基于自回归条件异方差-反向传播网络模型的日前边际电价预测Day-Ahead Marginal Price Forecasting Based on Autoregressive Conditional Heteroskedasticity-Back Propagation Network Model
余帆;沈炯;刘西陲;
摘要(Abstract)：
针对日前电力市场提出了一种基于自回归条件异方差分析的改进神经网络模型。首先利用自回归条件异方差分析得到边际电价序列的条件方差,然后以条件方差作为电价波动风险指标,建立基于历史电价、历史负荷和历史电价条件方差等输入量的自回归条件异方差?反向传播网络模型,并利用该模型对美国PJM电力市场的日前边际电价进行了预测。结果表明,引入自回归条件异方差分析可以有效提高传统反向传播网络的预测精度。
关键词(KeyWords)： 电力市场;日前边际电价;预测;自回归条件异方差(ARCH);神经网络
基金项目(Foundation):
作者(Author): 余帆;沈炯;刘西陲;
Email:
DOI:
参考文献(References)：
- [1]Francisco J N,Javier C.Forecaseting next-day electricity prices by time series models[J].IEEE Trans on Power Systems,2002,17(2):342-348.
- [2]Gao Feng,Guan Xiaohong,Cao Xiren,et al.Forecasting power market clearing price using neural networks[C].Proceedings of the3rd World Congress on Intelligent Control and Automation,Hefei,P.R.China,2002:1098-1102.
- [3]李郁侠,赵军科.混沌理论和快速BP神经网络在边际电价预测中的应用[J].武汉大学学报(工学版),2005,38(6):78-82.Li Yuxia,Zhao Junke.Application of chaotic theory and fast BPANN to forecast of marginal electricity price[J].Engineering Journal of Wuhan University,2005,38(6):78-82(in Chinese).
- [4]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.Li Caihua,Guo Zhizhong,Wang Zhiwei.Short-term marginal price forcasting with hybrid model[J].Automation of Electric Power System,2002,26(21):29-33(in Chinese).
- [5]谢培元,游大海,曾次玲.基于遗传算法优化BP网络的提前一天市场清算电价预测[J].电力自动化设备,2004,24(3):54-56.Xie Peiyuan,You Dahai,Zeng Ciling.Clearing price forecast for day-ahead market based on genetic-algorithm-optimized BP network[J].Electric Power Automation Equipment,2004,24(3):54-56(in Chinese).
- [6]Engle R F.Autoregressive conditional heteroskedasticity with estimates of the variance of U.K.inflation[J].Econometrica,1982,50(4):987-1008.
- [7]Bollersle T.Generalized autoregressive conditional heteroskedasticity[J].Journal of Econometrics,1986,(31):307-327.
- [8]Engle R,Kraft D.Multiperiod forecast error variances of inflation estimated from ARCH models[R].Washington D.C.:Bureau of the Census,1983.
- [9]Martin T H,Howard B D,Mark H B.Neural network design[M].北京:机械工业出版社,2002.
- [10]周明,严正,倪以信,等.含误差预测校正的ARIMA电价预测新方法[J].中国电机工程学报,2004,24(12):63-68.Zhou Ming,Yan Zheng,Ni Yixin,et al.A novel ARIMA approach on electricity price forecasting with the improvement of predicted error[J].Proceedings of the CSEE,2004,24(12):63-68(in Chinese).
- [11]高铁梅.计量经济分析方法与建模[M].北京:清华大学出版社,2006.
- [12]胡守仁,余少波,戴葵.神经网络导论[M].长沙:国防科技大学出版社,1993.