Optimal generation units start-up sequence during restoration of power system considering network reliability using bi-level optimization

During recent years power systems are operated near the nominal capacity of system equipment with low stability margin. Operation of power systems in this condition is extremely risky and power systems lose their stability with any intense failure of system equipment. In bulk power system, it is more critical and may cause a partial or overall blackout. The capability of black start to bring back system to a normal condition in the case of partial or overall shut down is very important in each power system. When a shutdown occurs, power plants with black start capability supply cranking power for non-black start power plants, pick up critical loads and energize required transmission lines. These actions should be done in minimum time interval to maximize system provided energy during the restoration process. Most important decision making during restoration process is the determination of start-up sequence of generation units. During the restoration process, units with black start capability (BS units) start at the beginning of the process to provide cranking power for non-black start units (NBS units). Hence the determination of NBS units is decision variable in the restoration problem. In this paper, this problem has been described as a bi-level optimization problem which in upper level determines the optimal start-up sequence of NBS units by using a Teaching-Learning Based Optimization algorithm and in lower level determines the optimal transmission path with minimum number of switching and maximum reliability between any two necessary buses using the searching path graph-based algorithm. The proposed approach has been implemented successfully on IEEE 24-bus RTS and IEEE 118-bus test systems.