<![CDATA[Wind turbine emulation faces significant challenges in achieving high dynamic performance while maintaining low-cost and sensorless control architectures suitable for laboratory validation. This paper proposes a software-based wind turbine simulator replicating the dynamic behavior of a 3 kW wind turbine under realistic wind conditions, including quasi-stationary, stochastic, and localized gust (Mexican Hat) profiles. The emulator is implemented using a three-phase induction motor driven by a four-switch voltage source inverter (4S-VSI), controlled via rotor field-oriented control and space vector modulation. A sliding-mode observer (SMO) is employed to estimate rotor speed and flux from stator current measurements, eliminating mechanical sensors. Additionally, an adaptive parameter estimator based on the reactive power method is incorporated into the control loop to identify the rotor resistance in real time. Under nominal loaded operation, the proposed scheme achieves speed tracking errors below 1%, torque errors below 6%, and rotor flux errors below 2% across all wind profiles. When a severe +100% rotor resistance variation is introduced, speed deviation reaches 10% and torque error approaches 20% prior to adaptation, while estimated quantities remain stable, demonstrating observer robustness. Once the reactive power–based adaptation is activated, speed error returns to nearly zero, torque error falls below 5%, stator current error remains under 3%, and flux deviation becomes negligible. The maximum observed speed overshoot under gust excitation is 13.27%, with a settling time of 0.31 s. A quantitative comparison with a conventional six-switch VSI shows that the proposed 4S-VSI reduces switching activity by approximately 43% (from 44.89 kHz to 25.72 kHz equivalent switching frequency), leading to lower switching losses and reduced hardware complexity without compromising dynamic performance. These results demonstrate that the proposed architecture achieves robust observer convergence, accurate wind profile emulation, and significant converter loss reduction, providing a cost-effective and computationally efficient platform for real-time validation of wind energy conversion systems.]]>
