Energy consumption prediction through linear and non-linear baseline energy model

• Published in: Indonesian Journal of Electrical Engineering and Computer Science
• Main Author: Mustapa R.F.; Dahlan N.Y.; Yassin I.M.; Mohd Nordin A.H.; Zabidi A.
• Format: Review
• Language: English
• Published: Institute of Advanced Engineering and Science 2019
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85073815574&doi=10.11591%2fijeecs.v17.i1.pp102-109&partnerID=40&md5=1a679d220513390f7aade201f39116e1

Accurate baseline energy models demand increase significantly as it lower the risk of energy savings quantification. It is achieved by performing energy consumption prediction with its respective independent variables through linear or non-linear modelling technique. Developing such model through linear modelling technique provide certain disadvantages due to the fact that the behavior of certain independent variables with respect to the energy consumption is non-linear in nature. Furthermore, linear modelling technique requires prior studies upon modelling to achieve accurate energy consumption prediction. Thus, to apprehend this situation, this paper main intention is to perform energy consumption prediction through a non-linear modelling technique to provide alternative option for developing a good and accurate baseline energy models. This study proposes energy consumption prediction based on Non-linear Auto Regressive with Exogenous Input – Artificial Neural Network (NARX-ANN) as a non-linear modelling technique that will be compared with Multiple Linear Regression Model (MLR) as linear modelling technique. A case study in Malaysian educational buildings during lecture week will be used for this purpose. The results demonstrate that NARX-ANN shows a higher accuracy through statistical error measurement. Copyright © 2020 Institute of Advanced Engineering and Science. All rights reserved.