Study of daily peak load forecasting by structured representation on genetic algorithms for function fitting

In recent years, the user who introduces the small-scale power generation facilities (solar photovoltaic generation, wind power generation, micro gas turbine, and fuel cell) increases the power system deregulation. The electric power system becomes more and more complicated. Therefore, we thought that the electric power demand forecasting was required in order to operate economically and with high efficiency. This paper presents a method of short-term load forecasting by STROGANOFF (i.e. structured representation on genetic algorithms for nonlinear function fitting). The STROGANOFF is an hierarchical technique of multiple regression analysis method and GA-based search strategy that approximates the value of prediction to the future data by the past data. Considering local information, the examination was carried out using the electric demand data of this campus with power generation facilities.