Fast Real-Time Constrained Predictive Current Control in Permanent Magnet Synchronous Machine-Based Automotive Traction Drives

The current control in permanent magnet synchronous machine (PMSM) drives is a challenging problem which has to deal with physical and real-time constraints. The goal of this paper is to provide a control solution that can deal effectively with these challenges. The objective is to adjust the stator voltages to obtain fast and monotonic transient current response while satisfying the existing constraints. To this end, using the PMSM model in the rotor coordinates, an explicit one-step ahead predictive control law which uses tools from Lyapunov theory is developed. First, a polytopic approximation of the quadratic constraints is used to reduce the complexity. Then, the optimization problem is reduced to a linear program. Furthermore, a piecewise affine explicit control law is obtained via multiparametric linear programming. As a consequence, the online computation of the control law is reduced to a point location problem yielding a fast control method suitable for real-time implementation. Second, the speed-dependent terms of the mathematical model are considered as disturbance. Then, a state and disturbance observer is designed using a quadratic Lyapunov function to guarantee asymptotic stability of the estimation error. Real-time results obtained in an industrial hardware-in-the-loop test-bench are reported and analyzed.