An improved invasive weed optimization algorithm for solving dynamic economic dispatch problems with valve-point effects

In this study, an improved invasive weed optimisation (CMIWO) algorithm is investigated to solve the dynamic economic dispatch (DED) problem with valve-point effects. In the proposed algorithm, a hybrid operator including selective crossover, random mutation and row crossover is proposed to improve the exploration and exploitation abilities. Moreover, a self-adaption repair method is developed and embedded into the proposed algorithm to repair infeasible solutions. To verify the optimisation performance of CMIWO, six well-known DED problems in three different-scale power systems are tested and compared with other algorithms that have been proposed in the literature. The experimental results show that CMIWO can find the more economical dispatch solutions compared to other algorithms, and the self-adaption repair method can successfully convert infeasible solutions into feasible solutions. The convergence ability of CMIWO is also verified after the detailed comparison.