Neural network control for dynamic voltage restorer

The dynamic voltage restorer is a power electronic device which has demonstrated its ability to protect sensitive loads from the effects of costly voltage sags. The dynamic voltage restorer is connected in series with the distribution feeder at medium voltage. A neural network control is proposed. As is known, the PWM converter is actually nonlinear, but the PI is a linear controller, so it can only guarantee the stability of this converter in a local area. Simulation results are shown to validate these control methods.