| An important element of effective power system operation is the well-planned short term scheduling of power generating units. The power system operator uses historical load data to schedule the available generating units to meet the hourly system loads in an economical and reliable manner. A Fuzzy Logic (FL) expert system is integrated with Artificial Neural Networks (ANN) for a more reliable operation of short-term load forecasting.; The 24 hours ahead forecasted load is obtained through two steps. First, the FL module carries the job of mapping the highly non-linear relationship between the weather parameters and it's impact on the daily electric load peak. Second, the 12 ANN modules are trained using historical hourly load and weather data, combined with the FL output data, to perform the final forecast. A comparison made between this model, an ANN model, and an Autoregressive Moving Average (ARMA) model shows the efficiency and accuracy of this new approach. |
