IEEE PES PERU – Role of Forecasting in Power Systems

Abstract:

Due to rapid industrialization and increased standard in the life style, the electric power requirement is increasing day-by-day. Various concerns related to progressive depletion of traditional fossil energy sources, increasing cost of fossil fuels, environmental protection, etc. have increased the interest in integrating renewable energy sources into the existing power systems. In the presence of renewable energy sources along with the volatile nature of demand and supply in electricity market environment, it has become a challenging task for the power utilities to perform proper scheduling of the generating units to serve the load demand at minimum operating cost. Forecasts are carried for various lengths of forecast horizons based on the requirements. A short term forecast is useful in dealing with the operational problems like frequency and power balance, voltage and reactive power support, and power quality issues. On the other hand, a long term forecast is useful for planning and operational problems like, economic scheduling, unit commitment and spinning reserve allocation. Power System forecasting include load, price, RES output, ancillary services.  Various techniques are suggested in past including both the conventional and non-conventional techniques. The operation and stability of electric power system is primarily governed by demand and supply balance. Various forecasting models based on Adaptive Wavelet Neural Network (AWNN) for hourly prediction, as well as day-ahead power forecast will be discussed.