Aplicação de Meta-heurísticas no Dimensionamento de Sistemas de Backup Híbridos para a Manutenção de Serviços Auxiliares em Subestações

Abstract

Auxiliary Services in substations are composed of fundamental systems for the operation and coordination of the electrical system. In cases of contingency in the main supply they require the maintenance of their loads to ensure the recomposition process, which is why backup systems are used, composed of alternative power sources that often use generator sets supplied by fossil fuels. Another alternative is the adoption of a micro-grid composed of a hybrid system, with renewable sources and batteries, which make the system more sustainable. However, this option requires a careful analysis for its implementation due to its cost and the intermittency of the generating sources. In this sense, this paper proposes two algorithms for sizing a microgrid, based on the Genetic Algorithm and Variable Neighborhood Search metaheuristics, which consider uncertainties in terms of the intermittency of generating sources and the duration of power outages through Monte Carlo simulations. The performance of the methods was analyzed through the results delivered by each algorithm in a sequence of execution repetitions, looking at the investment cost and execution time. The results obtained by the algorithms developed showed promise, finding equivalent and more attractive solutions in monetary terms compared to other methods used in specialized literature, with a resolution time of just a few seconds. A reduction in investment costs of more than 50% was observed when wind turbines were added to the problem.