An Improved Differential Evolution Approach for Short-term Hydrothermal Scheduling

Abstract

Short-term Hydrothermal Scheduling (STHS) plays one of the most important roles in power system operations. It is a complicated optimization problem that is di-cult to solve due to the consideration of various constraints. This paper proposes an improved difierential evolution approach for the solution of STHS problem. A regeneration operation is integrated into the proposed approach to avoid local optimums. Moreover, a novel constraints handling mechanism is introduced in order to avoid the use of penalty factors and improve the constraints handling e-ciency. The mechanism employs a repair strategy to preserve population feasibility as much as possible. Besides, a novel selection which makes a not clear distinction between the feasible and infeasible solutions is introduced to handle oversights of the repair strategy. The proposed approach is applied on a well-known hydrothermal test system to verify its feasibility and e-ciency. Numerical results are compared with those obtained by other population-based heuristic approaches reported in literature. It is observed from the comparisons that the proposed approach performs efiectively and can yield quite competitive solutions.