ANN based Electric Load Forecasting Applied to Real Time Data

The accuracy of electric load forecasts significantly affects the overall performance of power system. Some time due to complicated load pattern, forecasting becomes difficult. The object of this study is to develop more effective forecasting models, among others. This paper compares the electric load forecasting accuracy of ANN based techniques. This study investigates the time series techniques used to forecasts electric load, e.g. MA, linear trend, the exponential and parabolic trend. In the present study ANN based time series forecasting model has been developed, on hourly electric load consumption data. Data used to forecast is acquired from a distribution company located in Noida, Uttar Pradesh. These ANN based techniques are evaluated for the forecasting errors.