Experience with FNN models for medium term power demand predictions

被引:42
作者
Doveh, E [1 ]
Feigin, P [1 ]
Greig, D [1 ]
Hyams, L [1 ]
机构
[1] Technion Israel Inst Technol, Stat Lab, Fac Ind Engn & Management, IL-32000 Haifa, Israel
关键词
feedforward neural networks; medium term load forecasting; regularization; early stopping; time series prediction;
D O I
10.1109/59.761878
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
We relate experience of predicting medium term load demand for the Israeli power grid using neural network models. A variety of different regularization procedures are used for training and the results are compared. Furthermore we make suggestions on combining models and selecting input variables.
引用
收藏
页码:538 / 546
页数:9
相关论文
共 24 条
[1]   A neural network short term load forecasting model for the Greek power system [J].
Bakirtzis, AG ;
Petridis, V ;
Klartzis, SJ ;
Alexiadis, MC ;
Maissis, AH .
IEEE TRANSACTIONS ON POWER SYSTEMS, 1996, 11 (02) :858-863
[2]  
Bishop C. M., 1995, NEURAL NETWORKS PATT
[3]   Performance enhancement using nonlinear preprocessing [J].
Chow, T ;
Leung, CT .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1996, 7 (04) :1039-1042
[4]  
Hassibi B., 1992, ADV NEURAL INFORM PR, V5
[5]   MULTILAYER FEEDFORWARD NETWORKS ARE UNIVERSAL APPROXIMATORS [J].
HORNIK, K ;
STINCHCOMBE, M ;
WHITE, H .
NEURAL NETWORKS, 1989, 2 (05) :359-366
[6]  
LeCun Y., 1989, Advances in neural information processing systems, V2, P598
[7]   MULTILAYER FEEDFORWARD NETWORKS WITH A NONPOLYNOMIAL ACTIVATION FUNCTION CAN APPROXIMATE ANY FUNCTION [J].
LESHNO, M ;
LIN, VY ;
PINKUS, A ;
SCHOCKEN, S .
NEURAL NETWORKS, 1993, 6 (06) :861-867
[8]  
LEVIN AU, 1994, ADV NEUR INF P SYS, V6
[9]   Comparison of very short-term load forecasting techniques [J].
Liu, K ;
Subbarayan, S ;
Shoults, RR ;
Manry, MT ;
Kwan, C ;
Lewis, FL ;
Naccarino, J .
IEEE TRANSACTIONS ON POWER SYSTEMS, 1996, 11 (02) :877-882
[10]  
MACKAY DJC, 1995, MAXIMUM ENTROPY BAYE