Predictive-Control-Based Direct Power Control With an Adaptive Parameter Identification Technique for Improved AFE Performance

Abstract

This paper proposes predictive-control-based direct power control (DPC) with an adaptive online parameter identification technique for ac-dc active front ends (AFEs) to overcome model mismatch and parameter uncertainty. Based on least-squares estimation, the input inductance and input resistance of the AFE are calculated every sampling period using sampled input currents and input voltages. Because the online-tuned input inductance and resistance are updated for the controller, the proposed predictive-control-based DPC with the adaptive parameter identification technique can mitigate performance degradation resulting from the model uncertainty of the model predictive controller without any additional sensors. Therefore, the AFE generates sinusoidal input currents with a unity power factor despite parameter uncertainty.