Neural network based load forecasting and fuzzy logic EMS for Ngarenanyuki school microgrid

During the last decade a lot of efforts have been put into promoting the access to electricity to those millions of people living in rural un-electrified areas of underdeveloped countries. Despite the huge potential market of new consumers, the progress of electrification rates in such areas is very limited. Besides evident economic constraints, also lack in appropriate and replicable technical solutions contributes in hindering the process of electrification. As regards this issue, R&D activities in off-grid hybrid and combined Renewable Energy (RE) systems, which are often the most proper option, should also focus on solutions which can easily adapt to the extremely different local conditions thus reducing capital costs while making the electric service more affordable. In such a panorama, this paper presents the results of a research project, financed by a University Research Fund, which has the goal of realizing a hybrid micro-grid with a smart Energy Management System (EMS) in Ngarenanyuki (Tanzania) secondary school.