Metodologia para otimização de investimento e confiabilidade no planejamento de sistemas de distribuição de eletricidade

• Ano de defesa: 2021
• Autor(a) principal: Gomes, Allan Costa
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Palavras-chave em Português: Distribuição; Planejamento; Confiabilidade; Programação matemática; Ótimo de Pareto
• Link de acesso: http://www.repositorio.ufc.br/handle/riufc/59676

Electrical distribution systems are responsible for the uninterrupted supply of electricity to consumers. Therefore, these systems must have reliability among one of their qualities. The use of Protection and Sectionalizing Equipments (PSEs) in Electrical Distribution Systems (EDSs) is one of the main options used to ensure reliability. Through protection, it is possible to deenergized the system so that faults in the EDSs can be safely cleared, while through sectionalizing actions, it is possible to isolate the sectors affected by the contingencies, allowing the restoration of the service to the healthy sectors. However, the PSEs represent costs that must be considered by the planners of the EDSs. Then, it is notable the problem of how to use these assets optimally, considering both the finance and reliability aspects of EDSs. This dissertation aims to address the problem of determining the optimal number and placement of PSEs to optimize investment and reliability of EDSs. It is proposed for this problem a method based on a mixed integer linear programming model that allows obtaining the Pareto front between the objectives of cost and reliability, the reliability being measured by the Energy not Served (ENS) indicator. The model considers the simultaneous placement of manual and remote sectionalizing switches. The proposed method is a tool for the EDSs planner and helps the decision-making process for the placement of PSEs between the two conflicting objectives, cost and reliability. The method is applied to the Roy Billinton Text System 4 (RBTS4) and the Pareto front defined by a set of undominated optimal solutions is obtained as a result. Additionally, experiments are carried out to assess the impact of load transfer, simultaneous or non-simultaneous placement of PSEs, cost variation of remote switches and the operating time of manual switches. The results obtained indicate that the use of PSEs can drastically reduce the amount of ENS by up to 80% for the RBTS4, while the complete survey of the Pareto front can be performed in up to 23 minutes