TL;DR
This paper introduces a novel method for accurately and rapidly identifying commutation failures (CF) in LCC-HVDC systems by analyzing the energy stored in leakage inductance. It also proposes an improved fault current limiting strategy that considers the key variable influencing extinction angle recovery, leading to better mitigation of subsequent commutation failures (SCF) and enhanced system stability.
What Problem Does This Paper Address?
Line-commutated converter based high-voltage direct-current (LCC-HVDC) transmission systems are crucial for bulk power delivery. However, they are susceptible to commutation failures (CF), which can lead to subsequent commutation failures (SCF). SCFs can force the entire HVDC link to shut down, severely impacting power system stability. Existing CF identification methods often suffer from misjudgments or detection delays, limiting the effectiveness of mitigation strategies. Furthermore, previous approaches to suppress SCFs haven't adequately addressed the critical factors influencing the recovery of the extinction angle after a fault, hindering proactive system stabilization.
Methods
The authors propose a two-pronged approach:
- Improved CF Identification: Instead of relying on traditional metrics like extinction angle or voltage drops, this method leverages the energy stored in the leakage inductance of the converter transformer. The paper details the analysis of normal commutation processes and CF characteristics, demonstrating that this leakage inductance energy directly reflects the commutation state. A criterion based on the maximum leakage inductance energy across different phases is developed for accurate and timely CF detection.
- Fault Current Limiting Strategy for SCF Mitigation: This strategy focuses on a key variable that dictates the evolution of the extinction angle during the system's recovery phase after a fault. By considering this variable and excluding the distraction of commutation voltage, the authors propose a method to proactively manage the extinction angle. This aims to prevent SCFs by ensuring a stable and rapid recovery of the commutation voltage.
Datasets
The authors used the CIGRE LCC-HVDC benchmark model, a standard test network available in the PSCAD/EMTDC software, for their simulations. This model represents a monopolar 500 kV link with 12-pulse converters. No new datasets were introduced.
Key Insights and Ideas
- Leakage Inductance Energy as a CF Indicator: The core insight is that the energy stored in the leakage inductance is a more direct and sensitive indicator of the commutation state than traditional methods. This energy accumulates during the commutation overlap process, and its behavior changes significantly during a CF.
- Dynamic Extinction Angle Recovery: The paper highlights the importance of understanding and controlling the extinction angle's evolution during the post-fault recovery period. This dynamic behavior is crucial for preventing subsequent failures.
- Key Variable for Extinction Angle: Identifying and utilizing a specific "key variable" that directly influences extinction angle evolution allows for a more targeted and effective mitigation strategy.
- Robustness to Noise: The proposed CF identification method demonstrates robustness against measurement errors and signal noise, which is critical for real-world applications.
Key Results
- The proposed CF identification method achieved accurate and timely detection of commutation failures, outperforming traditional methods in simulations.
- The improved fault current limiting strategy successfully mitigated subsequent commutation failures (SCF), leading to a more stable and faster recovery of the DC voltage and current compared to conventional strategies.
- The proposed strategy showed a lower SCF probability across various fault levels and types (single-phase-to-ground and three-phase-to-ground faults).
- The method is robust to noise, maintaining its accuracy even with simulated noise levels up to 40 dB.
Limitations and Future Work
- Parameter Sensitivity: While the paper suggests a detection threshold ($k_{set} = 0.85$), it acknowledges that the specific value might need to be determined through dynamic simulation experiments for different HVDC projects.
- Future Work: The authors mention that future work will incorporate both the key variable of extinction angle and the cosine of the firing advance angle ($\cos\beta$) into the SCF mitigation strategy for further refinement.
Summary updated