Loss minimization of induction machines during torque transients

Abstract

This article presents a new design of an energy optimization approach to induction motor drives. The proposed control strategy is based on a designed cost function given as a weighting sum of power and energy models and subjected to the dynamic of a low-order current-fed induction motor model, which considers only the dynamics of the rotor flux and the mechanical equation as constraints to the problem. Such a strategy uses the concept of off-line optimal closed-loop control. Considering the case of a random desired torque, the resolution with suboptimal conditions allows a time-varying rotor flux that converges to the optimal solution and minimizes energy in a dynamic operation. This solution is presented in an analytical form and is dedicated to optimizing a transient regime characterized by a random load torque. This command law will become invariant with respect to time and will asymptotically tend to a stationary optimal solution. Aiming to check the validity of the proposed algorithm, a comparison study is conducted between rotor-field-oriented control operating with a suboptimal rotor flux and rated constant-flux control. The simulation and experimental results obtained for a 1.5-kW laboratory induction motor demonstrate the effectiveness of the proposed strategy.