A Proper Unit Commitment (UC) scheduling relies upon the accurate information of the forecasted demand and demand pattern, availability and capacity of the generators, required minimum/maximum up and down time of the generators, and heat rate. According to the experiences of a local utility company, the difference of the fuel cost can reach million dollars per day with different unit commitment scheduling. Accurate day-ahead short-term load forecasting plays important roles in proper unit commitment scheduling.; This thesis describes the procedure to improve the unit commitment scheduling by adjusting the unit commitment program and applying the day-ahead results from the newly developed short-term load forecasting program in the SCADA/EMS system of a utility company. Comparison of field records will also be provided. Moreover, economic impact studies are performed to represent the effect from load forecasting error on a production cost, and a suggested level of acceptable load forecasting error is offered. |