An ocean current prediction method for profiling float based on RBFNN

In order to obtain accurate small-scale ocean current velocity prediction data, this paper proposes an ocean current prediction method based on RBFNN applied to profiling float. First, a small-scale depth-averaged ocean current velocity (DACV) can be obtained based on profile buoys. After that, a prediction problem model of DACV was established. To solve the ocean current prediction model, a prediction method based on the Radial basis function network (RBFNN) is proposed, and the influence of parameters in RBFNN on the prediction performance is explored. Finally, the proposed method is validated on the ocean current dataset.