Demand forecasting and generation scheduling for a secure and cost-effective operation of power systems

This paper presents a practical approach for demand forecasting and generation scheduling in a power system with solar PV sources and a Compressed Air Energy Storage System. The demand-forecasting method is improved using an artificial neural network in the first stage. After forecasting the demand for several representative days, the cost-effective generation scheduling is achieved for PV generation units and storage systems at the second stage. The proposed forecasting and generation scheduling approaches are tested in IEEE-RTS-96 test system, and the results are discussed from a technical and economic point of view.