Simulating Crowbar Circuit Engagement and Disengagement by Modifying Rotor-Side Converter Reference Voltage (MPPT)

In wind power system simulation models, adjusting the rotor-side converter's reference voltage can simulate the dynamic behavior of crowbar circuits without requiring actual IGBTs or ideal switching devices. This method simplifies the modeling process while effectively reflecting the impact of crowbar protection mechanisms on the system. The implementation typically involves conditional logic blocks in simulation environments like MATLAB/Simulink, where reference voltage values are programmatically modified based on grid fault detection signals.

Crowbar circuits are commonly used in low-voltage ride-through (LVRT) scenarios for doubly-fed induction generators (DFIGs), with their core function being rotor-side converter protection during grid voltage dips. Traditional implementations require physical switching devices, whereas this method equivalently achieves crowbar "engagement" (protection activation) and "disengagement" (control restoration) by directly modifying the reference voltage's magnitude or phase. For instance, forcing the reference voltage to zero or a limited value during faults simulates crowbar resistor insertion; restoring the original reference voltage after fault clearance corresponds to disengagement. Code implementation would involve creating a state machine that switches between normal operation and fault protection modes, using voltage threshold comparisons to trigger transitions.

This model incorporates maximum power point tracking (MPPT) and rotor current decoupling control strategies. The MPPT algorithm captures optimal wind energy by adjusting generator speed, while decoupling control ensures independent regulation of active/reactive power components in the rotor-side converter. Dynamic reference voltage adjustments must coordinate with both strategies: during engagement, MPPT is temporarily disabled to prioritize protection, while disengagement requires smooth transition back to normal control mode to avoid power oscillations. Algorithm implementation would include priority-based control switching logic and ramp functions for reference voltage recovery.

The advantage of this method lies in reduced model complexity, making it suitable for studying control algorithms' impact on system transient response. However, practical applications must consider non-ideal characteristics like hardware switching delays and losses, which could be simulated through additional time delay blocks and efficiency factors in the control code.