Long-term load forecasting for fast developing utility using aknowledge-based expert system

摘要

The application of the classical forecasting methods, when appliednto a fast developing utility with a period characterized by fast andndynamic changes, are insufficient and may provide an invaluablendimension to the decision making process. In this paper, anknowledge-based expert system (ES) is implemented to support the choicenof the most suitable load forecasting model for medium/long term powernsystem planning. In the proposed ES, the detailed problem statementnincluding forecasting algorithms and the key variables (electrical andnnonelectrical variables) that affect the demand forecasts are firstlynidentified. A set of decision rules relating these variables are thennobtained and stored in the knowledge base. Afterwards, the best modelnthat will reflect accurately the typical system behavior over othernmodels is suggested to produce the annual load forecast. A practicalnapplication is given to demonstrate the usefulness of the developednprototype system