Abstract
A Prediction Method for Deck Motion of Aircraft Carrier Based on Particle Swarm Optimization and Kernel Extreme Learning Machine
Highlights
As a moving platform for aircraft at sea, aircraft carriers have been widely valued for the powerful offensive/defensive ability of carrier-based aircraft
The kernel function of the support vector machine (SVM) is introduced in extreme learning machine (ELM), which is renamed kernel extreme learning machine (KELM), in which kernel mapping is used as a substitute for random mapping
To satisfy the requirements of real-time calculation and high-accuracy of deck-motion prediction, ELM was introduced, and the concept of kernel mapping in SVM was introduced to overcome the difficulty in selecting the number of hidden nodes as well as the instability and poor generalization caused by random mapping in ELM
Summary
As a moving platform for aircraft at sea, aircraft carriers have been widely valued for the powerful offensive/defensive ability of carrier-based aircraft. With the aim of realizing deck-motion prediction, intelligent methods such as grey prediction and neural network prediction are introduced.[12,13,14] Among these methods, the grey prediction method requires the predicted object to conform to the exponential function law, but it is an ideal assumption and difficult to realize When this assumption cannot be satisfied, fitting and generalization ability will degrade. When using ELM, there is no mature method of selecting the number of hidden nodes which is a decisive factor deciding prediction accuracy This method suffers from model instability and a low generalization ability caused by random mapping in the hidden layer.[19] In order to solve these problems, a new network named kernel extreme learning machine (KELM) was introduced in Ref. 16.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
