Large language models in renewable energy systems: A comprehensive review of forecasting, control, policy, and fault diagnosis

The change toward a sustainable energy future demands intelligent solutions for planning, forecasting, and optimization in renewable energy systems. Recent advances in Artificial Intelligence (AI), particularly Large Language Models (LLMs) such as generative pretrained transformer, bidirectional encoder representations from transformers, large language model meta AI, and Falcon, provide new opportunities by enabling robust analysis of structured and unstructured data. LLMs can interpret technical documents, make heterogeneous datasets, and support applications like forecasting, fault detection, smart grid control, policy analysis, and public engagement. This paper systematically reviews the role of LLMs across renewable energy domains, classifies applications, and highlights their potential to bridge gaps in data-driven decision-making. We propose a taxonomy of LLM applications, analyze use cases, assess limitations, and outline future research trends. By integrating cutting-edge AI methods with sustainability goals, this work highlights the role of LLMs in building resilient, intelligent, and climate-conscious energy ecosystems.