Short-term load forecasting of UPPCL using ANN

Power sector reforms have been introduced in Uttar Pradesh, India in as early as 1999. Restructuring and unbundling of UP state electricity board was done by segregating power generation, transmission and distribution functions into autonomous and separately accountable entities. Uttar Pradesh power corporation India ltd. (UPPCL) was formed. An independent Regulatory Body was formed. Private sector participation was encouraged and tariff reform was introduced with the objective to rationalize tariff for full cost recovery and minimize cross subsidy. Power is procured through long-term PPA or energy exchange based on forecasting of day-ahead load demand. The present load forecasting method is based on previous year load and uses very crude method with large errors. This paper discusses role of ANN in day-ahead hourly forecast of the power system load in UPPCL so as to minimize the error in demand forecasting. A new artificial neural network (ANN) has been designed to compute the forecasted load of UPPCL. The data used in the modeling of ANN are hourly historical data electricity load. The ANN model is trained on hourly data from UPPCL from April, 2014 to June, 2014 and tested on out-of-sample data of two weeks. Simulation results obtained have shown that day-ahead hourly forecasts of load using proposed ANN is very accurate with very less error.