Load/price forecasting and managing demand response for smart grids: Methodologies and challenges

An important issue in smart grids is to manage demand-response (DR) to reduce peak electricity load and hence future investment in thermal generations and transmission networks. To be able to predict the behaviors of the grid and customers, one needs to establish appropriate models and estimate the corresponding parameters from measurements using statistical estimation techniques. Based on these models, one can then make prediction of its behavior in the future. To carry out DR optimization, appropriate models for the appliances are usually required so that proper control can be performed using the forecasted electricity price and model-based predicted energy consumption. Another common cost model involved in DR optimization is called the utility function, which measures the users' satisfaction on the appliances. This function is usually inversely proportional to the amount of energy consumed by the appliances so that the electricity cost can be reduced without significantly discomfort the users.