改進(jìn)的偏最小二乘法在分析諧波源中的應(yīng)用 

王富勇，李麗 
泉州電力技能研究院，福建 泉州 362000 
 

摘 要： 介紹了改進(jìn)的偏最小二乘法的基本原理和建模思想，并給出了建模步驟，論證了該方法的合理性。將該方法應(yīng)用于諧波源識(shí)別的分析中，對(duì)諧波阻抗和諧波發(fā)射水平進(jìn)行了定量分析，仿真結(jié)果表明該方法能夠避免因變量和自變量間多重關(guān)系，可以更準(zhǔn)確地識(shí)別諧波所在側(cè)，是電力系統(tǒng)數(shù)據(jù)建模的一種新方法。
關(guān)鍵詞： 偏最小二乘回歸；改進(jìn)的偏最小二乘法；方差；相關(guān)系數(shù)；解釋能力
中圖分類號(hào)：TM712 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1007-3175(2013)05-0039-04

Application of Improved Partial Least Squares Regression in Harmonic Source Analysis 

WANG Fu-yong, LI Li 
Quanzhou Electric Technical Research Institute, Quanzhou 362000, China 
 

Abstract: Introduction was made to the basic principle and modeling thought of the improved partial least squares regression method and the steps of modeling were given. This paper demonstrated the rationality of this method and used the method to analyze harmonic source discrimination, carrying out quantitative analysis of harmonic impedance and emission level. Simulation result shows that this method could avoid the multirelation between dependent variables and independent variables and identifies the locality side of harmonics more correctly. It is a new method for modeling data of electric system.
Key words: partial least squares regression; improved partial least squares regression; variance; correlation coefficient; explanatory ability

參考文獻(xiàn)
[1] Hamzah N, Mohamed A, Hussain A.Methods for determining utility and customer harmonic
contribution sat the point of common coupling[C]//National Power Engineering Conference, Bangi, Malaysia, 2003：167-171.
[2] 劉秀玲，王洪瑞，王永青，等. 臨界阻抗法在配電網(wǎng)諧波源探測(cè)中的應(yīng)用[J]. 電力系統(tǒng)自動(dòng)化，2007，31(14)：76-80.
[3] Yacamini R. Power system harmonic, part 4 interharmonics[J]. Power Engineering Journal,1996, 10(4)：185-193.
[4] Tilte G, Grady W M, Powers E J, et al. Introduction of the concept of friendliness power to characterize the harmonic emission of nonlinear loads[C]//Electric Ship Technologies Symposium, Philadelphia, 2005：306-312.
[5] Xiao X Y, Yang H G.The method of estimating customers harmonic emission level based on bilinear regression[C]//IEEE International Conference on Electric Utility Deregulation: Restructuring and Power Technologies, Hong Kong, 2004, 2：662-665.
[6] 車權(quán)，楊洪耕.基于穩(wěn)健回歸的諧波發(fā)射水平估計(jì)方法[J].中國(guó)電機(jī)工程學(xué)報(bào)，2004，24(4)：39-44.
[7] 黃舜，徐永海.基于偏最小二乘回歸的系統(tǒng)諧波阻抗與諧波發(fā)射水平的評(píng)估方法[J].中國(guó)電機(jī)工程學(xué)報(bào)，2007，27(1)：93-97.
[8] Li Chun, Xu Wilsun, Tayjasanant T. A criticalimpedance based method for identifying harmonic sources[J]．IEEE Transactions on Power Delivery, 2004, 19(2)：671-678．
[9] Jiang Xiao hua , King Ji , EmadiAli . A Power Harmonics Detection Approach Based on the Least Squares Energy Minimization Principle[C]//The 30th Annual Conference of the IEEE Industrial Electronics Society, 2004, 3：2934-2938.
[10] 王惠文，吳載斌，孟潔. 偏最小二乘回歸方法的線性與非線性方法[M]. 北京：國(guó)防大學(xué)出版社，2006.
[11] 林敏，杜光年，劉志斌. 偏最小二乘回歸方法的局限性及改進(jìn)算法[J]. 數(shù)學(xué)的實(shí)踐與認(rèn)識(shí)，2008，38(9)：72-76.