Adaptive B-spline network control for three-phase PWM AC-DC voltage source converter

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. Comparing 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 on-line 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 Lyapunov law. Simulation results are presented to illustrate the effectiveness of the proposed control strategy.