Combination of SOM and RBF Based on Incremental Learning for Acoustic Fault Identification of Underwater Vehicles

Lots of growing neural network models have been proposed to tackle the incremental learning problem, but they also bring about the problem of fast growing complex structure. In this paper, we present a combinational Neural Network of SOM (Self-Organizing Maps) and RBF (Radial Basis Function) based on incremental learning method. The experiment of acoustic fault sources identification of underwater vehicle shows that the proposed network has better generalization performance than traditional RBF network, and can improve the speed and accuracy of identification.