A novel approach for the analysis and design of sliding mode controllers for PWM DC/DC converters is presented. The main advantage of this non-linear controller is that there is no restriction on the size of the signal variations around the operating point. Small as well as large signal variations around the operating point are considered. Sliding mode controllers for buck, boost, buck-boost, and Cuk converters have been designed and discussed. These controllers have been simulated on a digital computer and their dynamic performances have been shown to be satisfactory. Finally, Lyapunov's second theorem has been used to verify the stability of the designed sliding mode controller for the Cuk converter.

Power electronic converters are periodic time-variant systems, because of their switching operation. The generalized state-space averaging method is a way to model them as time independent systems, defined by a unified set of differential equations, capable of representing circuit waveforms. Therefore, it can be a convenient approach for designing controllers to be applied to switched converters. This brief shows that the generalized state-space averaging method works well only within specific converter topologies and parametric limits, where the model approximation order is not defined by the topology number of components. This point is illustrated with detailed examples from several basic dc/dc converter topologies.

No abstract

No abstract