Predictive Control in Power Electronics and Drives: Basic Concepts, Theory, and Methods

In this chapter we revise basic principles and methods of model predictive control with a view towards applications in power electronics and drives. The simplest predictive control formulations use horizon-one cost functions, which can be related to well-established dead-beat controllers. Model predictive control using larger horizons has the potential to give significant performance benefits, but requires more computations at each sampling instant to solve the associated optimization problems. For particular classes of system models, we discuss practical algorithms, which make long-horizon predictive control suitable for power electronics applications.