Power system operational optimization using the kakapo optimization algorithm for dynamic economic dispatch

Introduction. Metaheuristic algorithms are effective for solving complex power system optimization problems characterized by nonlinearity, multimodality, and high dimensionality. Nature-inspired strategies based on adaptive biological behaviors offer significant potential to enhance search efficiency and convergence reliability. The recently published kakapo optimization algorithm (KOA) is employed in this study to address the dynamic economic dispatch (DED) problem over a 24-hour horizon in multi-unit power systems. Problem. The DED problem extends conventional economic load dispatch into a multi-hour planning horizon, considering hourly load variations, generator ramp-rate limits, valve-point effects, and transmission losses. These characteristics render DED highly nonconvex and nonlinear, posing challenges to conventional and metaheuristic techniques. Maintaining a robust balance between global exploration and local exploitation is critical to prevent premature convergence or suboptimal generation schedules. Goal. To apply kakapo optimization algorithm for the dynamic economic dispatch problem, aiming to generate economically optimal and operationally feasible generation schedules over a 24-hour dispatch horizon while preserving population diversity and search stability. Methodology. KOA models two synergistic behavioral phases of the kakapo. Exploration is inspired by lek mating and acoustic signaling, where higher-fitness solutions emit stronger «calls» that probabilistically attract weaker candidates toward promising regions. Exploitation mimics freezing and camouflage strategies, performing fine-grained local adjustments around promising solutions with adaptive step sizes. KOA is applied to a standard five-unit system over 24 hours and benchmarked against nine well-known metaheuristics. Results. KOA achieves the lowest total generation cost, rapid convergence, and high robustness. Statistical performance metrics – including mean, best, worst, standard deviation, and rank – consistently favor KOA, confirming its effectiveness for multi-dimensional, multi-modal DED problems. Scientific novelty. KOA introduces a biologically inspired, self-adaptive search framework that balances exploration and exploitation without external control parameters. Practical value. The algorithm provides a reliable, versatile, and computationally efficient optimization tool for complex power system dispatch problems, with potential applications in renewable integration, multi-objective optimization, and real-time adaptive operations. References 29, tables 4, figures 2.