【摘要】 为提高电压扰动信号分类识别的精度,提出了一种基于数学形态学与动态时间扭曲的新算法.该算法首先通过形态滤波器对信号进行滤波处理,然后利用dq变换提取滤波输出的特征,再通过动态时间扭曲分类器与参考模板进行匹配,最后获得有效的分类识别结果.用Matlab进行仿真分析的结果表明,该算法能有效识别各类扰动信号,准确率高,即使在强噪声环境下,识别精度也超过84%.更多还原

【Abstract】 A novel algorithm based on the mathematical morphology and the dynamic time warping was proposed for improving the accuracy of voltage disturbance classification.In this algorithm,firstly,a morphological filter is utilized for disturbance signals’ filtering;secondly,feature extraction of the filtered signals is done by a d-q transform;finally,the classification results can be obtained by template matching through a dynamic time warping classifier.A simulation was done in Matlab.The simulation result shows that the proposed algorithm has a classification accuracy of over 84% even in a strong noise environment.更多还原