High order accumulation based classification of transient power quality disturbances by mahalanobis distance
Xiao Xian · Dianli kexue yu jishu xuebao · 2013
A novel classification and recognition algorithm for transient power quality disturbance signals,which is based on high order accumulation and Mahalanobis distance,was proposed to complete the classification of impulse transient disturbances and oscillation transient disturbances in this paper.The proposed method extracted high order statistical characteristics of transient disturbances with high order accumulation and selected the maximum and the minimum of 2-order,3-order and 4-order accumulation as the 6D feature vector of disturbances.Then the Mahalanobis distance between feature vector of test sample and the standard template was calculated.The classification of two kinds of disturbances was completed with the calculation results according to the minimum distance criterion.Simulation results showed that low order accumulation could not be used independently in the classification of transient disturbances.The higher the dimension of the feature vector is,the more efficient the classification progress is.The principle of the proposed method is simple.It is an effective method to classify transient power quality disturbances with high accuracy.