Metaheuristic-Based Optimization for Cascaded Hydropower Systems

• Arnaldo M. P. Junior Radix Engineering and Software Development / UFF
• Kleyton P. Cotta Radix Engineering and Software Development
• Leonardo T. Vianna Radix Engineering and Software Development
• Tassiana N. Benamor Radix Engineering and Software Development
• Bruno J. M. Medeiros Radix Engineering and Software Development
• Maria E. L. Luz Radix Engineering and Software Development
• Matheus F. S. Pinheiro Radix Engineering and Software Development
• Maurício S. Magalhães Radix Engineering and Software Development
• Hugo R. V. Portuita Radix Engineering and Software Development
• Tassio Simioni Radix Engineering and Software Development
• Márcio L. Perin CERAN
• Paulo R. S. Munari CERAN
• Beatriz S. Pires CPFL

Resumo

This study presents a planning and optimization methodology for hydropower generation in run-of-river facilities. The research was applied to the Rio das Antas Energy Complex employing metaheuristic algorithms. This energy complex is composed of three hydropower plants with an installed capacity of 360 MW. The main objective is to improve the operation of these facilities, which is achieved through optimizing the use of available water resources. In order to attain this outcome, hydrological balance equations were applied, ensuring compliance with operational constraints while maximizing the plant’s generation efficiency. The study adopts three optimization algorithms: Particle Swarm Optimization (PSO), its variation Unified PSO (UPSO), and Differential Evolution (DE). The optimization process led to a significant improvement in the efficiency of the generating units of the power plants, surpassing the currently observed values by nearly 2%. This result demonstrates the effectiveness of the proposed approach in enhancing generation performance within the studied hydropower system. Additionally, the generating output recommended by the optimization algorithm was found to be consistent with both the scheduled dispatch and the historical generation patterns, which prioritize peak generation hours and result in fewer power unit shutdowns.