Ship Rolling Prediction Based on Gray RBF Neural Network

被引：3

作者：

Liu Lisang ^{[1
]}

Peng Xiafu ^{[1
]}

Zhou Jiehua ^{[1
]}

机构：

[1] Xiamen Univ, Dept Automat, Xiamen 361005, Fujian, Peoples R China

来源：

MEASURING TECHNOLOGY AND MECHATRONICS AUTOMATION, PTS 1 AND 2 | 2011年 / 48-49卷

关键词：

ship rolling sequence; prediction; gray model; RBF; GMRBF(2,1);

D O I：

10.4028/www.scientific.net/AMM.48-49.1044

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

To enhance the ship's seaworthiness and seakeeping capacity, a new prediction algorithm based on Gray RBF neural network is presented to forecast roll motion accurately. The second-order gray model GM(2,1) and RBF network are introduced firstly, then using AGO (accumulated generating operation) to weaken randomness and volatility of raw data, which would affect the accuracy of RBF network. On the other hand, the algorithm flow of GMRBF(2,1) is given. Further more, GMRBF(2,1) is applied in a sample of ship roll sequence and effectively improves large prediction error of second-order gray model. The simulation results prove that the new model is more accurate and stabilizer than traditional models.

引用

页码：1044 / 1048

页数：5