Title :
Hysteresis Control Method Based on RBF Neural Network
Author :
Xi Zi-qiang ; Guo Huiming ; Qi Lei
Author_Institution :
Sch. of Electr. & Electron. Eng., HuBei Univ. of Technol., Wuhan, China
Abstract :
The key to the effect of active power filter compensation is compensating current :in order to make it better tracking instruction current,using the RBF neural network of hysteresis controllers,with a "K-means, RLS" algorithm for training is applied.Simulation results show that, neural network control is effective for fast variables, clould improve the performance of hysteresis control, get very strong robustness to the system ,and it also verify the theoretical analysis is correct.
Keywords :
hysteresis; neural nets; power filters; radial basis function networks; RBF neural network; active power filter compensation; hysteresis control method; tracking instruction current; Active filters; Algorithm design and analysis; Control systems; Harmonic analysis; Hysteresis; Neural networks; Power harmonic filters;
Conference_Titel :
Power and Energy Engineering Conference (APPEEC), 2012 Asia-Pacific
Conference_Location :
Shanghai
Print_ISBN :
978-1-4577-0545-8
DOI :
10.1109/APPEEC.2012.6307488
