摘要
提出了一种基于S变换模时频矩阵相似度的短时电能质量扰动分类方法。首先,建立各种扰动的标准模时频矩阵,然后计算扰动信号模时频矩阵与标准模时频矩阵的相似度,按照相似度最大的原则将扰动分类。该方法直接利用S变换结果,而不用增加其它算法或变换,原理简单、计算方便。仿真结果显示,该分类方法在不同噪声水平下均能达到满意的分类正确率,是一种有效的短时电能质量扰动分类方法。
A classification method for short duration power quality disturbance （SDPQD） based on similarity of module time-frequency matrixes by S-transform is proposed. At first, the standard module time-frequency matrixes of SDPQD for voltage dips, swells, interruptions, harmonics, oscillatory transients, notches and spikes are constructed, then the similarity of disturbance signal＇s module time-frequency matrixes to standard module time-frequency matrixes is calculated and according to the principle of maximum similarity the disturbances are classified. The proposed method can directly adopt the results of S-transform when other algorithms or transforms are not to be added, so it is simple in principle and easy to calculate. Simulation results show that with the proposed method a satisfied correctness of identification can be achieved, so it is an effective method to classify the SDPQD.
出处
《电网技术》
EI
CSCD
北大核心
2006年第5期67-71,共5页
Power System Technology
关键词
电能质量
S变换
扰动
模时频矩阵
相似度
分类
噪声鲁棒性
Power quality
S-transform
Disturbances
module time-frequency matrixes
Similarity
Classification
Noise robustness
作者简介
刘守亮（1981-）,男,硕士研究生,研究方向为电能质量与电力市场;
肖先勇（1968-）,男,副教授,从事电能质量分析与控制、电力市场等方面的教学和科研工作;
杨洪耕（1949-）,男,教授,从事电能质量分析与控制、电力市场等方面的教学和科研工作.
