A Joint Optimization Model for Energy and Reserve Capacity Scheduling With the Integration of Variable Energy Resources

In this paper, a two-stage approach is proposed on a joint dispatch of thermal power generation and variable resources including a storage system. Although, the dispatch of alternate energy along with conventional resources has become increasingly important in the new utility environment. However, recent studies based on the uncertainty and worst-case scenario-oriented robust optimization methodology reveal the perplexities associated with renewable energy sources (RES). First, the load demand is predicted through a convolutional neural network (CNN) by taking the ISO-NECA hourly real-time data. Then, the joint dispatch of energy and spinning reserve capacity is performed with the integration of RES and battery storage system (BSS) to satisfy the predicted load demand. In addition, the generation system is penalized with a cost factor against load not served for the amount of energy demand which is not fulfilled due to generation constraints. Meanwhile, due to ramping of thermal units, the available surplus power will be stored in the backup energy storage system considering the state of charge of the storage system. The proposed method is applied on the IEEE-standard 6-Bus system and particle swarm optimization (PSO) algorithm is used to solve the cost minimization objective function. Finally, the proposed system performance has been verified along with the reliability during two worst-case scenarios, i.e., sudden drop in power demand and a short-fall at the generation end.