A neural network based alternator power angle estimation method for marine electrical network supervision and control

Gang Yao (1) , Guy Lebret (2) , Mourad Aït-Ahmed (3) , Pedro Neiva Kvieska (2) , Tianhao Tang (1)

1 Shanghai Maritime University
2 IRCCyN - Institut de Recherche en Communications et en Cybernétique de Nantes
3 IREENA - Institut de Recherche en Energie Electrique de Nantes Atlantique EA4642

A neural network based alternator power angle estimation method is presented in this paper. The method is dedicated to detect load commutations in marine electrical network and to issue scheduling variables to gain scheduling controller. The global objective is to deal with the unexpected parameter oscillations problem in transient process when load commutations happen.

Mots clés: alternators; commutation; marine systems; neurocontrollers; power generation control; power generation scheduling; alternator power angle estimation method; gain scheduling controller; load commutation detection; marine electrical network supervision-and-control; marine power generation; transient process; Control systems; Electric variables control; Neural networks; Power system transients; Propulsion; Scheduling; Switches; Three-term control; Electrical machine; Embarked networks; Estimation technique; Neural network

Domaines: Automatique / Robotique

Gang Yao, Guy Lebret, Mourad Aït-Ahmed, Pedro Neiva Kvieska, Tianhao Tang. A neural network based alternator power angle estimation method for marine electrical network supervision and control. Power Electronics and Applications, 2009. EPE '09. 13th European Conference on, Sep 2009, Barcelone, Spain. ⟨hal-01146961⟩