中文摘要：

电能质量扰动检测识别对电能质量的监测和治理改善都具有重要作用。为更好地识别电能质量扰动,提出了一种基于关联向量机和S变换的电能质量扰动识别方法。首先,通过S变换提取正弦信号、谐波、电压波动、电压暂降、电压暂升、暂态振荡、谐波暂降及谐波暂升等9种电能质量扰动的主要特征,然后用关联向量机对特征样本进行训练及分类。算例结果表明,该方法能有效地识别出电能质量扰动信号类型,识别时间短,且正确率极高,达98.8%,是应用于实时电能质量监测工程实际的很好选择。

英文摘要：

Classification of power quality（PQ）disturbances is significant for monitoring and compensation of PQ.A new approach based on relevance vector machine（RVM）and S-transform is presented to classify the PQ disturbances.Features of sine,harmonic,voltage fluctuation,voltage sag,voltage swell,oscillatory transient,voltage sag with harmonic and voltage swell with harmonic are extracted by S-transform first.Then RVM is applied for classification of the PQ disturbances.The results of example show that the approach based on S-transform and RVM can effectively classify the PQ disturbances,and it not only requires very short test time,but also has very high correctness,reaching 98.8%.The proposed method provides a good way for real-time PQ monitor system in practice.