Hassan Ghoudjehbaklou - Isfahan Technical University

Load forecasting in one of the most important part of any well planning and wise operation of power systems.Having accessed to a reasonably good load forecast, on one hand could reduce the unnecessary overhead costs and on the other hand can reduce the probability of a forced load shedding.In many different power system problems,in addition to peak load forecasts, Load Duration Curves (LDC) or another form of it namely Inverted Load Duration Curves(ILDC) have shown to be most useful. This paper discusses a new method for forecasting both ILDC and yearly peak loads. This method utilizes the Box & Jenkins Seasonal Autoregressive Integrated Moving Average(SARIMA) modelto estimate measurements noises. Extracting these noise from the measured values one is able to construct an estimated ILDC for each year.Then by using these estimated ILDCs a forecast of ILDC or peak load for latter years is possible. Tests on actual data indicate that this forecasting technique is superior to that of Box & Jenkins.