适配柔直系统快速性及可靠性需求的雷击干扰识别方法

直流线路遭受雷击干扰易导致行波保护等灵敏性较高的主保护误动作。在主保护动作前，可靠区分雷击干扰和短路故障将有效提高主保护的安全性，保障直流系统的安全稳定运行。但对于柔直系统而言，现有雷击干扰识别方法动作速度不足，难以在主保护动作之前判定雷击/故障性质，且在因数据传输异常或噪声干扰条件下出现远偏离实际值的坏数据时，动作性能有待进一步提升。为此，提出了一种具有足够速动性及安全性的雷击干扰识别方法。首先揭示了反向线模电压行波的特征。然后根据所揭示的特征分别提出了基于小波熵相对变化率识别算法的判据以及一种结合中值滤波的线性拟合函数一次项系数大小的判据，二者协同工作实现不同雷击及故障叠加场景下雷击非故障场景的识别与主保护的闭锁，从而有效增强主保护的安全性。基于PSCAD/EMTDC的仿真结果表明，在短至1 ms的时窗内即可在全线范围内完成对雷击干扰的辨识，并具有良好的抗坏数据干扰能力。

Lightning interference identification method for the speed and reliability of a flexible DC system

Lightning strikes on DC transmission lines can easily cause a malfunction in main protection with high sensitivity such as traveling wave protection. Before the main protection is activated, a reliable distinction between lightning interference and short circuit failure will effectively improve the security of main protection and guarantee the security and stability of DC systems. However, for the VSC-HVDC, the existing lightning interference identification scheme is inadequate in terms of speed, and finds it difficult to recognize the lightning strike/fault before the main protection activates. Also the action performance under the random interference such as sampling abnormality needs to be improved. To solve the above problems, this paper proposes a novel lightning interference identification criterion acting with high speed and safety. First, this paper reveals the characteristics of the reverse line mode traveling wave, according to the characteristics mentioned, and proposes a relative change rate of wavelet entropy criterion and fit coefficient criterion based on linear fixed function. These changes collaborate to deal with different lightning strikes and faults in the field of lightning strike non-fault scenes and the lock of the main protection. They thereby effectively enhance the security of primary protection. Based on simulation in PSCAD/EMTDC, the method can identify lightning interference in 1 ms within all the lines, and is provided with good anti-bad data interference ability.This work is supported by the National Natural Science Foundation of China (No. 51907069).