Abstract—The novel hybrid model for forecasting of time series characterized by nonlinear, nonstationary behavior is presented. Experimental studies on synthetic and real data are carried out and discussed. The application of developed approach for forecasting of price situation in the electricity market and for forecasting of the parameters of the expected operation conditions of the electric power system has demonstrated the accuracy improvement using the most common metrics. The system is examined using the benchmark data records from Australian national electricity market.
Index Terms—Hilbert-Huang Transform; artificial neural networks; forecasting; electricity price; power systems.
V. Kurbatsky and N. Tomin are with the Power Systems Department of Energy Systems Institute of Russian Academy of Sciences, Irkutsk, 664033 RUSSIA (kurbatsky@isem.sei.irk.ru, tomin@isem.sei.irk.ru). D. Sidorov and V. Spiryaev are with the Applied Mathematics Department of Energy Systems Institute of Russian Academy of Sciences, Irkutsk, 664033 RUSSIA (dsidorov@isem.sei.irk.ru, elden@mail.ru). 1 The term “intelligent” is used in reference to the approaches, methods, systems and complexes using artificial intelligence technologies
Cite: Victor Kutbatsky, Denis Sidorov, Nikita Tomin and Vadim Spiryaev, "Hybrid Model for Short-Term Forecasting in Electric Power System," International Journal of Machine Learning and Computing vol. 1, no. 2, pp. 138-147, 2011.