Abstract

Economic Load Dispatch (ELD) is a critical optimisation problem in power system operation that aims to minimise total fuel cost while satisfying system demand and generator operating constraints. The presence of valve-point loading effectsmakesthe ELD problem highly non-linear and non-convex, limiting the effectiveness of conventional optimisation methods. This paper proposes a hybrid optimisation approach that integrates Differential Evolution (DE) and Teaching–Learning-BasedOptimization(TLBO) to enhance solution quality and convergence performance. In the proposed DE–TLBO framework, DE is employed during the initial phase to perform global exploration of the searchspace,while TLBO is used in the later phase to intensify local exploitation and refine candidate solutions. A repair-basedconstrainthandling strategy is incorporated to ensure strict satisfaction of generator limits and power balance constraints at everyiteration.The effectiveness of the proposed approach is evaluated using the IEEE-40 generating unit test system under identical simulationconditions. Simulation results demonstrate that the proposed DE–TLBO hybrid algorithm achieves lower fuel cost and improved convergence characteristics compared to standalone DE and TLBO algorithms. The findings confirm that hybrid metaheuristic optimization provides a robust and efficient solution for large-scale, non-convex economic load dispatch problems. Keywords: Economic Load Dispatch; Differential Evolution; Teaching–Learning-Based Optimization; Hybrid Metaheuristic Optimization; Valve-Point Loading Effect; IEEE-40 Generating Unit System