A Predictive Control Method for LCC HVDC Participated Black Start With Commutation Failure Prevention

Line-commutated converter (LCC) based HVDC systems can significantly speed up the restoration of infeed power systems in blackout states. But they are fragile to commutation failure (CF) under the adverse black-start conditions. To improve the adaptability of LCC HVDC systems in such situations, a predictive control method is proposed in this paper. It includes the linear discrete time predictive model, receding horizon regulator, state estimator, and CF prevention (CFPREV) module. The model is based on the quasi-dynamic and transient-state analysis of load disturbances, where the black-start state is forecasted according to voltage sags caused by inrush currents. The regulator operates in real-time to optimize reference inputs for the HVDC controller. In the estimator, process error and measurement noise are modeled and the optimal system state is estimated with a Kalman filter. Moreover, the dc current setting for CFPREV is calculated using the predicted and estimated state variables. The CFPREV module and regulator work in coordination for secure and fast dc power recovery under abnormal conditions. Small-signal stability analysis and EMT simulations are implemented on the CIGRE benchmark model based black start test grid. The results show that the proposed method i) improves dc power response to black start load startups and ii) presents better recovery performance after the fault.