Title :
Multi-ELM-based real-time simulation credibility assessment method and tested on ship electric propulsion system
Author :
Sheng, Liu ; Pengfei, Zhi ; Lanyong, Zhang ; Bowen, Xing
Author_Institution :
Harbin Engineering University College of Automation, Harbin 150001, China
Abstract :
Simulation credibility analysis involves analyzing and calculating lots of indexes and data, which is very time-consuming. This paper firstly reviews the ELM´s applications in power engineering and then develops a Multi-ELM-based realtime simulation credibility assessment method. The training database come from the actual system. The trained ELMs are used in the simulation to get the credibility result. Because of the fast speed of ELM, the credibility assessment can be applied for real-time simulation. The proposed method is tested on a ship electric propulsion system, and the testing results show that it can exactly and rapidly assess the simulation credibility result.
Keywords :
Analytical models; Data models; Indexes; Propulsion; Testing; Training; Credibility assessment; Electric propulsion system; Extreme learning machine (ELM);
Conference_Titel :
Control Conference (CCC), 2015 34th Chinese
Conference_Location :
Hangzhou, China
DOI :
10.1109/ChiCC.2015.7261041