Deep Learning-Based Model Predictive Control for Resonant Power Converters

Resonant power converters offer improved levels of efficiency and density. In order to implement such systems, advanced control techniques are required take the most converter. this context, model predictive arises as a powerful tool that is able consider nonlinearities constraints, but it requires solution complex optimization problems or strong simplifying assumptions hinder its application in real situations. Motivated by recent theoretical advances field deep learning, paper proposes learn, offline, optimal policy defined formulation using neural networks so online use learned controller only evaluation network. The obtained can be executed very rapidly on embedded hardware. We show potential presented approach Hardware-in-the-Loop setup an FPGA-controlled resonant