Advanced predictive direct power control for grid-connected WECS-PMSG: comprehensive simulation and real-time validation under stochastic wind conditions.

As wind energy continues to play a significant role in the global energy mix, ensuring the efficiency, robustness, and stable operation of Wind Energy Conversion Systems (WECS) has become a critical concern. This highlights the need for advanced control strategies to enhance power regulation, improve dynamic performance, and ensure reliable grid integration under variable wind conditions. Among the proposed approaches, Direct Power Control (DPC) has attracted significant attention for its simplicity and fast dynamic response, making it suitable for real-time implementation in grid-connected applications. However, despite its advantages, conventional DPC suffers from several limitations, including high power ripples, increased Total Harmonic Distortion (THD) in currents and voltages, and variable switching frequency, which can significantly affect overall system performance and power quality. To address these limitations, this paper proposes an Advanced Predictive Direct Power Control (APDPC) strategy for grid-connected WECS based on a permanent magnet synchronous generator (PMSG). The PMSG is selected for its high efficiency, compact structure, and operational flexibility, enabling robust and coordinated control of both the Generator-Side Rectifier and the Grid-Side Inverter. The proposed APDPC is implemented for both converters within a unified control framework, ensuring seamless power flow regulation. Unlike the conventional DPC approach, at each sampling period, three voltage vectors are directly selected based on the predicted evolution of active and reactive powers, without the need for complex control loops, hysteresis controllers, switching table, or exhaustive evaluation of all candidate switching states. The corresponding application times of these vectors are subsequently optimized using a multi-objective cost function that simultaneously minimizes the instantaneous errors in active and reactive powers on both sides of the system. To assess its effectiveness, the proposed control strategy was validated through MATLAB/Simulink® simulations under both constant and variable wind speed conditions. The results clearly demonstrate a substantial enhancement in dynamic performance compared with conventional DPC, including a 37% faster response, complete elimination of overshoot and undershoot, and a 61% reduction in steady-state error. In addition to these dynamic improvements, the proposed APDPC significantly enhances power quality, achieving reductions of approximately 40-45% in torque and power ripples, along with up to a 50% reduction in the THD of the grid current and grid voltage. These findings are further reinforced by comparative analysis with recent control strategies, highlighting its superior tracking accuracy and overall performance. Finally, real-time implementation on the dSPACE DS1104 platform confirms the practical feasibility and effectiveness of the proposed approach under realistic operating conditions.