电力市场运营模式研究及其出清电价预测
|
摘要
当今世界,电力工业体制正在发生深刻的变革,打破垄断、引入竞争,建立统一、规范、开放、有序、竞争的电力市场,成为发展的必然趋势。随着世界各国电力工业改革的发展趋势,我国于上世纪90年代也开始了以打破垄断、引入竞争、放松管制为目标的电力市场化改革。如何合理制定相应的运营模式以及怎样根据电力市场的相关历史数据准确的预测出未来的市场出清电价,对于市场中的各个参与者都具有十分重要的意义。
本文首先围绕建立区域电力市场交易模式的问题,进行了深入研究。在分析比较了世界上几个典型的电力市场交易模式后,提出了我国区域电力市场交易模式的选择原则,并对我国未来电力市场发展的方向和市场运营模式及可能实施的双边交易方式进行了分析与论述,为我国区域电力市场设计了批发竞争的市场模式。旨在对我国电力市场改革方向与运营方式及在未来市场中为参与者的最优竞价策略进行探索性研究。
其次本文介绍了电力市场出清电价结算机制,论述了出清电价的形成过程,并分析了出清电价的影响因素。继而根据实际电力市场的出清电价数据对其变化特点和分布特征进行了研究,得出不同时段的出清电价时间序列特性不同,具有不同的变化趋势,提出了分时段预测出清电价的思路。接着对当前较常见的预测方法的优缺点进行了探讨后,提出了出清电价的预测模型。为了提高电价预测的精度,本文提出了以下两个方面的改进:1、在输入因素的选取上,选取更能表征或影响电价变化的因素作为输入向量;2、在输入因素预处理上,对输入向量进行处理使其更有规律性,更能表征电价的变化。
本文采用最小二乘支持向量机建立了两个预测模型:1、利用相似搜索技术来生成训练集和输入变量,并对次日电价进行实例预测,取得了较好的效果。预测结果表明采用这种模型预测采用相似搜索技术处理后,预测的效果比传统的方法有较大的提高。2、提出采用小波变换和最小二乘支持向量机相结合的电价预测模型。选择合适的小波基和分解尺度,将电价序列分解成一个低频分量和多个高频分量,从而进行电价特征的提取。利用所获得的电价各分量,对电价各分量分别建立最小二乘支持向量机预测模型,最后汇总叠加各分量预测结果得到最终的预测电价。实验结果表明,此模型显著的提高了出清电价预测的精度。
In today's world,a profound transform of electric power industry mechanisms is taking place. Its inevitable trend is breaking monopoly,inducing competition and setting up a uniform, normative,open,orderd and competitive electricity market. China attempted to have a reformation of electric power market in 1990s,breaking the traditional integrative management mode,and carrying out separation of net and plant, transition and distribution. How to establish corresponding elecricity market operation mode rationally and how to use the relative historic data to forecast the future market clearing electricity price is a very meaningful work for every participator in the power market.
Firstly,The problem of establishing regional electric power market exchange pattern is researched thoroughly in the paper. After compared of several typical electric power market exchange patterns worldwide,the principle of regional power market trading mode is put forward. In this paper, China power market's orientation, structure and possible bilateral exchange manner are analyzed and discussed. And the regional power markets of wholesale competition model in China is designed in the paper. The purpose of this paper is to exploratory research the reformational orientation and the operational manner of China power market and optimal bidding strategies for participants in future market.
Secondly,The electric power market pricing mechanism is introduced, and the establishment and the influence factors of the electricity price are discussed. According to actual market clearing price's data,vary and distributing character of MCP data is researched.we obtained that different period data of MCP have different time series characteristic,the change of MCP have different trend. So the forecasting MCP idea based on period of time is put forward. Then, analyzed and compared the adventages and disadventages of some current forecasting methods, the forecasting models of market clearing price is proposed. To improve the accurary of price forecasting,two aspects of improvements are proposed:(1)selector of the factors which can show the change of the price sequence accurately;(2)the pretreatment of the input variables.
Two price forecasting model using Support Vector Machine is proposed: (1)similarity searching technique is applied to building the training set and input variables .The electricity price of next day is forecasted successfully by this model. The precision of forecasting based on similarity searching technique is higher than traditional methods.(2)The forecasting price model based on Wavelet Transform and Support Vector Machine is proposed. Selected of appropriate wavelet basis and decomposition scale, the market clearing price series is decomposed into one low-frequency and some high-frequency sub-series in the wavelet domain,So MCP feature is extracted form the market clearing price series. Using each new component of the original electricity price signal, LSSVM forecasting models for each sub-series is independently set up, then adding up to the forecasting results of all sub-series, the whole forecasted electricity price series are obtained. The results of forecasting show that this model impove the forecasting accurate of MCP remarkablly.
本文首先围绕建立区域电力市场交易模式的问题,进行了深入研究。在分析比较了世界上几个典型的电力市场交易模式后,提出了我国区域电力市场交易模式的选择原则,并对我国未来电力市场发展的方向和市场运营模式及可能实施的双边交易方式进行了分析与论述,为我国区域电力市场设计了批发竞争的市场模式。旨在对我国电力市场改革方向与运营方式及在未来市场中为参与者的最优竞价策略进行探索性研究。
其次本文介绍了电力市场出清电价结算机制,论述了出清电价的形成过程,并分析了出清电价的影响因素。继而根据实际电力市场的出清电价数据对其变化特点和分布特征进行了研究,得出不同时段的出清电价时间序列特性不同,具有不同的变化趋势,提出了分时段预测出清电价的思路。接着对当前较常见的预测方法的优缺点进行了探讨后,提出了出清电价的预测模型。为了提高电价预测的精度,本文提出了以下两个方面的改进:1、在输入因素的选取上,选取更能表征或影响电价变化的因素作为输入向量;2、在输入因素预处理上,对输入向量进行处理使其更有规律性,更能表征电价的变化。
本文采用最小二乘支持向量机建立了两个预测模型:1、利用相似搜索技术来生成训练集和输入变量,并对次日电价进行实例预测,取得了较好的效果。预测结果表明采用这种模型预测采用相似搜索技术处理后,预测的效果比传统的方法有较大的提高。2、提出采用小波变换和最小二乘支持向量机相结合的电价预测模型。选择合适的小波基和分解尺度,将电价序列分解成一个低频分量和多个高频分量,从而进行电价特征的提取。利用所获得的电价各分量,对电价各分量分别建立最小二乘支持向量机预测模型,最后汇总叠加各分量预测结果得到最终的预测电价。实验结果表明,此模型显著的提高了出清电价预测的精度。
In today's world,a profound transform of electric power industry mechanisms is taking place. Its inevitable trend is breaking monopoly,inducing competition and setting up a uniform, normative,open,orderd and competitive electricity market. China attempted to have a reformation of electric power market in 1990s,breaking the traditional integrative management mode,and carrying out separation of net and plant, transition and distribution. How to establish corresponding elecricity market operation mode rationally and how to use the relative historic data to forecast the future market clearing electricity price is a very meaningful work for every participator in the power market.
Firstly,The problem of establishing regional electric power market exchange pattern is researched thoroughly in the paper. After compared of several typical electric power market exchange patterns worldwide,the principle of regional power market trading mode is put forward. In this paper, China power market's orientation, structure and possible bilateral exchange manner are analyzed and discussed. And the regional power markets of wholesale competition model in China is designed in the paper. The purpose of this paper is to exploratory research the reformational orientation and the operational manner of China power market and optimal bidding strategies for participants in future market.
Secondly,The electric power market pricing mechanism is introduced, and the establishment and the influence factors of the electricity price are discussed. According to actual market clearing price's data,vary and distributing character of MCP data is researched.we obtained that different period data of MCP have different time series characteristic,the change of MCP have different trend. So the forecasting MCP idea based on period of time is put forward. Then, analyzed and compared the adventages and disadventages of some current forecasting methods, the forecasting models of market clearing price is proposed. To improve the accurary of price forecasting,two aspects of improvements are proposed:(1)selector of the factors which can show the change of the price sequence accurately;(2)the pretreatment of the input variables.
Two price forecasting model using Support Vector Machine is proposed: (1)similarity searching technique is applied to building the training set and input variables .The electricity price of next day is forecasted successfully by this model. The precision of forecasting based on similarity searching technique is higher than traditional methods.(2)The forecasting price model based on Wavelet Transform and Support Vector Machine is proposed. Selected of appropriate wavelet basis and decomposition scale, the market clearing price series is decomposed into one low-frequency and some high-frequency sub-series in the wavelet domain,So MCP feature is extracted form the market clearing price series. Using each new component of the original electricity price signal, LSSVM forecasting models for each sub-series is independently set up, then adding up to the forecasting results of all sub-series, the whole forecasted electricity price series are obtained. The results of forecasting show that this model impove the forecasting accurate of MCP remarkablly.
引文
[1]Hong Y Y,Hsiao C Y.Locational marginal price forecasting in Deregulated electricity markets using artificial intelligence[J].IEEE Proc.Gener.Transim.Distrib,2002,149(5):621-626.
[2]Guo Jau Jia,Peter B Luh.Selecting input factors for clusters of Gaussian radial basis function networks to improve market clearing price prediction[J].IEEE Trans.Power Syst,2003,18(3):665-672.
[3]Yamin H Y,Shahidehpour S M,Li Z.Adaptive short-term electricity price forecasting using artificial neural networks in the restructured power markets[J].Electrical power and energy systems,2004,26(8):571-581.
[4]潘昌,纪昌明.基于模糊回归的电价预测,水电能源科学.2003,21(3):86-88.
[5]李邦云,袁贵川,丁晓群.基于相似搜索和加权回归技术的短期电价预测,电力自动化设备.2004,24(1):42-45.
[6]Francisco J,Nogales,Javier Contreras,et al.Forecasting next-day electricity prices by time series models[J].IEEE Transaction on Power Systems,2002,17(2):342-348.
[7]Cuaresma J C,Hlouskova J,Kossmeier S,et al.Forecasting electricity spot-prices using linear univariate time-series models[J].Applied energy,2004,77(1):87-106
[8]Hu Zhaoyang,Sun Weizhen,Wang Zhen,et al.Short term,medium term and long-term forecasting of electricity price with consideration to market power[J].Automation of Electric Power Systems,2003,27(22):16-22.
[9]李彩华,郭志忠,王志伟.混合式短期边际电价预测模型[J].电力系统自动化,2002,26(21):29-33.
[10]Huang Rixing,Kang Chongqing,Xia Qing.Systems marginal price forecasting in electricity market[J].Automation of Electric Power Systems,2001,24(22):9-12.
[11]王锡凡,张显.短期电价综述,电力系统自动化.2006,30(3):92-101.
[12]康重庆,夏清,胡左浩.电力市场中预测问题的新内涵,电力系统自动化.2004,28(18):1-6.
[13]苏娟,杜松怀,周兴华.电力市场现货电价预测方法研究状况综述,继电器.2005,33(16):78-84.
[14]赵庆波,周原冰,郭蓉.模糊神经网络在电力系统边际电价预测中的应用[J].电网技术,2004,28(7):45-48.
[15]杨朴,游大海,谢培元.基于遗传算法和BP神经网络的电价预测[J].水电能源科学,2003,21(2):84-86.
[16]GAO Feng guan,Xiaohong,CAO Xiren,et al.Forecasting power market market clearing price and quantity using a neural network method[A].Power Engineering Society Summer Meeting,2000,2183-2188.
[17]魏平,李均利,陈刚等.基于小波的改进神经网络MCP预测方法及应用[J].电力系统自动化,2004,28(11):17-21.
[18]A Wang,B Ramsay.Prediction of system marginal price in the UK power pool using neural network[J].Proceedings of IEEE International Conference on Neural Networks,1997,4:2116-2120.
[19]B R Szkuta,L A Sanabria,T S Dillon.Electric price short-term forecasting using artificial neural networks[J].IEEE Trans.On PWS,1999,14(3):851-857.
[20]周明,聂艳丽,李庚银,等.基于小波分析的短期电价ARIMA预测方法[J].电网技术,29(9):50-55.
[21]B.Scholkopf,C.Burges,andV.Vapnik.Incorporating Invariances in Support Vector Learning Machines[C].Artificial Neural Networks-ICANN'96,Berlin,Springer Lecture Notesin Computer Science,1996,1112:47-52.
[22]MullerKR,SmolaAJ,etal.Predictiong Time Series with Support Vector Machines.[C].Proceeding of the International Conference on Artificial Neural Networks.Lausaunne,Switzerland,Springer,1997:999-1004.
[23]DruckerH,BurgesCJC,KaufmanL,SmolaA,VapnikVN.Support Vector Regression Machines.[C].Advances in Neural Information Processing Systems,Cambridge,MA,MITPress,1997:155-161.
[24]B.E.Boser,I.M.Guyon,and V.N.Vapnik.A training Algorithm for Optimal Margin Classifiers.[C].5~(th) Annual ACM Workshop on COLT,Pittsburgh.1992:144-152.
[25]BrownM,Gunn,SR.Empirical Data Modelling Algorithms:Additive Spline Models and Support Vector Machines[C].Proc of UKACC International Conference on Control,Swansea,UK,1998,1:709-714.
[26]张浩然,韩正之,李昌刚.基于支持向量机的非线性系统辨识[J].系统仿真学报,2003,15(1):119-121.
[27]Suykens JAK.Vandewalle J,DeMoor B.Optimal Control by Least Squares Support Vector Machines.Neural Networks,2001,14:105-113.
[28]deKrujf BJ,deVries TJA.On Using a Support Vector Machine in Learning Feed-Forward Control.Proc 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics.Como,Italy,2001,1:272-277.
[29]潘昌,纪昌明.基于模糊回归的电价预测[J],水电能源科学,2003,21(3):86-88.
[30]蓝信军,杨期余,江辉.加权模糊回归方法在电力负荷预测中的应用[J],长沙电力学院学报(自然科学版),2003,18(4):34-37.
[31]程吾.多元粗糙模糊回归预测[J].四川师范大学学报(自然科学版),2003,26(4):335-337.
[32]姜飞,龙子泉,林峰.模糊神经网络在电力短期负荷预测中的应用[J],控制理论与应用,2003,22(8):23-26.
[33]Vanaja Iyer,Chun Che Fung,Tanmas Gedeon.A Fuzzy-Neural Approach to Electricity Load and Spot-price Forecasting in a Deregulated Electricity Market.Conference on Convergent Technologies for Asia-Pacific Region,2003,4:1479-1482.
[34]赵庆波,周原冰,曾鸣等.模糊神经网络在电力系统边际电价预测中的应用[J],电网技术,2004,28(7):45-48.
[35]侯迪波,杨丽明,周泽魁.基于模糊神经网络的周期性结构预测方法研究[J],浙江大学学报(工学版),2004.38(3):308-311.
[36]郑丽颖.混沌神经网络及模糊混沌神经网络的研究与应用[D],哈尔滨工程大学,2002,9.
[37]杨洪明,段献忠.电价的混沌特性分析及其预测模型研究[J].电网技术,2004,28(3):60-64.
[38]刘广建,胡三高,戴俊良.电力系统边际电价的混沌特性及预测[J].中国电机工程学报,2003,23(5):6-8.
[39]龚剑扬,司锡才,郜丽鹏等.混沌序列的模糊神经网络预测[J],弹箭与制导学报2003,52-58.
[40]王卫宁,汪秉宏,史晓平.以混沌理论为基础的神经网络预测方法[J],运筹与管理,2003,12(6):58-61.
[41]孙雅明,张智晟.相空间重构和混沌神经网络融合的短期负荷预测研究[J],中国电机工程学报,2004,24(1):44-48.
[42]丁军威,孙雅明.基于混沌学习算法的神经网络短期负荷预测[J],电力系统自动化,2000,1(25).
[43]NOGALES F J,CONTRERAS J,CONEJO A J,et al.Forecasting next-day electricity prices by time series models[J].IEEE Trans.ON pws.2002.17(2):342-348.
[44]KIM C L.YU I K,SONG Y H.Prediction of system marginal price of electricity using wavelet transform analysis[J].Energy Conversion and Management,2002.43(14):1839-1851.
[45]LIU Zhengguo.FU Xiangming.GAO Fuying,et al.Forecasting market clearing price in Zhejiang generation market using neural network[J].Automation of Electric Power Systems.2002,26(9):49-52.
[46]李邦云,袁贵川,丁晓群.基于相似搜索和加权回归技术的短期电价预测[J].电力自动化设备,2004,24(1):42-45.
[47]杨莉,邱家驹,江道灼.基于BP网络的下一交易日无约束市场清算价格预测模型[J].电力系统自动化,2001,25(19):11-14.
[48]夏吉广,张维存,尹怡欣.基于RBF网络的电力市场清算电价预测[J].中国制造 业信息化,2005,34(5):138-140.
[49]宋永华,孙锦鑫.英国将推出电力市场的新模式.电网技术,1999,8:69-73.
[50]朱成章.各国电力企业改革模式探讨.华东电力,1999,1:49-53.
[51]宋新明.电力市场输电服务分析与研究[学位论文],北京,中国电力科学研究院,2000.
[52]张粒子,郑华.区域电力市场电价机制.北京:中国电力出版社.2004.4-61.
[53]姚建刚.区域电力市场竞价交易结构与模式的探讨.电力系统自动化.2003,27(22).
[54]姚建刚,章建,银车来.电力市场运营及其软件开发.中国电力出版社.2002.
[55]傅书遏,自晓民,张扬等.区域电力市场模式及运营方式.电力系统自动化.2003,5(29).
[56]Fraser A M,Swinney H L.Independent coordinates for strange attractors from mutual information[J].Physical Review A,1986,33:1134-1140.
[57]张显,王锡凡.短期电价预测综述[J].电力系统自动化,2006,30(3):92-101.
[58]ZHAO Qing bo,ZHOU Yuan bing,GUO Rong,et al.Application of FuzzyNeural Network in Power System Marginal Price Forecasting[J].Power System Technology,2004,28(7):45-48.
[2]Guo Jau Jia,Peter B Luh.Selecting input factors for clusters of Gaussian radial basis function networks to improve market clearing price prediction[J].IEEE Trans.Power Syst,2003,18(3):665-672.
[3]Yamin H Y,Shahidehpour S M,Li Z.Adaptive short-term electricity price forecasting using artificial neural networks in the restructured power markets[J].Electrical power and energy systems,2004,26(8):571-581.
[4]潘昌,纪昌明.基于模糊回归的电价预测,水电能源科学.2003,21(3):86-88.
[5]李邦云,袁贵川,丁晓群.基于相似搜索和加权回归技术的短期电价预测,电力自动化设备.2004,24(1):42-45.
[6]Francisco J,Nogales,Javier Contreras,et al.Forecasting next-day electricity prices by time series models[J].IEEE Transaction on Power Systems,2002,17(2):342-348.
[7]Cuaresma J C,Hlouskova J,Kossmeier S,et al.Forecasting electricity spot-prices using linear univariate time-series models[J].Applied energy,2004,77(1):87-106
[8]Hu Zhaoyang,Sun Weizhen,Wang Zhen,et al.Short term,medium term and long-term forecasting of electricity price with consideration to market power[J].Automation of Electric Power Systems,2003,27(22):16-22.
[9]李彩华,郭志忠,王志伟.混合式短期边际电价预测模型[J].电力系统自动化,2002,26(21):29-33.
[10]Huang Rixing,Kang Chongqing,Xia Qing.Systems marginal price forecasting in electricity market[J].Automation of Electric Power Systems,2001,24(22):9-12.
[11]王锡凡,张显.短期电价综述,电力系统自动化.2006,30(3):92-101.
[12]康重庆,夏清,胡左浩.电力市场中预测问题的新内涵,电力系统自动化.2004,28(18):1-6.
[13]苏娟,杜松怀,周兴华.电力市场现货电价预测方法研究状况综述,继电器.2005,33(16):78-84.
[14]赵庆波,周原冰,郭蓉.模糊神经网络在电力系统边际电价预测中的应用[J].电网技术,2004,28(7):45-48.
[15]杨朴,游大海,谢培元.基于遗传算法和BP神经网络的电价预测[J].水电能源科学,2003,21(2):84-86.
[16]GAO Feng guan,Xiaohong,CAO Xiren,et al.Forecasting power market market clearing price and quantity using a neural network method[A].Power Engineering Society Summer Meeting,2000,2183-2188.
[17]魏平,李均利,陈刚等.基于小波的改进神经网络MCP预测方法及应用[J].电力系统自动化,2004,28(11):17-21.
[18]A Wang,B Ramsay.Prediction of system marginal price in the UK power pool using neural network[J].Proceedings of IEEE International Conference on Neural Networks,1997,4:2116-2120.
[19]B R Szkuta,L A Sanabria,T S Dillon.Electric price short-term forecasting using artificial neural networks[J].IEEE Trans.On PWS,1999,14(3):851-857.
[20]周明,聂艳丽,李庚银,等.基于小波分析的短期电价ARIMA预测方法[J].电网技术,29(9):50-55.
[21]B.Scholkopf,C.Burges,andV.Vapnik.Incorporating Invariances in Support Vector Learning Machines[C].Artificial Neural Networks-ICANN'96,Berlin,Springer Lecture Notesin Computer Science,1996,1112:47-52.
[22]MullerKR,SmolaAJ,etal.Predictiong Time Series with Support Vector Machines.[C].Proceeding of the International Conference on Artificial Neural Networks.Lausaunne,Switzerland,Springer,1997:999-1004.
[23]DruckerH,BurgesCJC,KaufmanL,SmolaA,VapnikVN.Support Vector Regression Machines.[C].Advances in Neural Information Processing Systems,Cambridge,MA,MITPress,1997:155-161.
[24]B.E.Boser,I.M.Guyon,and V.N.Vapnik.A training Algorithm for Optimal Margin Classifiers.[C].5~(th) Annual ACM Workshop on COLT,Pittsburgh.1992:144-152.
[25]BrownM,Gunn,SR.Empirical Data Modelling Algorithms:Additive Spline Models and Support Vector Machines[C].Proc of UKACC International Conference on Control,Swansea,UK,1998,1:709-714.
[26]张浩然,韩正之,李昌刚.基于支持向量机的非线性系统辨识[J].系统仿真学报,2003,15(1):119-121.
[27]Suykens JAK.Vandewalle J,DeMoor B.Optimal Control by Least Squares Support Vector Machines.Neural Networks,2001,14:105-113.
[28]deKrujf BJ,deVries TJA.On Using a Support Vector Machine in Learning Feed-Forward Control.Proc 2001 IEEE/ASME International Conference on Advanced Intelligent Mechatronics.Como,Italy,2001,1:272-277.
[29]潘昌,纪昌明.基于模糊回归的电价预测[J],水电能源科学,2003,21(3):86-88.
[30]蓝信军,杨期余,江辉.加权模糊回归方法在电力负荷预测中的应用[J],长沙电力学院学报(自然科学版),2003,18(4):34-37.
[31]程吾.多元粗糙模糊回归预测[J].四川师范大学学报(自然科学版),2003,26(4):335-337.
[32]姜飞,龙子泉,林峰.模糊神经网络在电力短期负荷预测中的应用[J],控制理论与应用,2003,22(8):23-26.
[33]Vanaja Iyer,Chun Che Fung,Tanmas Gedeon.A Fuzzy-Neural Approach to Electricity Load and Spot-price Forecasting in a Deregulated Electricity Market.Conference on Convergent Technologies for Asia-Pacific Region,2003,4:1479-1482.
[34]赵庆波,周原冰,曾鸣等.模糊神经网络在电力系统边际电价预测中的应用[J],电网技术,2004,28(7):45-48.
[35]侯迪波,杨丽明,周泽魁.基于模糊神经网络的周期性结构预测方法研究[J],浙江大学学报(工学版),2004.38(3):308-311.
[36]郑丽颖.混沌神经网络及模糊混沌神经网络的研究与应用[D],哈尔滨工程大学,2002,9.
[37]杨洪明,段献忠.电价的混沌特性分析及其预测模型研究[J].电网技术,2004,28(3):60-64.
[38]刘广建,胡三高,戴俊良.电力系统边际电价的混沌特性及预测[J].中国电机工程学报,2003,23(5):6-8.
[39]龚剑扬,司锡才,郜丽鹏等.混沌序列的模糊神经网络预测[J],弹箭与制导学报2003,52-58.
[40]王卫宁,汪秉宏,史晓平.以混沌理论为基础的神经网络预测方法[J],运筹与管理,2003,12(6):58-61.
[41]孙雅明,张智晟.相空间重构和混沌神经网络融合的短期负荷预测研究[J],中国电机工程学报,2004,24(1):44-48.
[42]丁军威,孙雅明.基于混沌学习算法的神经网络短期负荷预测[J],电力系统自动化,2000,1(25).
[43]NOGALES F J,CONTRERAS J,CONEJO A J,et al.Forecasting next-day electricity prices by time series models[J].IEEE Trans.ON pws.2002.17(2):342-348.
[44]KIM C L.YU I K,SONG Y H.Prediction of system marginal price of electricity using wavelet transform analysis[J].Energy Conversion and Management,2002.43(14):1839-1851.
[45]LIU Zhengguo.FU Xiangming.GAO Fuying,et al.Forecasting market clearing price in Zhejiang generation market using neural network[J].Automation of Electric Power Systems.2002,26(9):49-52.
[46]李邦云,袁贵川,丁晓群.基于相似搜索和加权回归技术的短期电价预测[J].电力自动化设备,2004,24(1):42-45.
[47]杨莉,邱家驹,江道灼.基于BP网络的下一交易日无约束市场清算价格预测模型[J].电力系统自动化,2001,25(19):11-14.
[48]夏吉广,张维存,尹怡欣.基于RBF网络的电力市场清算电价预测[J].中国制造 业信息化,2005,34(5):138-140.
[49]宋永华,孙锦鑫.英国将推出电力市场的新模式.电网技术,1999,8:69-73.
[50]朱成章.各国电力企业改革模式探讨.华东电力,1999,1:49-53.
[51]宋新明.电力市场输电服务分析与研究[学位论文],北京,中国电力科学研究院,2000.
[52]张粒子,郑华.区域电力市场电价机制.北京:中国电力出版社.2004.4-61.
[53]姚建刚.区域电力市场竞价交易结构与模式的探讨.电力系统自动化.2003,27(22).
[54]姚建刚,章建,银车来.电力市场运营及其软件开发.中国电力出版社.2002.
[55]傅书遏,自晓民,张扬等.区域电力市场模式及运营方式.电力系统自动化.2003,5(29).
[56]Fraser A M,Swinney H L.Independent coordinates for strange attractors from mutual information[J].Physical Review A,1986,33:1134-1140.
[57]张显,王锡凡.短期电价预测综述[J].电力系统自动化,2006,30(3):92-101.
[58]ZHAO Qing bo,ZHOU Yuan bing,GUO Rong,et al.Application of FuzzyNeural Network in Power System Marginal Price Forecasting[J].Power System Technology,2004,28(7):45-48.