فریبرز پرتوی راد

Forecasting electricity demand is one of the most important issues of the electrical energy system. Considering the structural changes in electricity demand and the stylized facts of electricity consumption in different sectors of demand, forecasting the amount of electricity demand will clarify the prospects of changes in the Iran's electric energy system in the medium and long term. By using new approaches, this prediction will have higher reliability. In this research, using the state-space approach and combining it with Markov regime switching, the main sources of uncertainties were included in the model. By using the data of electric energy feed-in the system to supply electricity demand and the average real price of electricity and temperature and the number of customers in the ten-year period of ۲۰۱۳-۲۰۲۲, the parameters of the model were estimated based on the state-space approach and Markov regime switching. State-space approach in the form of time-varying parameters and Markov switching approach in the form of variance fluctuations were included in the model. The results showed that the model based on this integrated approach gives a more accurate prediction than the classical model of electricity demand. The standard error of the estimated equations is reduced to ۰.۱ (in the competing model, the standard error of the corresponding equation is ۰.۰۳, and in the integrated approach, it is ۰.۰۰۲ for peak and ۰.۰۰۴ off-peak periods). While this system has a serious challenge in providing electricity demand, the sensitivity of electricity demand to the real price of electricity and temperature changes is decreasing and the demand for marginal costumer is increasing.