Browse > Article
http://dx.doi.org/10.7837/kosomes.2021.27.7.1124

Active Front End Rectifier Control of DC Distribution System Using Neural Network  

Kim, Seongwan (Ocean Polytech Team, Korea Institute of Maritime and Fisheries Technology)
Jeon, Hyeonmin (Division of Marine System Engineering, Korea Maritime & Ocean University)
Kim, Jongsu (Division of Marine System Engineering, Korea Maritime & Ocean University)
Publication Information
Journal of the Korean Society of Marine Environment & Safety / v.27, no.7, 2021 , pp. 1124-1128 More about this Journal
Abstract
As regulations of emissions from ships become more stringent, electric propulsion systems have been increasingly used to solve this problem in vessels ranging from large merchant ships to small and medium-sized ships. Methods for improving the efficiency of the electric propulsion system include the improvement of power sources; the use of a system linked to environmentally friendly power sources, such as batteries, fuel cells, and solar power; and the development of hardware and control methodology for rectifiers, power conversion devices, and propulsion motors. The method using a phase-shifting transformer with diodes has been widely used for rectification. Power semiconductor devices with grid connection to an environmentally friendly power source using DC distribution, a variable speed power source, and the application of small and medium-sized electric propulsion systems have been developed. Accordingly, the demand for active front-end (AFE) rectifiers is increasing. In this study, a method using a neural network rather than a conventional proportional-integral controller was proposed to control the AFE rectifier. Tested controller data were used to design a neural network controller trained through MATLAB/Simulink. The neural network controller was applied to a rectification system designed using PSIM software. The results indicated the effectiveness of improving the waveform and power factor DC output stage according to the load variation. The proposed system can be applied as a rectification system for small and medium-sized environmentally friendly ships.
Keywords
Neural network; AFE rectifier; Electric propulsion system; DC distribution system; Eco-friendy ship;
Citations & Related Records
Times Cited By KSCI : 1  (Citation Analysis)
연도 인용수 순위
1 Kim, J. S., D. H. Seo, and J. H. Choi(2013), CHarmonic and Torque Ripple Reduction of Electric Propulsion System using 12-Pulse Diode Rectifier by Auxiliary Supply,Journal of the Korean Society of Marine Environment & Safety, Vol. 19, No. 1, pp. 66-70.   DOI
2 Zhang, S.(2021), Comparison of Diode Rectifier and Active-Front-End for Synchronous Generators Suppling DC Loads for Maritime Applications, 2021 International Aegean Conference on Electrical Machines and Power Electronics (ACEMP) & 2021 International Conference on Optimization of Electrical and Electronic Equipment (OPTIM), pp. 259-264.
3 Chai, M., D. R. Bonthapalle, L. Sobrayen, S. K. Panda, D. Wu, and X. Chen(2018), Alternating current and direct current-based electrical systems for marine vessels with electric propulsion drives, Applied Energy, Vol. 231, pp. 747-756.   DOI
4 Jeon, H. M. and J. S. Kim(2018), A Study on Application of Electric Propulsion System using AFE Rectifier for Small Coastal Vessels, Journal of the Korean Society of Marine Environment & Safety, Vol. 24, No. 3, pp. 373-380.   DOI
5 Hwang, J. Y., C. H. Jeon, H. M. Jeon, and J. S. Kim(2019), A Study on the Application of Domestic ferry to a Battery Propulsion Ship, Journal of the Korean Society of Marine Environment & Safety, Vol. 25, No. 7, pp. 945-952.   DOI
6 Robert, H. N.(1992), III.3-Theory of the Backpropagation Neural Network, Academic press, pp. 65-93.