A meta-heurística de busca dispersa aplicada no planejamento da expansão de sistemas de transmissão

• Autor(a) principal: Arcari, Inedio [UNESP]
• Data de Publicação: 2014
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/111130
• topic: Energia elétrica - Distribuição; Algoritmos; Energia eletrica - Transmissão; Electric power distribution

This work presents a theoretical analysis and computational implementation of a specialized Scatter Search algorithm to solve the static transmission network expansion planning (TNEP) problem of electric power systems. The objective of such planning problems is to determine a set of circuits among the candidates in which not only satisfy the demands but also the minimum investment cost is at hand. This problem is considered as a complex mixed integer nonlinear programming (MINLP) problem that has a lot of local optimum problem. The scatter search is an evolutionary method with the objective of maintaining a set of diverse and high-quality candidate solutions. The proposed scatter search algorithmhas been applied in engineering optimization problems especially in electric power system problems and has presented high quality solutions. The diversity sets ensure to avoid getting trapped in a local optimum. Another important factor is that the proposed methodology reduces the search space and consequently the number of combinations is reduced. In this work, a high quality solution of TNEP is obtained using the greedy constructive heuristic algorithms such as Garver, and Villasana-Garver-Salon that work based on Transport model and DC model respectively. In this work, in order to generate the initial solutions, a controlled disturbance has been added in the costs of the transmission lines in order to obtain diverse and high quality solutions that lead to find the global optimum for some problems even in the initial generation step. Moreover, the proposed scatter search algorithm presents a local improvement phase during the implementation. In order to show the effectiveness of the proposed algorithm, 5 case studies are conducted such as Garver 6-bars and 15 branches , IEEE 24-bars and 41 branches , South Brazilian 46-bars and 79 branches, Colombian 93-bars and 155 branches, and the North-Northeast 87-bars ...