Search

• An overview of state-of-the-art commutation failure (CF) is provided. • A review of the impact of multi-infeed structure on CF is undertaken. • The advantages and limitations of different CF prevention methods are examined. • Suggestions and potentially important future research directions are put forward. [Display omitted] Commutation failure (CF) is the main fault type in high voltage direct current (HVDC) systems with line commutated converters. To mitigate CFs, considerable researches have been done either in modelling or control methods. However, with the operation of more and more large-capacity HVDC projects, the coupling characteristics between HVDC and power grid are more complex. The existing reviews tend to focus only on the CFs of a single HVDC itself. This paper provides a more systematic overview of CFs, covering fifteen specific aspects in six areas, including four types of system structures, such as multi-infeed HVDC, and various types of CFs. The mechanism, discrimination, modeling, and prevention of CFs are reviewed extensively in combination with the latest literature. In particular, the advantages and limits of various CFs prevention methods are analyzed and compared in order to reveal the latest research results. Finally, it points out that the analysis and control of multi-infeed HVDC and hierarchical infeed HVDC, the application of artificial intelligence (AI) technology, the improvement of communication-based on 5G technology, harmonic problems, and control and coordination strategies of new auxiliary equipment are challenges and research directions of HVDC CFs in the future. [ABSTRACT FROM AUTHOR]

With increasing wind farm scale, dynamic modeling and simulating the low-voltage ride-through (LVRT) process of wind farms are time- and source-consuming. Thus, a dynamic equivalent model of a wind farm considering the LVRT process is necessary and vital for power system analysis. However, the crowbar status of each doubly-fed induction generator (DFIG), voltage, and wind speed differences among will influence the LVRT process and equivalent results. To consider these three factors comprehensively and eliminate the error, this paper proposes a three-stage dynamic equivalent modeling approach, using accurate crowbar identification and voltage differences among DFIGs. Stage I: DFIGs are divided by crowbar status. A new accurate crowbar identification method is proposed to consider the interference from other DFIGs, using the experimental two-machine system and several dynamic simulations to correct fault voltage. Stage II: The often-overlooked voltage differences are considered, and DFIGs are divided into different clusters for accurate fault dynamics. Stage III: To eliminate errors resulting from wind speed differences, the DFIG active power reference is corrected based on the analytical function of the wind farm LVRT process. The proposed model can accurately identify crowbar status and consider the voltage and wind speed differences without detailed model simulation results, improving the accuracy. Case studies are undertaken on a wind farm with fifty DFIGs. Simulation results indicated that the proposed model can match the detailed model under different wind scenarios and fault voltages, and it also increases simulation efficiency, which is suitable for studying the effect of LVRT on power systems. [Display omitted] • A three-stage dynamic equivalent method is proposed to model the wind farm. • Fault voltage is corrected to identify crowbar status given DFIGs' mutual impact. • Voltage differences among wind turbines are considered to reduce equivalent error. • The simulation results demonstrate the effectiveness of proposed equivalent method. • The proposed model can capture the LVRT process of large-scale wind farm. [ABSTRACT FROM AUTHOR]