A dynamic programming scheme that optimizes the efficiency of an induction machine drive operated in closed cycle and has both control and state constraints is developed. Application of rotor flux feedforward field orientation control for an induction machine reduces the system equations to contain only three state variables: rotor flux, velocity, and position. Maximum stator current and rotor velocity are set as constraints. Saturation effects are modeled to provide a state-dependent constraint on the rotor flux magnitude. Load is treated as a function of the rotor position, which is appropriate for many mechanical system applications. State trajectories of the system that optimize machine efficiency are found by dynamic programming. Flux trajectories for the optimal solution are found to vary significantly over the machine cycle. The validity of the energy optimization is investigated experimentally on a feedforward, field-oriented induction machine.
IEEE Industry Applications Society Annual Meeting, pp.574 - 580