilayer perceptron

作者：Liao, Shenghao[1];Wang, Lijun[1];Wang, Sisi[1];Jia, Baozhu[2];Yin, Jianchuan[2];Li, Ronghui[2]

机构：[1]Guangdong Ocean Univ, Naval Architecture & Shipping Coll, Zhanjiang 524088, Peoples R China;[2]Guangdong Ocean Univ, Guangdong Prov Key Lab Intelligent Equipment South, Zhanjiang 524088, Peoples R China

年份：2025

卷号：336

外文期刊名：OCEAN ENGINEERING

收录：SCI-EXPANDED(收录号：WOS:001511075500001)、、EI(收录号：20252418589587)、Scopus(收录号：2-s2.0-105007807870)、WOS

基金：This work was partially supported by National Science Foundation of China (Grant NO. 52171346and 52271361) , the Fund of Guangdong Provincial Key Laboratory of Intelligent Equipment for South China Sea Marine Ranching (Grant NO. 2023B1212030003) and the Key Area Project of Ordinary Universities in Guangdong Province (Grant NO. 2024ZDZX3054) .

语种：英文

外文关键词：Tidal prediction; Multilayer perceptron; Empirical mode decomposition; Hybrid model

外文摘要：To address the complex characteristics of tidal variations, including spatial heterogeneity, temporal variability, and uncertainty, this study proposes a hybrid tide prediction model that enhances the multilayer perceptron (MLP) using empirical mode decomposition (EMD). Firstly, the EMD method is employed to automatically decompose the original tidal time series into multiple subsequences with specific frequencies and dynamic features, effectively extracting regular signals that are easier to model and predict. Secondly, the MLP model, combined with sliding time window techniques, is employed for accurate prediction of low nonlinearity subsequences. Finally, the predictions of all decomposed subsequences are synthesized through a reconstruction process, resulting in a comprehensive and accurate prediction of the overall tidal series. Simulation tests were conducted using measured data from multiple tidal stations. The results demonstrate that the proposed model achieves higher prediction accuracy compared to traditional tidal prediction models, effectively capturing the dynamic characteristics of complex tidal variations. Therefore, the EMD-MLP model proposed in this study provides an efficient and reliable solution for tidal prediction, offering robust decision support for applications in marine engineering, disaster prevention, and tidal energy development.