Wind Features Extracted from Weather Simulations for Wind-Wave Prediction Using High-Resolution Neural Networks

Chih-Chiang Wei

doi:10.3390/jmse9111257

Abstract

Nearshore wave forecasting is susceptible to changes in regional wind fields and environments. However, surface wind field changes are difficult to determine due to the lack of in situ observational data. Therefore, accurate wind and coastal wave forecasts during typhoon periods are necessary. The purpose of this study is to develop artificial intelligence (AI)-based techniques for forecasting wind–wave processes near coastal areas during typhoons. The proposed integrated models employ combined a numerical weather prediction (NWP) model and AI techniques, namely numerical (NUM)-AI-based wind–wave prediction models. This hybrid model comprising VGGNNet and High-Resolution Network (HRNet) was integrated with recurrent-based gated recurrent unit (GRU). Termed mVHR_GRU, this model was constructed using a convolutional layer for extracting features from spatial images with high-to-low resolution and a recurrent GRU model for time series prediction. To investigate the potential of mVHR_GRU for wind–wave prediction, VGGNet, HRNet, and Two-Step Wind-Wave Prediction (TSWP) were selected as benchmark models. The coastal waters in northeast Taiwan were the study area. The length of the forecast horizon was from 1 to 6 h. The mVHR_GRU model outperformed the HR_GRU, VGGNet, and TSWP models according to the error indicators. The coefficient of mVHR_GRU efficiency improved by 13% to 18% and by 13% to 15% at the Longdong and Guishandao buoys, respectively. In addition, in a comparison of the NUM–AI-based model and a numerical model simulating waves nearshore (SWAN), the SWAN model generated greater errors than the NUM–AI-based model. The results of the NUM–AI-based wind–wave prediction model were in favorable accordance with the observed results, indicating the feasibility of the established model in processing spatial data.

Highlights

Taiwan is located at the turning point of most typhoon paths in the western North Pacific–East Asia region [1]
The Weather Research and Forecasting (WRF) numerical model [8] developed by the National Center for Atmospheric Research–National Oceanographic and Atmospheric Administration (NOAA) is a commonly used, mesoscale community numerical weather prediction (NWP) model [9]
The contributions of this study are as follows: (1) wind field simulation results are valuable for Taiwan, those for typhoon periods, when coastal areas are under strong winds and waves

Summary

Introduction

Taiwan is located at the turning point of most typhoon paths in the western North Pacific–East Asia region [1]. We adopted AI-based techniques for forecasting wind–wave processes near coastal areas during typhoons. The contributions of this study are as follows: (1) wind field simulation results are valuable for Taiwan, those for typhoon periods, when coastal areas are under strong winds and waves. The wind field data obtained from mesoscale meteorological models are highly accurate in reflecting the characteristics of wind fields, which can provide reasonable forced conditions for the simulation of typhoon waves [17]. We employed an NWP WRF model to simulate wind fields, integrated the wind field and in situ observational data, and used them as the input data for the AI-based wind–wave prediction models. Regarding spatial data processing, the HRNet [16] model is constructed using convolutional layers, which facilitates feature extraction from images with high-to-low resolution. The GRU layer enables memory block units to determine the memory time length and provide the most effective memory length of time series data on the time axis

Related Work

Comparison with the Wave Numerical Simulations

WRF Simulations

Typhoon Testing and Prediction

Model Performance Levels

Comparison with the SWAN Simulations