On-line continuous control set MPC for PMSM drives current loops at high sampling rate using qpOASES

Several Continuous Control Set Model Predictive Control algorithms have been successfully proposed for synchronous motor drive current control applications. The choice of the optimal input for the system results into the resolution of a constrained quadratic programming problem. However, most of the MPC algorithms found in literature do not consider any constraints in the identification of the optimal inputs, mainly because of computational burden limit. In the electric drives field this limitation is particularly strict, since industrial applications requires control frequency higher than 5 kHz. This paper deals with the analysis of an on-line continuous control set MPC which solves in real-time mode the quadratic problem, considering also the input constraints. The input feasible domain is mathematically derived in this paper and the information is properly given to a quadratic programming problem solver. The open-source solver qpOASES run real-time the MPC algorithm on a high performance RD hardware, at rate of 6 kHz. The proposed control scheme is applied as current regulator for an interior permanent magnet synchronous motor. The performances achieved with the proposed constrained MPC are compared to those of another MPC that computes only the unconstrained solution, coupled with an over-modulation strategy. Both steady state and transients experimental tests are reported, considering the operating conditions where the voltage constraints are violated.