Model Predictive Control for DC-DC Converters
The control of power converters is often challenging because of the switched nonlinear characteristics.
Currently, there are two main MPC approaches for power converters, namely finite-set (or direct) MPC (FCS-MPC) and control-set MPC (CCS-MPC). The former approach exploits the discrete nature of converters which has finite number of switching states, leading to the fact that the method is simple and intuitive and in practice it often provides a faster response time. However, this approach faces some problems such as lack of stability guarantees and difficulties in maintaining a constant switching frequency, which is an important issue in commercial applications. On the other hand, in the second approach, the control inputs are continuous-time signals which are sent to a modulator (PWM). This approach is able to decouple the switching frequency from the controller sampling time and can operate the converter at a fixed frequency. The main aim of this thesis is to investigate different MPC approaches for common DC-DC Converters (Buck, Boost, Cuk ...) and improve a method to guarantee a better performance for the system.
Topic Areas:
Helpful/Required Prerequisites:
Simulation: Matlab.
Language: English.
Project start:
Estimated time requirements:
Theory: 40%
Simulations: 40%