Disturbance Observer Based Back-Stepping Control for Buck Converters with Unmatched Disturbances: An RBF Neural Network Approach

In this paper, a novel disturbance observer-based back-stepping control (DOBC) strategy combining with a radial basis function neural network (RBFNN) is proposed for DC-DC buck converters. Firstly, the state space average model of buck converter is established and it is turned into a general second-order model through coordinate transformation. Model parameter uncertainties and external disturbances are considered in two parts instead of a lumped disturbance. Based on the general second-order model of buck converter, RBFNN is used to estimate parameter uncertainties and a special disturbance observer is used to observe external disturbances. Then, back-stepping method is used to obtain the controller. Simulations are given to verify the advantages of the presented approach.