The Number of Nodes Effect to Predict the Electrical Consumption in Seven Distinct Countries
This paper presents a machine learning-based approach for forecasting electrical consumption in seven selected countries across different geographical categories. The data, sourced from The International Energy Agency, is analysed and condensed to focus on specific nations: Northern (Norway, Canada)...
| Main Authors: | , , , |
|---|---|
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Politeknik Negeri Malang
2023
|
| Subjects: | |
| Online Access: | https://jartel.polinema.ac.id/index.php/jartel/article/view/826 https://doi.org/10.33795/jartel.v13i4.826 |
| Summary: | This paper presents a machine learning-based approach for forecasting electrical consumption in seven selected countries across different geographical categories. The data, sourced from The International Energy Agency, is analysed and condensed to focus on specific nations: Northern (Norway, Canada), Southern (Chile, Australia), Four-season (France, Japan), and a Tropical country (Colombia). The unique electrical consumption patterns influenced by regional climate characteristics make this study compelling for machine learning applications. From the dataset comprising over 132,000 records from January 2010 to May 2023 across 53 countries, a refined dataset focusing on 791 data points from seven specifically chosen countries to simplify the study. A significant part of the paper details the machine learning design for electrical consumption forecasting. Specifically, Artificial neural network architecture is proposed to predict consumption. The input features encompass the year, month, and country, with the output being the anticipated electrical usage. |
|---|