Ship rolling motion prediction based on extreme learning machine
Huixuan Fu, Yuchao Wang, Zhang Hongmei · 2015
The traditional time series predictive models are not able to achieve a satisfying prediction effect in the problem of a non-linear system and nonstationary time series. To solve these problems, ship rolling time series prediction, which is based on Extreme Learning Machine, was proposed. Extreme Learning Machine is a new single-hidden layer learning algorithm for Feedforward Neural Network, don't need to set up a large number of network training parameters, it's superior to the traditional Neural Network learning algorithm. The simulation experiments used multiple-input/single-output (MISO) Extreme Learning Machine prediction model and BP Neural Network prediction model. The results indicated that Extreme Learning Machine was more accurate than BP Neural Network.