Intelligent ocean wave height prediction system using light GBM model



Sea wave height, Light GBM, Machine learning, Forecasting, Wave variables


Forecasting the heights of marine waves is an important tool for offshore and coastal engineering and a huge undertaking in marine detection and warning. However, a precise forecast of the Sea Wave Height (SWH) is challenging and outstanding to waves' volatility and fluctuation characteristics. Therefore, our research proposes an Intelligent Ocean Wave Height Prediction system using a light gradient boosting machine learning. Wave speed, peak wave direction, zero up crossing wave period, wave period, and SWH are among the wave-based properties we extract. Then the inputs are fed into the Light GBM, which performs these high-dimensional inputs admirably, and the model is simple to interpret. Furthermore, because LightGBM is noise-insensitive and can work with unnecessary data in time-window-size data, the proposed method can be used to estimate wave height. Consequently, our proposed approach outperforms when compared to the existing techniques.