Novel Ocean Wave Height and Energy Spectrum Forecasting Approaches: An Application of Semi-Analytical and Machine Learning Models

Accurate and reliable wave forecasting is crucial for optimizing the performance of various marine operations, such as offshore energy production, shipping, fishing. Meanwhile, predicting height achieving sustainability a renewable source, it enables harnessing power efficiently based on water-energy nexus. Advanced models, machine learning models semi-analytical approach, have been developed to provide more accurate predictions ocean waves. In this study, Sverdrup Munk Bretschneider (SMB) Emotional Artificial Neural Network (EANN) Wavelet (WANN) approach will be used estimate parameters in Gulf Mexico Aleutian Basin. The accuracy reliability these approaches evaluated, spatial temporal variability field investigated. available characteristics are generate hourly, 12-hourly, daily datasets. WANN SMB model shows good prediction significant both case studies. model, specifically time scale, Nash–Sutcliffe Efficiency (NSE) peak deviation coefficient (DCpeak) were determined 0.62 0.54 buoy 0.64 0.55 buoy, respectively, height. context testing phase at NSE DCpeak indices exhibit values 0.85 0.61 0.72 while EANN strong tool hourly (Aleutian (NSEEANN = 0.60 DCpeakEANN 0.88), 0.80 0.82)). addition, findings pertaining spectrum density demonstrate that exhibits superior comparison particularly with regard accurately estimating (DCpeakEANN= 0.41), (DCpeakEANN 0.59)).