Condition assessment is an indispensable monitoring step in marine electric propulsion, and is also fundamental in automatic control and condition based maintenance. Recognizing the shortcomings of slow convergence of the intelligent neural-network as well as local optimization, a new method of condition assessment of marine electric propulsion system using support vector machine SVM is applied in this paper. It determined the kernel function and classification method. Using training sampling and K-multiple principal component analysis to optimize the parameters of the kernel function, it obtained a model of condition assessment compatible with SVM. Simulation using MATLAB shows that it can provide high precision and is suitable for generalization as well as improving ship safety.
Condition Assessment of Marine Electric Propulsion System Using Support Vector Machine
Second International Conference on Transportation Information and Safety ; 2013 ; Wuhan, China
ICTIS 2013 ; 2156-2163
2013-06-11
Conference paper
Electronic Resource
English
Engineering Index Backfile | 1918