WANN and ANN based Urban Load Forecasting for Peak Load Management

摘要

The need for electricity is increasing everyday with an increase in population and development of our civilization. Though the need for electricity is reaching new levels every day the resources available for the generation of electric power is being depleted rapidly and proper planning and utilization of the generated power has become extremely essential. However, to properly plan and match the supply with demand it is necessary to have an approximate prior knowledge of the demand that would be required to be met in near future. For this purpose, Short Term Load Forecasting (STLF) has an important role to play. The prior hourly or daily prediction of load can be extremely helpful in identifying the peak load periods. There are several techniques for the prediction of electrical load. Recently Artificial Neural Network (ANN) has been gaining importance. This paper attempts to compare between ANN and the hybrid technique of Wavelet decomposed Artificial Neural Network (WANN) to find out the technique which gives a better forecasting result for electric loads.