Artificial intelligence

Abstract: The work covers non-traditional methods of forecasting, in particular, methods using artificial neural networks. The article considers such vital moments as: neural network configuration, normalization of input data, also random factors which have influence on the accuracy of load forecasts, are taken into account. Comparative characteristics of effectiveness of artificial neural networks and artificial neural networks with fuzzy logic are given.

Keywords: artificial neural networks, energy, fuzzy logic, short-term forecasting

References:

1. Petrashev S.M., development and implementation of technologies neural mathematical modeling to solve applied problems of heating, 1997, Odessa state. Polytechnic University, Odessa, 17 p.
2. Kovalenko M.V., short-term forecasting coherent fuel and energy resources in the domestic sector in period heating season, 2005, NTU "HPI", Kharkov, 20 p.
3. Danyluk O.V., theoretical foundations and modeling of power systems based on the technology of artificial neural least 2003, Institute of Electrodynamics of NAS of Ukraine, Kyiv, 36 p.
4. Abbass H.A., Pareto euro-evolution: constructing ensemble of neural networks using multi-objective optimization, 2003, Proceedings of CEC’03, N3, 2074-2080 p.
5. Alotto P. & Nervi, M.A. An efficient algorithm for the optimization of problems with several local minima, 2001, International Journal for Numerical Methods in Eng., N50, 847-868 p.
6. Bao-Liang Lu, Kita H. & Nishikawa Y. Inverting feed forward neural networks using linear and nonlinear programming, 1999, IEEE Transactions on Neural Networks, Vol. 10, N 6, 1271 – 1290 p.
7. Shumilova G.P., Gotman N.E., Startseva T.B. Electrical loads prediction in operational control electric power system structures based on neural network, 2008. Syktyvkar, 78 p.
8. Hsy Y., Ho K. Fuzzy expert systems: An application to short term load forecasting, 1992, IEE Proceedings, Vol. 139, No 6., 471-477 p.
9. Bakirtzis A.G., Theocharis J.B., Kiartzis S.J. Satsios K.J. Short term load forecasting using fuzzy neural networks, 1995, IEEE Trans. on Power Systems, Vol. 10, No 3, 1518 – 1523 p.
10. Srinivasan D., Liew A.C. Chang C.S. Forecasting daily load curves using a hybrid fuzzy-neural apporoach, 1994, IEE Proc - Generat. Transmiss. Distrib., Vol. 141, No 6., 561 -567 p.