Basic Ai Models For Electric Load Forecasting: Bibliometric Analysis Approach From Scopus

DOI

10.2991/978-94-6463-694-9_40How to use a DOI?

Keywords

Bibliometric; Load forecasting; Artificial intelligence; AI; Forecasting method

Abstract

Artificial intelligence (AI) has revolutionized the field of electricity load forecasting, which is crucial for efficient power system planning and operation. The growing integration of renewable energy sources, particularly with their intermittent nature, has further highlighted the need for accurate load forecasting to optimize system performance and enhance grid stability. AI methods, such as neural networks and machine learning, have emerged as effective tools to address the complexity and nonlinearity of electric load data, outperforming traditional approaches. This paper uses a bibliometric analysis approach with the SCOPUS database, examining publications from 1997 to 2023 to analyze research trends in basic AI approaches for load forecasting. The analysis of 72 publications shows a notable rise in interest post-2018, particularly in short-term and very short-term load forecasting. Key countries leading this research include China, India, Indonesia, Australia, and the USA, reflecting their growing energy demands and efforts to integrate AI with smart grids and renewable energy systems. Recent trends show an increasing exploration of novel methods such as ensemble learning and Bayesian optimization to enhance model accuracy and stability. Future studies should explore more advanced AI techniques to better address the evolving challenges of load forecasting in the context of increasingly complex power systems.

Research purpose:

This paper aims to analyse research trends in basic artificial intelligence approaches for load forecasting in the following aspects: the number of publications over time, the key countries, the main authors.

Research motivation:

Artificial intelligence (AI) has revolutionized the field of electricity load forecasting, which is crucial for efficient power system planning and operation. The growing integration of renewable energy sources, particularly with their intermittent nature, has further highlighted the need for accurate load forecasting to optimize system performance and enhance grid stability. Investigating the current application of AI in electricity load forecasting is essential to identify future trends in this area.

Research design, approach, and method:

In this study, bibliometric analysis was employed to investigate global scholarly interest in applying AI to electrical load forecasting. Data was extracted from the Scopus database, examining publications from 1997 to 2023. The analysis was conducted using Biblioshiny, VOSviewer, and Microsoft Excel to assess performance metrics, keyword co-occurrence, and international scientific collaboration.

Main findings:

The research finds increasing attention to AI-based electric load forecasting from 1997 to 2023, with a notable spike in publication productivity after 2018. Analysing 72 publications, the study highlights significant contributions from China, India, Indonesia, Australia, and the USA, driven by rising energy demand. Advanced techniques like LSTM, deep learning, and ensemble learning have gained prominence since 2017, showing improvements in accuracy and efficiency for short-term load forecasting in smart grid systems.

Practical/managerial implications:

AI-based load forecasting can significantly support power system operation by providing more accurate predictions, especially with the rise of renewable energy integration. Bibliometric analysis shows a growing interest in the research topic of adopting basic AI techniques, such as neural networks and machine learning, to help power companies optimize energy distribution, reduce operational costs, and enhance grid stability. Considering AI models in load forecasting is necessary for countries with an increasing portion of renewable energy in their power systems.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Nguyen Hoang Lan
AU  - Nguyen Thi Huyen Trang
AU  - Ha Thu Hang
AU  - Toan-Vu Le
PY  - 2025
DA  - 2025/04/30
TI  - Basic Ai Models For Electric Load Forecasting: Bibliometric Analysis Approach From Scopus
BT  - Proceedings of the International Conference on Emerging Challenges: Sustainable Strategies in the Data-driven Economy (ICECH 2024)
PB  - Atlantis Press
SP  - 597
EP  - 612
SN  - 2352-5428
UR  - https://doi.org/10.2991/978-94-6463-694-9_40
DO  - 10.2991/978-94-6463-694-9_40
ID  - Lan2025
ER  -