Reconfiguração e restauração ótima de sistemas de distribuição primária de energia elétrica

• Ano de defesa: 2012
• Autor(a) principal: Carlos Henrique Nogueira de Resende Barbosa
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Reconfiguração; Restauração; Sistemas de distribuição; Otimização multiobjetivo; Confiabilidade; Engenharia elétrica
• Link de acesso: http://hdl.handle.net/1843/BUOS-97HFBK

Smart reconfiguration on Electric Power Primary Distribution Systems has been playing a key role in enabling power utilities to comply with the ever increasing requirements imposed by the regulatory agency and government, the deregulated market, and consumers. By these circumstances, power utilities must search for a tradeoff among technical and economic factors, quality of service, security, to name a few aspects. The combinatorial problem of identifying optimal network configurations has been investigated for fifty years at least, but opportunities for improvements can still be envisioned. Particularly in this work, a methodology to tackle reconfiguration and restoration problems for three-phase unbalanced radial feeders using a multiobjective approach with real-time requirements was developed. Achieved solutions ensure the optimal or near-optimal operating conditions to the electrical system evaluated. On the reconfiguration problem, the solution is aimed at the best compromise between available energy and demand plus power losses, the best loading distribution, the flattest voltage profile, among other criteria, as well as it prioritizes safer configurations in terms of operating conditions. On restoration, priority is given to the network configuration able to be carried out in the shortest time, reenergyzing the maximum number of loads without causing overloads, being much similar to the pre-fault configuration and oriented to service availability and redundant equipment. Complementarily, technical or operational constraints may not be violated. Both the sharp searching process for feasible space and the network configuration assessment by means of power flow calculation were addressed. A reasoning scheme for maneuvering sequence was devised to minimally ensure safety and feasibility of the solution and a decision-making procedure was used to classify efficient solutions according to user preferences. Results from computational experiments found for five systems showed that the proposed improvements are effective as well as promising in identifying the optimal configurations in small and medium size, single or many-feeder systems. Comparisons made to other published simulation results were carried out. In some cases, as the 70-bus test system, a better solution than the previous known in the literature was found.