中文摘要：

针对均值滤波对突变点敏感及传统自适应维纳滤波所取邻域的局限性，给出了一种新的电能质量检测去噪算法。该算法首先通过均值滤波去除噪声从而减小噪声在维纳滤波参数估计的影响；再对纳滤波结果进行阈值处理，从而提取突变点信息。实验结果表明，该算法不仅能去除大量的噪声而且能很好地保留突变点信息。

英文摘要：

Mean filter is sensitive to brake points . The parameters of the traditional adaptive Wiener filter are estimated from neighborhood. Because of the limit of the neighborhood, the problem of the estimation accuracy arises. Aimed to resolve the problem , this paper presented a new de-noising method of power quality detection . The method cut down the influence of the noises to the estimate of the parameter of Wiener filtering by meaning filtering first. To keep the brake points , thresh filtering was operated on the result of the Wiener filtering . Experiment indicates that this method not only de-noise well but also keep the brake point nicely.