一种电压暂降检测新方法的研究

Study for a New Method of Detecting Voltage Sags

针对电网中电压暂降是引起电能质量问题的主要原因，本文提出了一种新的电压暂降检测方法——协方差与自回归模型结合法。在应用过程中，首先采用二自回归（AR）模型对仿真出来的电压信号进行建模，再利用协方差方法对信号进行提取模型参数，然后计算表征距离测度的欧氏距离平方（d2），再通过比较相应的欧氏距离平方是否大于一个阀值来判断回路中是否有电压暂降的发生。根据对大量的电压暂降事故数据收集并进行分析表明，该方法都能够快速准确的检测是否产生电压暂降故障。

The voltage sag is one of the main causes of power quality problems. Aiming to this problem, this paper proposes a new voltage sag detection method which combines covariance and auto regression model. The process of the method is as follows： Building a 2-order AR model for the voltage signals simulated; extracting the AR model parameters by using the covariance method; calculating the distance measurement characterized by Euclidean distance squares （d2）; and then comparing whether the corresponding Euclidean distance squares are greater than a threshold or not to determine whether there occurs a sag. This paper takes a large number of accident data and analyses, showing that the method can detect voltage sag fast and accurately.