Electric Energy Demand Forcasting with GRNN for Energy Saving Strategy

Abstract:

In order to reasonable distribute electric energy and develop an energy-saving plan, electric energy demand forecasting which can represent a fundamental information is necessary. Power consumption has long duration, obvious nonlinear effect and the energy demand may be seen as a temporal series when its data are conveniently arranged. Neural networks have proved to be a very powerful tool which can predict a future value by studying the past one. In this paper, we propose a time series prediction model based on Generalized Regression Neural Network (GRNN) to predict the evolution of the monthly demand of electric consumption. GRNN is trained in just one presentation of the training patterns and is capable of providing fast and accurate results. The performance of the algorithm is evaluated using mean absolute and mean square errors. Tests are carried out by using the power consumption data of a campus distribution subsystem and the results obtained are found to be compatible with the actually data.

• Periodical: Applied Mechanics and Materials (Volumes 198-199)
• Pages: 639-643
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.198-199.639
• Online since: September 2012
• Authors: Lan Guang Zhao, Jing Hou, Jin Xiang Pian, Feng Zhong Zhang
• Keywords: Data Processing, Electric Energy Demand Forecasting, General Regression Neural Network (GRNN)

References

[1] K. Nose-Filho, A.D.P. Lotufo and C.R. Minussi: IEEE Trondheim PowerTech, (2011).

[2] K.D. Ranaweera, G.G. Karady and G.R. Farner: IEEE Trans. On Power Systems, vol. 12, pp.1388-1392, (1997).

[3] D. Carmona, M.A. Jaramillo, E. Gonzalez and J.A. Alvarez: IEEE annual conference of the industial electronics society, vol. 3, p.1860–1865, (2002).

[4] Q. Zhang and T. Liu: IEEE international conference on information management and engineering, pp.461-463, (2010).

[5] P. Milena, R.R. Milan, J. Boris and D. Vesna: IEEE 9th symposium on neural network applications in electrical engineering, (2008).

[6] Z. Yun, Z. Quan, S. Caixin, L. Shaolan, L. Yuming and S. Yang: IEEE Trans. On Power Systems, vol. 23, pp.853-858, (2008).

DOI: 10.1109/tpwrs.2008.922249

[7] D. X. Niu, H. Q. Wang and Z. H. Gu: International conference on machine learning and cybernetics, vol. 7, pp.4076-4082, (2005).

[8] C. Xia, J. Wang and K. Mcmenemy: Electricla power and energy systems, vol. 32, pp.73-750, (2010).

[9] Y. Sun, H.Z. Zhang and Y.N. Chang: International conference on intelligent computation technology and automation, pp.63-66, (2011).

[10] D.F. Specht: IEEE trans. on neural networks, vol. 2, pp.568-576, (1991).

[11] D. Tomandl and A. Schober: Neural Netw., vol. 14, p.1023–1034, (2001).