Reliability constrained unit commitment with combined hydro and thermal generation embedded using self-learning group search optimizer

This paper proposes a reliability constrained unit commitment problem with combined hydro and thermal generation embedded (RCHTUC), solved by a SLGSO (self-learning group search optimizer). The RCHTUC problem aims at minimizing the sum of fuel costs and start-up costs of thermal plants subject to various operation constraints. Furthermore, the problem takes into account the combination of hydro and thermal systems and the reliability constraints in hydrothermal power systems so as to respond to unforeseen outages and changes of load demands. In order to solve the RCHTUC problem, a SLGSO is developed from the GSO (group search optimizer), applying adaptive covariance matrix to design the optimum searching strategy and employing Lévy flights to increase the diversity of group. This paper reports on the simulation results obtained by the proposed method. The results are compared with those obtained by other methods on different hydrothermal systems over the scheduling horizon. The simulation results demonstrate the efficiency of the SLGSO for tackling the RCHTUC problem.