Even if renewable energy generation is improving and diffusing rapidly, reliable energy access is still a major issue for a consistent part of the global population, with more than 1 billion people still lacking energy access globally. The vast majority of this share of population is living in remote rural areas of developing countries, experiencing major issues in terms of living conditions. While consistent efforts have been done in the past decades to solve the problem, still a lot of work has to be done and novel approaches need to be implemented.
In the past, most of the new energy connections were achieved through national grid extension, which is proving to be a non-adequate short-term solution for a consistent share of the remaining part of the population living in rural areas. This is the reason why decentralised solutions, such as Solar Home Systems and DC micro-grids, are becoming more appealing as alternative ways to improve energy access in developing countries.
In this framework, this Master's thesis will focus on DC solar micro-grids as a solution to the energy access problem. More specifically, the aim will be to develop a methodology to gather, process and analyse data, for planning and evaluation of remote DC micro-grid networks in rural areas of developing countries. One of the main novelty aspects of this proposed methodology is the integrated implementation of Geographic Information Systems and concepts derived from the mathematical field of Graph Theory, together with an electrical analysis.
The methodology is clearly divided into three consecutive steps. The first step focuses on gathering and processing ground-level data using GIS, to compare different micro-grid layouts in term of geometrical length. The second step consists of a graph theory-based dual-objective optimisation algorithm to design meshed micro-grids from a set of starting topologies. The third step implements a DC power flow tool to analyse the operational behaviour of the optimised layouts. The proposed methodology is explained in detail throughout the report, with an example of its application to a sample of villages in different world-wide locations.
The results of this first application of the proposed methodology allow to draw some conclusions on the methodology itself and on the comparison of different micro-grid topologies. First of all, the huge potential of the combination of GIS tools and graph theory applied to micro-grid planning is shown. The results of the layout comparison show how typically implemented micro-grid layouts are generally outperformed by micro-grids designed using novel concepts and this integrated approach. Nonetheless, each specific case studies has peculiar characteristics and conditions that need to be taken carefully into account and can lead to totally different kinds of optimal solutions. It is hence of vital importance to have a methodology which is at the same time well-structured and flexible to adapt to changes and modification of parameters in order to perfectly reflect the specific needs and characteristics of each different rural electrification project.