Simplified sensorless predictive control of grid-side converter in wind energy conversion systems under distorted grid conditions

This paper proposes a new direct power control (DPC) strategy without voltage sensors for grid-connected pulse width modulation converters in wind energy conversion systems. This method relies on model predictive DPC (MPDPC) combined with virtual flux (VF) estimation. A novel VF estimator, based on an adaptive linear neuron (ADALINE) configured as a quadrature signal generator, is designed to eliminate the need for grid voltage sensors. In parallel, a simplified MPDPC approach is introduced, optimizing the selection process of the optimal control vector by reducing the number of equations involved. This simplification lowers computational demands while maintaining high control performance. The combination of the simplified MPDPC and the proposed VF estimator results in a sensorless voltage control strategy, referred to as VF-SMPDPC. The proposed strategy is validated through real-time simulations using the OPAL-RT system. A comprehensive comparative analysis demonstrates the superior performance of the ADALINE-based estimator over other VF estimation methods. Furthermore, the VF-SMPDPC strategy significantly outperforms conventional MPDPC, particularly under distorted grid conditions, proving its effectiveness and robustness in practical applications. • A novel sensorless predictive control strategy is developed for grid-connected converters. • An ADALINE-based virtual flux estimator eliminates the need for voltage sensors. • The proposed estimator achieves high accuracy and robustness under grid disturbances. • The proposed control scheme combines the VF estimator with a computationally efficient predictive control method. • Real-time OPAL-RT tests confirm that the proposed control scheme outperforms conventional MPDPC.