Title :
Active power line conditioner with a neural network control scheme
Author :
Mei, Tong ; Xiaowu, Shu
Author_Institution :
Coll. of Electr. Eng., Zhejiang Univ., Hangzhou, China
Abstract :
A new control scheme based on artificial neural network (ANN) for a voltage-type active power line conditioner (APLC) is proposed. APLC is a very efficient apparatus for compensating harmonics caused by nonlinear loads. The proposed NN controller can provide appropriate switch control signals for APLC to generate compensation currents to reduce wave distortion in single phase, three-phases three-wires and three-phases four-wires distribution system. The training data are obtained using evolutionary algorithms. Computer simulations confirm the validity of the proposed NN controller.
Keywords :
compensation; evolutionary computation; harmonics suppression; learning (artificial intelligence); neurocontrollers; power distribution control; power system harmonics; power system simulation; artificial neural network; compensation currents; computer simulation; evolutionary algorithms; harmonics compensation; neural network control scheme; neural network controller; nonlinear control process; nonlinear loads; single phase distribution system; switch control signals; three-phases four-wires distribution system; three-phases three-wires distribution system; training data; voltage-type active power line conditioner; wave distortion reduction; Artificial neural networks; Computer simulation; Control systems; Evolutionary computation; Neural networks; Phase distortion; Signal generators; Switches; Training data; Voltage control;
Conference_Titel :
TENCON '02. Proceedings. 2002 IEEE Region 10 Conference on Computers, Communications, Control and Power Engineering
Print_ISBN :
0-7803-7490-8
DOI :
10.1109/TENCON.2002.1182615