Title :
Adaptive B-spline network control for three-phase PWM AC-DC voltage source converter
Author :
Cheng, K. W Eric ; Wang, H.Y. ; Sutanto, D.
Author_Institution :
Dept. of Electr. Eng., Hong Kong Polytech., Kowloon, Hong Kong
Abstract :
A neural network control method-adaptive B-spline neural network for three-phase AC-DC voltage source converters that realizes a sinusoidal AC input current and unity power factor is discussed in this paper. Compared to the other PWM techniques, the main advantage of the neural network is that it has excellent merit for nonlinear control and is adaptive enough to fit the environment change. Since the training for the network is online in this paper, it is more robust to external disturbances. B-spline neural network is used because it is characterized by a local weight updating scheme with the advantages of fast convergence speed and low computation complexity. This is fairly important for real-time control application. The stability of the network control strategy can be shown using the Lyapunov law. Simulation results are presented to illustrate the effectiveness of the proposed control strategy
Keywords :
AC-DC power convertors; Lyapunov methods; PWM power convertors; control system analysis; control system synthesis; neurocontrollers; rectifying circuits; robust control; splines (mathematics); Lyapunov law; adaptive B-spline network control; computation complexity; control design; control simulation; convergence speed; external disturbance robustness; local weight updating scheme; network control strategy; neural network control method; nonlinear control; real-time control application; sinusoidal AC input current; stability; three-phase PWM AC-DC voltage source converter; unity power factor; Adaptive control; Adaptive systems; Analog-digital conversion; Neural networks; Programmable control; Pulse width modulation; Reactive power; Robustness; Spline; Voltage control;
Conference_Titel :
Power Electronics and Drive Systems, 1999. PEDS '99. Proceedings of the IEEE 1999 International Conference on
Print_ISBN :
0-7803-5769-8
DOI :
10.1109/PEDS.1999.794608