| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 5.4 MB | Adobe PDF |
Authors
Advisor(s)
Abstract(s)
O objetivo da presente dissertação foi desenvolver algoritmos de reconfiguração topológica de redes de distribuição de energia elétrica para minimização de perdas. Para o efeito são apresentados dois algoritmos.
O primeiro algoritmo apresentado é baseado num modelo heurístico em que a procura da melhor topologia da rede é feita tendo em consideração a identificação de caminhos topológicos conexos que minimizam as perdas. Estes caminhos são realizados iterativamente entre os barramentos adjacentes de acordo com as regras estabelecidas no algoritmo.
Tendo em consideração a natureza combinatória do problema foi desenvolvido um segundo algoritmo que teve como base um modelo evolutivo, algoritmo genético. O objetivo é alargar o espaço de buscas de soluções. Posteriormente os AG’s foram combinados com um modelo heurístico, initial searching point, para obter soluções ainda mais adaptadas ao problema e com bom desempenho.
Os dois algoritmos foram desenvolvidos em Python. A biblioteca Pandapower foi utilizada para modelar a rede em estudo e para os cálculos de trânsito de energia.
Estes algoritmos foram aplicados à rede IEEE 33 barramentos e ambos tiveram redução de perdas superiores a 30%. A redução de perdas resultante da aplicação do primeiro algoritmo foi de 31% e, do segundo algoritmo, a redução de perdas foi de 35%.
The objective of the present dissertation is to develop algorithms for the topological reconfiguration of electrical energy distribution networks to minimize losses. For this purpose, two algorithms are presented. The first one is based on a heuristic model in which the search for the best network topology takes into account the identification of connected topological paths that minimize losses. These paths are obtained iteratively between adjacent buses according to the rules established in the algorithm. Due to the combinatorial nature of the problem, a second algorithm was developed. This one is based on an evolutionary model, the genetic algorithm. Its objective is to expand the space for searching for solutions. Subsequently, the GA was combined with a heuristic model, the initial searching point. The solutions obtained were even more adapted to the problem and showed a better performance. Both algorithms use Python as their coding language. And the library used to model the network under study and to calculate the energy transit was Pandapower. They were applied to the IEEE 33 bus network, and both achieved a loss reduction of over 30%. The first one resulted in a reduction of 31% and the second one a reduction of 35%.
The objective of the present dissertation is to develop algorithms for the topological reconfiguration of electrical energy distribution networks to minimize losses. For this purpose, two algorithms are presented. The first one is based on a heuristic model in which the search for the best network topology takes into account the identification of connected topological paths that minimize losses. These paths are obtained iteratively between adjacent buses according to the rules established in the algorithm. Due to the combinatorial nature of the problem, a second algorithm was developed. This one is based on an evolutionary model, the genetic algorithm. Its objective is to expand the space for searching for solutions. Subsequently, the GA was combined with a heuristic model, the initial searching point. The solutions obtained were even more adapted to the problem and showed a better performance. Both algorithms use Python as their coding language. And the library used to model the network under study and to calculate the energy transit was Pandapower. They were applied to the IEEE 33 bus network, and both achieved a loss reduction of over 30%. The first one resulted in a reduction of 31% and the second one a reduction of 35%.
Description
Dissertação de natureza científica para obtenção do grau de Mestre em Engenharia Eletrotécnica
Keywords
Reconfiguração de redes de distribuição Perdas Algoritmos Genéticos Pandapower Reconfiguration of electrical energy distribution networks Energy loss Genetic Algorithm
Citation
COSME, Eveni Pereira – Reconfiguração topológica de redes para redução de perdas. Lisboa: Instituto Superior de Engenharia de Lisboa. 2023. Dissertação de Mestrado.
Publisher
Instituto Superior de Engenharia de Lisboa