Short-Term Load Forecasting Method Based on Fractal Theory

In terms of the present short-term load forecasting(STLF) methods, whether the linear or the nonlinear, neither could meet the STLF requirements better with the rapid developments of electrical power systems and electrical power markets, and so a novel STLF method was proposed based on fractal theory in this paper. Firstly, the paper investigated the fractal characteristics of power system loads based on fractal theory, then gave out the calculating method of the correlative dimension and embedded dimension according to G-P algorithm. Next, the paper discussed the C-C algorithm and revised it, and then used it to work out the time-delay. Finally, the paper established the STLF model and put into practice. The simulation results indicate that the proposed method possesses higher precision, and is an ideal STLF predictor. Key-Words: Short-term load forecasting, Fractal theory, Embedded dimension, Reconstruction, G-P algorithm, C-C algorithm