A priority vector based technique for load forecasting

The priority-vector-based load forecasting technique uses pairwise comparisons to extract relationships from presorted historical hourly load and weather records for up to two years. The presorting is done to identify seasonal boundaries and to categorized the day types. The technique is adaptive in that it internally generates the coefficients of relationships among the governing variables (weather parameters) and the load. As these relationships change over time, such coefficients are automatically updated. The resulting linear method is robust and fairly accurate. This technique was used to forecast the hourly loads for a week, using a 168 h lead time, in different seasons. The only forecast variable used was the dry-bulb temperature. When tested for historical data in the service area of a Virginia electric utility for four weeks in different seasons of the year, the average forecast error remained mostly under 4%. Only 23 individual errors, out of a total of 672 cases, exceeded 6%