| dc.description.abstract | The generation of electricity through renewable sources has been relevant to meet population growth worldwide. Thus, the type of energy from the photovoltaic solar system is one of the most applied at the moment, especially in countries with a tropical climate, as is the case in Brazil. However, it is observed that a large number of inhabitants, located in rural and even remote areas in relation to large urban centers, do not have an electricity supply due to variable issues, such as access and distance from generating sources located in the load centers. Therefore, the present work aims to propose more efficient consumer facility configurations, which use the photovoltaic system with independent or shared electricity generation, disconnection to the concessionaire’s network, also called off-grid, using the Design Construtal Method. Moreover, it aims at the projection and evolution of connection this network system of distribution of electric energy, applying iterations with Artificial Neural Networks (ANNs). As a result, it is identified that it is possible to share the infrastructure share of power generation sources of up to 16 supplies, in a planned evolution projection of design. In analysis of the RNAs, the result showed an average obtained of 4.39 supplies connected to an electric power generating source, it was observed that 25% of the iterations presented results of average3.22 deliveries, 25% presented an average of 4.89 deliveries, in this same condition. | en |
