An hybrid technique to hydrothermal scheduling

This paper reports on a new hybrid method to solve the short-term operation planning problem of hydrothermal power systems. It uses evolutionary programming to determine the optimal schedules of all hydro, and thermal power plants in a closed optimization. Neural network is trained to satisfy the volume constraint imposed by the hydro plant. Numerical experiments show, that the proposed method is able to solve the complex non-linear optimization problem with its wealth of constraints in acceptable computational time.