Short-term hydrothermal generation scheduling solved with a mixed-binary evolutionary particle swarm optimizer

In this paper, a mixed-binary evolutionary particle swarm optimizer (MB-EPSO) is applied to short-term hydrothermal generation scheduling problem (HGSP) in power systems. The HGSP has been formulated taking into account two decision variables simultaneously: water discharge (continuous) and thermal states (binary). The constraint handling technique used in the evolutionary algorithm is based on a strategy to generate and to keep the decision variables in feasible space through the correction operators, which were applied to hydro, thermal and system constraints. Such operators not only improve the quality of the final solutions but also significantly improve the convergence of the search process due to the use of feasible solutions. The results and effectiveness of the proposed technique are compared to the ones previously reported in the literature such as particle swarm optimizer (PSO), genetic algorithms (GA), and dynamic programming (DP), among others.

► We introduce heuristic operators to model the hydro, thermal and system constraint. ► The hydrothermal generation scheduling has been solved using a mixed-binary algorithm. ► The model can produce better solutions than other references techniques. ► The methodology presents a high adaptability to different optimization problems. ► An approach of the realistic case has been successfully analyzed (SIC).