ANN FOR THE ENERGY CONSUMPTION FORECASTING IN BUSINESS BUILDING

Abstract

In recent years, analysing the data gathered during production process of the energy has become an important issue in the energy sector in order to increase the efficiency of the produced energy. In the predictive machine learning approach, the main goal is generalisation. This means the ability of the model to give sensible predictions for situations not identical to what has already been observed in the training data. The main factors determining the predictive performance are  relevance of  sensory features, the amount of training data available and the adaptation of the complexity of the model to the task at hand. For the purpose of energy conservation, we present in this paper an introduction to the use of learning machines used as a data mining tool applied to buildings energy consumption data from a measurement campaign. The learning stage was done for a first part of the data and the predictions were done for the last month.Performances of the model & contributions of significant factors were also derived. The results show good performances for the model.