The hybrid cascaded (HC) LCC-MMC HVDC scheme, wherein the LCC serves as the rectifier, and a single LCC is connected in series with multiple MMCs in parallel as the inverter, amalgamates the merits of both LCC and MMC technologies. However, the utilization of thyristors in the LCC inverter still introduces a potential risk of commutation failure (CF). To mitigate this concern and bolster the stability of the hybrid system, an analysis of the LCC CF and its recovery process is undertaken, along with a mathematical investigation of the LCC's reactive power (RP) characteristics in the inverter. Subsequently, a mitigation strategy rooted in RP feedback control (RPFC) for the subsequent CF (SCF) within the HC-HVDC system is proposed. The transient performance of the system under AC fault conditions is evaluated with the proposed control approach in comparison to conventional methods. The outcomes indicate that this proposed approach effectively harnesses the RP support capabilities of MMCs, thereby endowing the LCC with enhanced resistance to potential SCF and improved post-fault recovery performance for the overall system.

In a multi-outfeed HVDC system, rectifier stations are closely spaced and tightly coupled, which facilitates the propagation of disturbances to adjacent HVDC links. When a commutation failure (CF) happens in a faulted HVDC system, the resulting transient overvoltage (TOV) at the sending end (SE) can cause a voltage rise in an adjacent unfaulted HVDC system, potentially triggering cascading CF and thereby expanding the fault impact. To mitigate these risks, a refined optimization method is developed in this study for the regulation of rectifier-side TOV, specifically incorporating the immunity constraints of adjacent links against cascading CF. First, the impact and evolution process of how TOV at the SE of a faulted HVDC system leads to cascading CF in adjacent unfaulted HVDC links are analyzed, and the mathematical boundary for rectifier-side bus potential is established, which serves to prevent CF in unfaulted HVDC system. Accordingly, an optimized DC current regulation scheme is proposed, aiming to constrain transient voltage rises while ensuring the commutation security of adjacent HVDC links. Finally, simulation studies based on a multi-outfeed HVDC system model in PSCAD/EMTDC are carried out to demonstrate the performance of the proposed method.

Commutation failure (CF) can cause DC current surges, impaired system transmission power and even leading to HVDC blocking. The prediction of CF combined with early control triggering is beneficial for avoiding CFs. However, current analyses of CF identification mainly focus on real-time detection or prediction based on limited factors, which are insufficient for reliable CF prediction. Therefore, a CF prediction method for the line-commutated converter high-voltage direct current (LCC HVDC) system based on CNN-LSTM-Attention is proposed. Firstly, key indicators that represent the commutation state are selected by analyzing the factors that influencing CF. Secondly, a CF prediction model is constructed based on CNN-LSTM Attention. Additionally, the input data sampling principle is clarified and the commutation state discrimination component based on DC current variation is built for the model training. On this basis, a method for predicting CF is proposed. Finally, the accuracy and efficiency of the proposed method is validated based on the CIGRE LCC-HVDC Benchmark system and the Bin-Jin HVDC system in China.