Model Predictive Control of Load-Commutated Inverter-Fed Synchronous Machines

This paper considers torque regulation of a variable-speed synchronous machine fed by a line-commutated rectifier and a load-commutated inverter. The proposed control approach is model-predictive control where both the rectifier and inverter firing angles are considered as control inputs. Conventional controllers assign different tasks to the rectifier and inverter firing angle. In contrast, the model-predictive controller coordinates the firing angles and this improves the dynamic performance and disturbance rejection. In particular, the proposed controller handles line side under voltage conditions better than a conventional PI controller. The nonlinear model-predictive torque controller has been implemented on an embedded system and applied in an experimental test bed. The experiments confirm that the controller is able to successfully ride through line side under voltage conditions.