High dynamic performance constrained optimal control of induction motors

This paper deals with field oriented explicit model predictive control of induction motors. Constraints are inherently handled by model predictive controllers. Taking into account the inverter limitations and maximum admissible stator current magnitude, the torque and flux can be optimally controlled, yielding optimal dynamic performance and disturbance rejection. Casting the problem to solve as a constrained quadratic finite time optimal control problem, an explicit controller can be computed off-line, the resulting piecewise affine control law requiring to run a binary search-tree in order to select the optimal linear state feedback. An observer is used to estimate motor electrical state and disturbance, which allows to effectively takes into account the induced voltage and cross-coupling between the d and q axis. The resulting controller requires little tuning and can automatically be generated for different motor parameters.