Otimização multiobjetivo usando Algoritmo Genético de Classificação Não Dominada para a operação ótima de Pequenas Centrais Hidrelétricas em cascata

Abstract

Brazil has one of the most diversified energy matrices in the world, with a significant share of renewable sources, particularly hydropower. Given the increasing demand for electricity, the construction of large Hydroelectric Power Plants (HPPs) presents contemporary challenges, especially considering environmental, social, cultural, and regulatory issues. Therefore, there is a growing need to focus on small-scale projects, such as Small Hydroelectric Power Plants (SHPPs), as well as on optimizing electricity generation. This study aims to develop a multi-objective model for the optimal operation of a cascade of SHPPs along the Sapucaí Paulista River, namely Dourados, São Joaquim, Palmeiras, and Retiro. The methodology consists of applying the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to a multi-objective model based on the functions of generated power, Operation and Maintenance (O&M) cost, and head losses across the cascade of SHPPs. The turbine and generator efficiency curves revealed that operators can identify an efficient operating range for the system, enabling power generation with reduced water resource consumption. This methodology proves to be innovative in the context of SHPPs, supported by statistical tests applied to validate the model’s objective functions. The NSGA-II results showed a significant increase in power generation in most months of the year, particularly in the second semester when energy resources are more constrained. Compared to the actual power generated by the facilities, the gains in May and June were 6% and 10%, respectively. In July, August, September, and October, gains exceeded 100%, while November and December saw gains of 88% and 64%, respectively. The application of NSGA-II proves to be an excellent technique to support decision-making and enhance the efficiency of electricity generation in SHPPs.