Finite-time Adaptive Neural Backstepping Control for Three-Phase AC/DC Converters Under Uncertain Disturbances

In order to suppress the influences of disturbances, like parameter uncertainties and unknown load, a finite-time adaptive neural backstepping control (FTANBC) strategy is proposed for three-phase AC/DC converters in this paper, where the neural networks are designed to compensate uncertain disturbances. Furthermore, the finite-time Lyapunov stable theory is combined with the adaptive neural backstepping to achieve good dynamic performance and enhance the system robustness. In addition, the FTANBC controller is constructed in the alpha beta frame, which eliminates the use of the phase locked loop. Compared with the PI method, the proposed method has good disturbance rejection ability. The overshoot of the DC voltage under the FTANBC method is minimized by about 90% in the start-up period, the settling time has been minimized by about 95% with various disturbances. Simulation results verify the correctness of the developed method.