Large Language Models in Novel Power Systems: Theories, Methods, and Applications

Submission Deadline: 01 May 2027 View: 60 Submit to Special Issue

Guest Editor(s)

Prof. Haixiang Zang

Email: zanghaixiang@hhu.edu.cn

Affiliation: School of Electrical and Power Engineering, Hohai University, Nanjing, China

Homepage:

Research Interests: renewable energy integration, application of artificial intelligence in power systems

Prof. Wei Liu

Email: wliu@njust.edu.cn

Affiliation: School of Automation, Nanjing University of Science and Technology, Nanjing, China

Homepage:

Research Interests: AI-based control of distribution networks & micro-grids, data-model-driven assessment & optimization

Dr. Lilin Cheng

Email: straw@hhu.edu.cn

Affiliation: School of Electrical and Power Engineering, Hohai University, Nanjing, China

Homepage:

Research Interests: renewable power forecasting, smart distribution powergrid optimization

Summary

With the widespread integration of distributed renewable energy sources and the introduction of AI technologies, novel power systems have undergone significant changes in operation modes, data scale, and security requirements. Traditional system analyzing and decision-making tools are gradually showing limitations. Meanwhile, massive amounts of multimodal data are generated during power system operation, which have not yet been fully exploited. In recent years, large language models (LLMs) represented by GPT have achieved breakthrough progress, demonstrating powerful capabilities in logical reasoning, cross-task generalization, and multi-modal fusion. However, their application in power systems still faces a series of core challenges, such as the lack of domain-specific power system knowledge, the difficulty of meeting the high reliability requirements of power systems in terms of model interpretability and security, and the challenge of coordinating computing power and electric power resources.

This Special Issue aims to bring together original research results at the intersection of LLMs and power systems, promoting the deepening of LLM technologies into power engineering practice such as renewable energy integration, hybrid energy systems, and smart grids.

For the convenience of manuscript submission, we encourage (but is not limited to) the following specific themes:
- LLM-driven forecasting for power system load and renewable power output
- Novel power system optimal dispatch and operation based on LLM-driven agents
- LLM-based applications in power equipment fault diagnosis and maintenance
- LLM-assisted decision-making risk management for electricity market trading
- Multimodal data fusion and process in novel power systems
- Interpretability assessment of domain-specific LLMs in novel power systems

Keywords