湖南电力
湖南電力
호남전력
HUNAN ELECTRIC POWER
2015年
2期
21-24
,共4页
配电网%智能分类识别%SVM%小波包时间熵
配電網%智能分類識彆%SVM%小波包時間熵
배전망%지능분류식별%SVM%소파포시간적
distribution network%intelligent classification%SVM%Wavelet-packet Time Entropy
文中提出了一种基于小波包时间熵与支持向量机( SVM )的配电网运行状态智能识别方法。仿真结果表明,该配电网运行状态智能分类识别方法中选取的小波包时间熵特征量维数相对较低,抗干扰能力强,分类识别所需训练样本少,分类速度快,识别准确率高,可较好应用于配电网运行状态的在线智能识别。
文中提齣瞭一種基于小波包時間熵與支持嚮量機( SVM )的配電網運行狀態智能識彆方法。倣真結果錶明,該配電網運行狀態智能分類識彆方法中選取的小波包時間熵特徵量維數相對較低,抗榦擾能力彊,分類識彆所需訓練樣本少,分類速度快,識彆準確率高,可較好應用于配電網運行狀態的在線智能識彆。
문중제출료일충기우소파포시간적여지지향량궤( SVM )적배전망운행상태지능식별방법。방진결과표명,해배전망운행상태지능분류식별방법중선취적소파포시간적특정량유수상대교저,항간우능력강,분류식별소수훈련양본소,분류속도쾌,식별준학솔고,가교호응용우배전망운행상태적재선지능식별。
This paper proposes an intelligent identification method for distribution network operation state based on wavelet-packet time entropy and SVM. The simulation results show that the method has good performance on accurate state identification with low dimension of eigenvector, good anti-interference ability, less training samples and high classification speed. Also it can be well used in online application of intelligent identification for distribution network operation state.