Prediction of Ocean Current Velocity in the Western Pacific Ocean Based on EEMD-SSA-BiLSTM Model

The Western Pacific region is an important region of global ocean dynamics, and its currents have significant impacts on the global climate system, marine ecosystems, and economic activities of coastal countries. Aiming at the time-series prediction problem of complex ocean currents in the Western Pacific Ocean, in order to improve the prediction accuracy, this paper adopts the idea of decomposition and integration, combines the ensemble empirical modal decomposition (EEMD) technique and bidirectional long and short-term memory network (BiLSTM), and at the same time, introduces the Sparrow Search Algorithm (SSA) to optimize the neural network, and builds the combined EEMD-SSA-BiLSTM prediction model. The prediction results are evaluated by the performance of coefficient of determination (R2), mean absolute error (MAE), root mean square error (RMSE) and mean absolute percentage error (MAPE). The results show that the EEMD-SSA-BiLSTM method has higher prediction accuracy and stability in the time-series prediction of the western Pacific currents, and is able to effectively capture the complex dynamic features of current changes.