Electronic transformers (ECT) need long-term operation, and often operate in formidable and complex conditions. So it is difficult in the process of ECT application of field calibration. A new method of field calibration is presented to solve the problem. Based on the de-noising preprocessing of sample data, this paper establishes the ARMA model, and adopts the frequency bands division-based singular value decomposition total least squares (SVDTLS) algorithm to estimate the models' frequencies, and extracts amplitudes and phases of signal, which can suppress the interference of filed random noise and harmonics waves efficiently. The experimental results indicate that this method with high precision can meet the requirements of field calibrating of 0.2 class ECT.%由于电子式互感器需要长期运行的工作特性以及所处复杂恶劣的工况环境,现场校准成为电子式互感器应用过程中的难题.提出一种电子式互感器现场校准算法,在对采样数据进行降噪预处理的基础上,采用基于频段划分的奇异值总体最小二乘算法对建立的ARMA模型进行频率估计,再提取信号幅值和相位等特征参量,可以有效抑制现场随机噪声和谐波的干扰.试验结果表明,该算法具有较高的计算精度,可以满足0.2级电子式互感器现场校准要求.
展开▼
