Most of Brazil’s and China’s electricity comes from renewable sources through hydroelectric power plants. The operation of these plants can be modeled as a multi-objective optimization problem, in which the generation of energy, the water level in the reservoirs, and the minimal alteration of water flow after the optimization process are maximized. In this context, the work proposes to study the performance of different evolutionary meta-heuristics applied to the problem of electric dispatch in cascade mode power plants (at least two plants on a riverbed). The work has a special focus on the Multi-Objective Evolutionary Swarm Hybridization (MESH) algorithm boosted by the Special-Crowd-Distance operator, to solve three conflicting objectives in an Operation Multi-Reservoir System (OMRS). Experiments in this work indicate that MESH version proposed here is competitive when compared to standard evolutionary meta-heuristics: the Multi-Objective Evolutionary Algorithm Based on Decomposition (MOEA/D), versions of Non-dominated Sorting Genetic Algorithm (NSGA II and III), and the standard MESH solving the real-world problem here addressed.