Application of evolutionary techniques to short-term optimization of hydrothermal systems

This paper presents the implementation of evolutionary techniques for solving short-term optimization problems of hydrothermal power systems where a complete model for the hydroplants is considered. Beyond the usual water conservation equations and bounds on variables, this formulation can handle some other important aspects of hydro plants and reservoirs that are rarely taken into account. Environmental and ramping costs of thermal units and transmission losses are also included in this formulation. The optimization process consists of a simulation of both the hydraulic and the thermal subsystems to obtain feasible (but no optimal) solutions, and the application of evolutionary methods to improve the global profit of the system gradually. The algorithms are applied to a real case with several coupled hydroplants and their results are compared with the ones obtained by other classical optimization methods to show the capacity of these methods to find global or near-global optimum solutions.