Performance analysis of an autonomous induction generator under different operating conditions using predictive control

This paper presents a Model Predictive Control scheme combined with rotor flux oriented vector control designed for an autonomous squirrel-cage induction generator driven by a variable speed wind turbine in a standalone system. The induction generator is excited using PWM inverter/rectifier connected to a single capacitor on the DC side. The model here adopted allows the saturation effect to be taken into account represented here by means of variable inductance, but in this paper a linear MPC is investigated, with different sampling time with respect to the plant. In particular, a fine tuning of MPC parameters, a special variable-step smoothing technique, a prefiltering action of rotor speed show the ability to regulate the DC bus voltage at a constant value fast and smoothly in presence of different wind disturbances and of different loads.