OPTIMAL SMOOTHING FOR TREND REMOVAL IN SHORT-TERM ELECTRICITY DEMAND FORECASTING

Changes to the electricity supply industry, in particular the progress towards deregulation, have prompted an increased interest in short te rm electricity demand forecasting. Trend removal has been identified a s a key element in forecasting, anti perhaps the area that carl most b enefit from the application of advanced time series analysis. A new ap proach to trend removal is presented eased on fixed interval, or optim al, smoothing techniques. Forecasts derived in this way are compared w ith more conventional approaches. Preliminary; results indicate the ne w method to be promising and deserving of further development.