Security-constrained unit commitment reaction to load and price forecasting errors

In daily power markets, forecasting electricity prices and loads is the most essential task and basis for security-constrained unit commitment (SCUC) and risk management. An approach to predict the market behavior is to use the historical prices, loads and other required information to forecast the future prices and loads. This paper forecasts the loads and prices with artificial neural networks (ANN) and does the unit commitment for the IEEE 30 bus test system using the ANN results. Price and load forecasting done with multilayer perceptron (MLP) and radial basis function (RBF). Forecasted load and price of MLP and RBF have been applied to IEEE 30 bus test system separately. This paper shows, what happened with unit commitment results when the error of the load and price forecasting going up or down. The unit commitment results show reactions against this changing error. When the error is low (lower than 2 percent) unit commitment acts properly and if the error become high then there are two reactions, the first one is instability of the bus test system network and the second is extra cost for network. This paper extended your view in risk management and making decision.