Probabilistic Forecasting of Energy Consumption using Bayesian Dynamic Linear Models

This study aims to conduct a systematic literature review on the development of mathematical models for forecasting energy consumption using a probabilistic approach, particularly focusing on the Bayesian Dynamic Linear Model (BDLM). The research method employed is Systematic Literature Review (SLR)...

Full description

Saved in:
Bibliographic Details
Main Authors: Hartika Aulia, Syaharuddin Syaharuddin, Vera Mandailina, Hamenyimana Emanuel Gervas, Hameed Ashraf
Format: Article
Language:English
Published: Syiah Kuala University 2024-08-01
Series:Aceh International Journal of Science and Technology
Subjects:
Online Access:https://jurnal.usk.ac.id/AIJST/article/view/38291
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study aims to conduct a systematic literature review on the development of mathematical models for forecasting energy consumption using a probabilistic approach, particularly focusing on the Bayesian Dynamic Linear Model (BDLM). The research method employed is Systematic Literature Review (SLR), utilizing literature sources indexed in Scopus, DOAJ, and Google Scholar, with publication dates ranging from 2014 to 2024. The findings of the research indicate that the application of BDLM has made a significant contribution to the optimization of energy management, especially in sectors such as industry and commercial buildings. The study highlights the effectiveness of BDLM in accurately predicting energy consumption through a probabilistic approach that efficiently manages uncertainty. However, the research also emphasizes that BDLM presents limitations and challenges that warrant attention, including the complexity involved in parameter determination and model validation processes, as well as the importance of addressing potential biases and considering factors such as deployment impacts. This research provides deep insights into the potential and challenges in the development of mathematical models for forecasting energy consumption, while also offering directions for further research in this field.
ISSN:2088-9860