Auto-regressive Recurrent Neural Network Approach for Electricity Load Forecasting
this paper presents an auto-regressive network called the Auto-Regressive Multi-Context Recurrent Neural Network (ARMCRN), which forecasts the daily peak load for two large power plant systems. The auto-regressive network is a combination of both recurrent and non-recurrent networks. Weather component variables are the key elements in forecasting because any change in these variables affects the demand of energy load. So the AR-MCRN is used to learn the relationship between past, previous, and future exogenous and endogenous variables. Experimental results show that using the change in weather components and the change that occurred in past load as inputs to the AR-MCRN, rather than the basic weather parameters and past load itself as inputs to the same network, produce higher accuracy of predicted load. Experimental results also show that using exogenous and endogenous variables as inputs is better than using only the exogenous variables as inputs to the network.
[1] A. Bakirtzis, V.Petridis, and S. Klartizis, `A neural network short term
load forecasting model for the greek power system', In IEEE Tran. On
Power Systems, 2 (11), 858--863, (1996).
[2] B.Q.Huang, T.Rashid, and T.Kechadi, `A new modi ed network based
on the elman network', in In Proceedings of IASTED International
Conference on Arti cial Intelligence and Application, ed., M. H.
Hamza, volume 1 of ISBN: 088986-404-7, pp. 379--384, Innsbruck,
Austria, (2004). ACTA Press.
[3] W. Charytoniuk and M-S. Chen, `Very short-term load forecasting
using neural networks', In IEEE Tran. On Power Systems, 15(1), 1558-
-1572, (2000).
[4] B.J Chen, M.W Change, and C.J. Lin, `Eunite network competition:
Electricity load forecasting', Technical report, In EUNITE 2001
symposium, a forecasting competition, (2001).
[5] K.W.Yeung Dit-Yan, `A locally recurrent neural network model for
grammatical interface', in processing Proceedings of the International
Conference on Neural Information Processing, pp. 1468--1473, (1995).
[6] G. Dorffner, `Neural networks for time series processing', In Neural
Network World, 6, 447--468, (1996).
[7] I. Drezga and S. Rahman, `Input variable selection for ann-based
shortterm load forecasting', In IEEE Trans. On Power Systems, 13(4),
1238--1244, (November 1998).
[8] I. Drezga and S. Rahman, `Short term load forecasting with local an
predictors', In IEEE Trans. On Power Systems, 14(3), 844--850,
(August 1999).
[9] M. A. El-Sharkawi and Dagmar Niebur Editors, `Application of
artificial neural networks to power systems', In IEEE press, 96 TP 112-
0, (1996).
[10] J.L. Elman, `Finding structure in time', In Cognitive Science, 14(2),
179--211, (1990).
[11] D. Esp, `Adaptive logic networks for east slovakian electrical load
forecasting', Technical report, In EUNITE 2001 symposium, a
forecasting competition, (2001).
[12] J. J. Hop eld, `Learning algorithms and probability distributions in
feed-forward and feed-back networks', In Proceedings of the National
Academy of Sciences of the USA, volume 84, pp. 8429--8433, USA,
(1987).
[13] K. Kalaitzakis, G.S. Stavrakakis, and E.M. Anagnostakis, `Shot term
load forecasting based on arti cial neural networks parallel
implementation', In Eclectic Power System Research 63, pp. 185--196,
Isanbul, Turky, (2002).
[14] M.S. Kandil, S.M. El-Debeiky, and N.E. Hasanien, `Overview and
comparison of long-term forecasting techniques for a fast developing
utility', In part I, Electric Power Systems Research 58, pp. 11--17,
New York, (2001).
[15] I. King and J. Tindle, `Storage of half hourly electric metering data and
forecasting with arti cial neural network technology', Technical
report, In EUNITE 2001 symposium, a forecasting competition, (2001).
[16] T. Kohonen, Self-Organizing Maps, Springer Verlag, Berlin,
Heidelberg, 1995.
[17] W. Kowalczyk, `Averaging and data enrichment: two approaches to
electricity load forecasting', Technical report, In EUNITE 2001
symposium, a forecasting competition, (2001).
[18] A. Lendasse, M. Corttell, V.Wertz, and M. Veleysen, `Prediction of
electric load using kohonen maps-application to the polish electricity
consumption', In the proceeding of the American Control conference
Anchorage, pp. 8-10, AK, (May 2002).
[19] L. Lewandowski, F. Sandner, and P. Portzel, `Prediction of electricity
load by modeling the temperature dependencies', Technical report, In
EUNITE 2001 symposium, a forecasting competition, (2001).
[20] A. D. Papalxopoulos and T. C. Hiterbe, `A regression-based approach
to short-term load forecasting', In IEEE Tran. On Power Systems, 5(4),
1535--1547, (1990).
[21] D. Park, M. Al-Sharkawi, R. Marks, A. Atlas, and M. Damborg,
`Electric load forecasting using an arti cial neural network', In IEEE
Tran. On Power Systems, 6(2), 442--449, (1991).
[22] S. Rahman, `Formulation and analysis of a rule-based short-term load
forecasting algorithm', Proceedings of the IEEE, 39, pp.161--169,
(1994).
[23] T. Rashid, B.Q. Huang, and M.-T Kechadi, `A new simple recurrent
network with real-time recurrent learning process', In the14th Irish
Arti cial Intelligence and Cognitive Science (AICS'03), volume 1, pp.
169--174, Dublin, Ireland, (2003).
[24] T. Rashid and T. Kechadi, `A practical approach for electricity load
forecasting', In the proceeding WEC'05, The 3rdWorld Enformatika,
volume 5, pp. 201--205, Isanbul, Turky, (2005). ACTA Press.
[25] D. Srinivasan, `Evolving arti cial neural networks for short term load
forecasting', in In Neuro computing, volume 23, pp. 265--276, (1998).
[26] William H. Wilson, `Learning performance of networks like elman's
simple recurrent netwroks but having multiple state vectors', Workshop
of the 7th Australian Conference on Neural Networks, Australian
National University Canberra, (1996).
[27] B-L. Zhang and Z-Y. Dong, `An adaptive neural-wavelet model
forshort term load forecasting', In Electric Power Systems Research 59,
pp. 121--129, New York, (2001).
[1] A. Bakirtzis, V.Petridis, and S. Klartizis, `A neural network short term
load forecasting model for the greek power system', In IEEE Tran. On
Power Systems, 2 (11), 858--863, (1996).
[2] B.Q.Huang, T.Rashid, and T.Kechadi, `A new modi ed network based
on the elman network', in In Proceedings of IASTED International
Conference on Arti cial Intelligence and Application, ed., M. H.
Hamza, volume 1 of ISBN: 088986-404-7, pp. 379--384, Innsbruck,
Austria, (2004). ACTA Press.
[3] W. Charytoniuk and M-S. Chen, `Very short-term load forecasting
using neural networks', In IEEE Tran. On Power Systems, 15(1), 1558-
-1572, (2000).
[4] B.J Chen, M.W Change, and C.J. Lin, `Eunite network competition:
Electricity load forecasting', Technical report, In EUNITE 2001
symposium, a forecasting competition, (2001).
[5] K.W.Yeung Dit-Yan, `A locally recurrent neural network model for
grammatical interface', in processing Proceedings of the International
Conference on Neural Information Processing, pp. 1468--1473, (1995).
[6] G. Dorffner, `Neural networks for time series processing', In Neural
Network World, 6, 447--468, (1996).
[7] I. Drezga and S. Rahman, `Input variable selection for ann-based
shortterm load forecasting', In IEEE Trans. On Power Systems, 13(4),
1238--1244, (November 1998).
[8] I. Drezga and S. Rahman, `Short term load forecasting with local an
predictors', In IEEE Trans. On Power Systems, 14(3), 844--850,
(August 1999).
[9] M. A. El-Sharkawi and Dagmar Niebur Editors, `Application of
artificial neural networks to power systems', In IEEE press, 96 TP 112-
0, (1996).
[10] J.L. Elman, `Finding structure in time', In Cognitive Science, 14(2),
179--211, (1990).
[11] D. Esp, `Adaptive logic networks for east slovakian electrical load
forecasting', Technical report, In EUNITE 2001 symposium, a
forecasting competition, (2001).
[12] J. J. Hop eld, `Learning algorithms and probability distributions in
feed-forward and feed-back networks', In Proceedings of the National
Academy of Sciences of the USA, volume 84, pp. 8429--8433, USA,
(1987).
[13] K. Kalaitzakis, G.S. Stavrakakis, and E.M. Anagnostakis, `Shot term
load forecasting based on arti cial neural networks parallel
implementation', In Eclectic Power System Research 63, pp. 185--196,
Isanbul, Turky, (2002).
[14] M.S. Kandil, S.M. El-Debeiky, and N.E. Hasanien, `Overview and
comparison of long-term forecasting techniques for a fast developing
utility', In part I, Electric Power Systems Research 58, pp. 11--17,
New York, (2001).
[15] I. King and J. Tindle, `Storage of half hourly electric metering data and
forecasting with arti cial neural network technology', Technical
report, In EUNITE 2001 symposium, a forecasting competition, (2001).
[16] T. Kohonen, Self-Organizing Maps, Springer Verlag, Berlin,
Heidelberg, 1995.
[17] W. Kowalczyk, `Averaging and data enrichment: two approaches to
electricity load forecasting', Technical report, In EUNITE 2001
symposium, a forecasting competition, (2001).
[18] A. Lendasse, M. Corttell, V.Wertz, and M. Veleysen, `Prediction of
electric load using kohonen maps-application to the polish electricity
consumption', In the proceeding of the American Control conference
Anchorage, pp. 8-10, AK, (May 2002).
[19] L. Lewandowski, F. Sandner, and P. Portzel, `Prediction of electricity
load by modeling the temperature dependencies', Technical report, In
EUNITE 2001 symposium, a forecasting competition, (2001).
[20] A. D. Papalxopoulos and T. C. Hiterbe, `A regression-based approach
to short-term load forecasting', In IEEE Tran. On Power Systems, 5(4),
1535--1547, (1990).
[21] D. Park, M. Al-Sharkawi, R. Marks, A. Atlas, and M. Damborg,
`Electric load forecasting using an arti cial neural network', In IEEE
Tran. On Power Systems, 6(2), 442--449, (1991).
[22] S. Rahman, `Formulation and analysis of a rule-based short-term load
forecasting algorithm', Proceedings of the IEEE, 39, pp.161--169,
(1994).
[23] T. Rashid, B.Q. Huang, and M.-T Kechadi, `A new simple recurrent
network with real-time recurrent learning process', In the14th Irish
Arti cial Intelligence and Cognitive Science (AICS'03), volume 1, pp.
169--174, Dublin, Ireland, (2003).
[24] T. Rashid and T. Kechadi, `A practical approach for electricity load
forecasting', In the proceeding WEC'05, The 3rdWorld Enformatika,
volume 5, pp. 201--205, Isanbul, Turky, (2005). ACTA Press.
[25] D. Srinivasan, `Evolving arti cial neural networks for short term load
forecasting', in In Neuro computing, volume 23, pp. 265--276, (1998).
[26] William H. Wilson, `Learning performance of networks like elman's
simple recurrent netwroks but having multiple state vectors', Workshop
of the 7th Australian Conference on Neural Networks, Australian
National University Canberra, (1996).
[27] B-L. Zhang and Z-Y. Dong, `An adaptive neural-wavelet model
forshort term load forecasting', In Electric Power Systems Research 59,
pp. 121--129, New York, (2001).
@article{"International Journal of Electrical, Electronic and Communication Sciences:64391", author = "Tarik Rashid and B. Q. Huang and M-T. Kechadi and B. Gleeson", title = "Auto-regressive Recurrent Neural Network Approach for Electricity Load Forecasting", abstract = "this paper presents an auto-regressive network called the Auto-Regressive Multi-Context Recurrent Neural Network (ARMCRN), which forecasts the daily peak load for two large power plant systems. The auto-regressive network is a combination of both recurrent and non-recurrent networks. Weather component variables are the key elements in forecasting because any change in these variables affects the demand of energy load. So the AR-MCRN is used to learn the relationship between past, previous, and future exogenous and endogenous variables. Experimental results show that using the change in weather components and the change that occurred in past load as inputs to the AR-MCRN, rather than the basic weather parameters and past load itself as inputs to the same network, produce higher accuracy of predicted load. Experimental results also show that using exogenous and endogenous variables as inputs is better than using only the exogenous variables as inputs to the network.
", keywords = "Daily peak load forecasting, neural networks, recurrent neural networks, auto regressive multi-context neural network.", volume = "1", number = "10", pages = "1540-9", }
