Sensorless explicit model predictive control of the DC-DC buck converter with inductor current limitation

This paper investigates the digital control of DC- DC buck converters. It proposes a new explicit model predictive controller for a buck DC-DC converter. The controller features an inherent current limitation over a wide range of loads and supply voltages using no current sensor. It achieves good reference tracking and good rejection of line and load transients. The complexity of the controller is limited as it consists of a table of optimal state feedback gains that have been derived off-line and that are chosen according to the measured state. Feasibility is validated experimentally on a fixed point DSP.