Electricity use profiling and forecasting at microgrid level

Short-Term Load Forecasting (STLF) is nowadays a crucial and integral part of the energy production procedure to the emerging technologies for demand side management. The numerous approaches and algorithms proposed take advantage of the advances in information, metering and control technologies to address the challenges of distributed generation and intermittent energy sources on the one hand and the electricity markets on the other. This paper describes a flexible and easily customized, modular toolbox, called Divinus, for electricity use profiling and forecasting in microgrids. Divinus supports algorithms for forecasting and profiling that can be used independently or combined and its architecture consists of several interconnected well-defined components where each one interacts directly with the other. In this work, we have implemented Self-Organizing Maps for profiling and k-Neighbors for forecasting. In order to test the functionalities of the platform, we used electricity consumption data of the TEISTE campus in Evia, Greece from January 2010 till March 2018. From the tests that have been carried out so far, we have observed that the proposed approach yields high accuracy and acceptable mean errors.