發(fā)布時(shí)間：2018-04-29 02:29

  本文選題：電力系統(tǒng) + 動(dòng)態(tài)狀態(tài)估計(jì)　； 參考：《電網(wǎng)技術(shù)》2014年01期

【摘要】：無跡卡爾曼濾波(unscented Kalman filter,UKF)是種非線性濾波方法。由于假設(shè)系統(tǒng)噪聲的方差為常數(shù),UKF的估計(jì)結(jié)果會(huì)受到未知系統(tǒng)噪聲的影響。為減小未知系統(tǒng)噪聲對動(dòng)態(tài)狀態(tài)估計(jì)的影響,提出了種改進(jìn)的自適應(yīng)UKF(adaptive unscented Kalman filter,AUKF)算法。該方法通過在UKF中引入漸消記憶指數(shù)加權(quán)的Sage-Husa噪聲統(tǒng)計(jì)估值器,能夠估計(jì)時(shí)變系統(tǒng)噪聲的均值和方差。利用IEEE57和IEEE 118測試系統(tǒng),在典型日負(fù)荷條件下對AUKF方法的有效性進(jìn)行仿真驗(yàn)證,結(jié)果表明所提出的AUKF算法與傳統(tǒng)UKF方法相比,在不增加計(jì)算復(fù)雜度的同時(shí),能夠提高狀態(tài)估計(jì)精度。
[Abstract]:Unscented Kalman filter (UKF) is a nonlinear filtering method. The estimation results of the system noise assuming that the variance of the system noise is constant are affected by the unknown system noise. In order to reduce the effect of unknown system noise on dynamic state estimation, an improved adaptive UKF(adaptive unscented Kalman filter algorithm is proposed. This method can estimate the mean and variance of noise in time-varying system by introducing a Sage-Husa noise statistical estimator weighted by fading memory exponent in UKF. Using IEEE57 and IEEE 118 test system, the validity of AUKF method is simulated under typical daily load conditions. The results show that the proposed AUKF algorithm has no computational complexity compared with the traditional UKF method. It can improve the accuracy of state estimation.
【作者單位】： 華北電力大學(xué)電氣與電子工程學(xué)院;
【基金】：國家自然科學(xué)基金項(xiàng)目(51077053,51277074)~~
【分類號】：TN713;TM711

【參考文獻(xiàn)】

相關(guān)期刊論文 前8條

1 李大路;李蕊;孫元章;;混合量測下基于UKF的電力系統(tǒng)動(dòng)態(tài)狀態(tài)估計(jì)[J];電力系統(tǒng)自動(dòng)化;2010年17期

2 李欽;項(xiàng)鳳雛;顏偉;盧建剛;余娟;陳俊;李世明;;基于SCADA及PMU多時(shí)段量測信息的獨(dú)立線路參數(shù)估計(jì)方法[J];電網(wǎng)技術(shù);2011年02期

3 傅書，

本文編號：1817999