郑凌蔚;刘士荣;谢小高;

• 1:杭州电子科技大学自动化研究所

摘要(Abstract)：

将光伏发电系统看成基于气象参数的非线性黑箱模型,用非线性自回归外推模型对不同天气条件下的光伏发电系统进行辨识。采用了对系统维数不敏感的基于方差分析展开的改进小波神经网络对系统进行非线性自回归外推模型辨识,辨识数据和验证数据均取自实际光伏发电系统。实例研究结果表明:与Sigmoid网络函数法、树分割法及基本小波神经网络法相比,基于改进小波神经网络的非线性自回归外推模型能更好地反应各种不同天气条件下光伏发电系统的动态行为;天气波动的剧烈程度对辨识效果影响较大。

关键词(KeyWords)： 光伏发电系统;非线性自回归外推;模型辨识;改进小波神经网络;方差分析

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(51007015);; 浙江省重大科技专项(2009C11G2040039)~~

作者(Author): 郑凌蔚;刘士荣;谢小高;

Email:

DOI: 10.13335/j.1000-3673.pst.2011.10.035

参考文献(References)：

• [1]Villalva M,Gazoli J,Filho E.Comprehensive approach to modelingand simulation of photovoltaic arrays[J].IEEE Trans on PowerElectronics,2009,24(5):1198-1208.
• [2]Ropp M,Gonzalez S.Development of a MATLAB/Simulink modelof a single-phase grid-connected photovoltaic system[J].IEEE Transon Energy Conversion,2009,24(1):195-202..
• [3]徐青山,卞海红,高山,等.计及旁路二极管效应的太阳能模组能评估[J].中国电机工程学报,2009,29(8):103-108.Xu Qingshan,Bian Haihong,Gao Shan,et al.Performance evaluationof photovoltaic module with bypass diode effect[J].Proceedings ofthe CSEE,2009,29(8):103-108(in Chinese).
• [4]Alemeida R,Busquet S,Camez C,et al.Large scale intergration ofmicro-generation to low voltage grids[EB/OL].[2003].http//www.microgrid.eu/micro2000/delivarables/Deliverable_DA1.pdf.
• [5]翟载腾,程晓舫,丁金磊,等.最大功率条件下串联太阳电池电流方程的确定[J].中国电机工程学报,2007,27(14):87-90.Zhai Zaiteng,Cheng Xiaofang,Ding Jinlei,et al.Determination ofserial solar panel’s current equation based on maximum power[J].Proceedings of the CSEE,2007,27(14):87-90(in Chinese).
• [6]翟载腾,程晓舫,杨臧健,等.太阳电池一般电流模型参数的解析解[J].太阳能学报,2009,30(8):1078-1082.Zhai Zaiteng,Cheng Xiaofang,Yang Zangjian,et al.Analyticsolutions for solar cell model parameters[J].ACTA Energiae SolarisSinica,2009,30(8):1078-1082(in Chinese).
• [7]焦阳,宋强,刘文华.光伏电池实用仿真模型及光伏发电系统仿真[J].电网技术,2010,34(11):198-203.Jiao Yang,Song Qiang,Liu Wenhua.Practical simulation model ofphotovoltaic cells in photovoltaic generation system andsimulation[J].Power System Technology,2010,34(11):198-203(inChinese).
• [8]刘伟,彭冬,卜广全,等.光伏发电接入智能配电网后的系统问题综述[J].电网技术,2009,33(19):1-6.Liu Wei,Peng Dong,Bu Guangquan,et al.A survey on systemproblems in smart distribution network with grid-connectedphotovoltaic generation[J].Power System Technology,2009,33(19):1-6(in Chinese).
• [9]González-Morán C,Arboleya P,Reigosa D,et al.Improved modelof photovoltaic sources considering ambient temperature and solarirradiation[C]//IEEE PES/IAS Conference on Sustainable AlternativeEnergy,Valencia,Spain:IEEE,2009.
• [10]卢继平,白树华.风光氢联合式独立发电系统的建模及仿真[J].电网技术,2007,31(22):75-80.Lu Jiping,Bai Shuhua.Model and simulation of conjoint independentpower generation system consisting of power generation by windenergy,solar energy and hydrogen energy[J].Power SystemTechnology,2007,31(22):75-80(in Chinese).
• [11]Zhu Yucai.Estimation of nonlinear ARX models[C]//Proceedings ofthe 41st IEEE Conference on Decision and Control.LasVegas,USA:IEEE,2002.
• [12]Al-Smadi A,Wilkes D.Parameters and order estimation fornon-Gaussian ARMA processes[C]//Proceedings of the IEEEConference on Bringing Together Education,Science andTechnology.Tampa,USA:IEEE,1996.
• [13]Rahim H,Ibrahim F,Taib M,et al.Model order selection criterionfor monitoring haemoglobin status in dengue patients using ARXmodel[J].International Journal on Smart Sensing and IntelligentSystems,2008,1(2):403-319.
• [14]Chen S,Wang X,Harris C.NARX-based nonlinear systemidentification using orthogonal least squares basis hunting[J].IEEETrans on Control Systems Technology,2008,16(1):78-84.
• [15]Zhang Q,Benveniste A.Wavelet networks[J].IEEE Transactions onNeural Networks,1992,3(6):889-898.
• [16]Zhang J,Walter G.Wavelet neural networks for function learning[J].IEEE Trans on Signal Processing,1995,43(6):1485-1497.
• [17]Ramesh K,Shukor S,Aziz N.Development of sigmoidnet basedNARX model for a distillation column[J].Chemical Product andProcess Modeling,2008,3(2):1-14.
• [18]Siegelmann H,Horne B,Giles C.Computational capabilities ofrecurrent NARX neural networks[J].IEEE Trans on Systems,Manand Cybernetics,Part B:Cybernetics,1997,27(2):208-215.
• [19]Billings S,Wei Hualiang.A new class of wavelet networks fornonlinear system identification[J].IEEE Trans on Neural Networks,2006,16(4):862-874.
• [20]Wei H,Billings S,Liu J.Term and variable selection for nonlinearsystem identification[J].International Journal of Control,2004,77(1):86-110.
• [21]Billings S,Chen M,Korenberg J.Identification of MIMO nonlinearsystems using a forward regression orthogonal estimator[J].International Journal of Control,1989,49(6):2157-2189.
• [22]Chen S,Billings S,Luo W.Orthogonal least squares methods andtheir application to nonlinear system identification[J].InternationalJournal of Control,1989,50(5):1873-1896.
• [23]郑凌蔚,刘士荣,徐青鹤.基于非线性扩散粒子群算法的光伏微网并网点恒定潮流控制[J].电网技术,2010,34(10):152-156.Zheng Lingwei,Liu Shirong,Xu Qinghe.Constant power flowcontrol at grid-connected point of photovoltaic microgrid based onnonlinear diffusion particle swarm optimization[J].Power SystemTechnology,2010,34(10):152-156(in Chinese).