Global energy consumption is growing, and traditional fossil energy sources are environmentally unfriendly and non-renewable. Energy consumption and carbon emissions have become major challenges for sustainable green development. Renewable energy sources such as wind, solar, and nuclear power have emerged as viable alternatives to fossil fuels. Although these clean energy sources have long been applied in energy production, their output capacity is constrained by factors like weather conditions, making stable electricity generation difficult. Combined with high equipment maintenance costs and the energy and economic losses incurred during outages, these limitations hinder the development of clean energy. However, now the continuous advancement of artificial intelligence technology offers new pathways to enhance the resilience of renewable energy systems and support the transition to a green, low-carbon economy.
Generative artificial intelligence (Generative AI) refers to a machine learning model that generates novel and original content without explicit programming. It has been widely used in different fields and has had a huge impact on several industries such as entertainment, finance, and education. The iterative development speed of generative AI has surpassed our imagination. Models such as Google's Gemini series can perform various complex tasks, demonstrating outstanding capabilities in mathematical reasoning and multimodal processing. Developers have even introduced agent models capable of autonomously executing multi-step tasks and achieving practical implementation. These advancements unlock immense potential for generative AI in intelligent renewable energy systems.
This article entitled “Generative AI-based spatiotemporal resilience, green and low-carbon transformation strategy of smart renewable energy systems”, which was published in the journal Frontiers of Engineering Management, systematically explores the role of generative AI in smart renewable energy systems. It delves into the lifecycle development of both critical and non-critical subsystems within renewable energy systems, examines generative AI application models across multiple scenarios, and constructs a comprehensive analytical framework covering planning, monitoring, and maintenance optimization.
Renewable energy systems may fail due to diverse mechanisms originating from both internal and external factors. This paper proposes a framework that distinguishes factors such as the type of external impact events and the criticality of internal subsystems, exploring spatiotemporal resilience under various failure modes. Certain external impact events directly destroy subsystem functionality, causing overall failure—these are termed critical incidents. Others degrade system performance without causing shutdown—these are termed impact incidents. Still others have minimal effects—these are termed minor incidents. By deeply analyzing the impacts on critical and non-critical subsystems under these three event types, this paper constructs a maintenance cost model under multiple scenarios and further proposes a spatiotemporal resilience model.
Using system data, generative AI can analyze and forecast energy consumption and carbon emissions from renewable energy systems by leveraging their operational data sets. Furthermore, combining external information, generative AI can perform big data analysis on energy production and consumption, dynamically feedback energy information, and accurately capture the supply and demand information. It can respond quickly to supply and demand information and operating conditions, and adaptively adjust the optimal strategies to avoid additional carbon emissions. Synthesizing internal and external information, generative AI can promote sustainable green and low-carbon practices across renewable energy systems.
In conclusion, this paper explores how generative AI can be efficiently integrated with renewable energy, proposing intelligent maintenance strategies under different failure modes and a green and low-carbon transformation strategy. It provides a paradigm for achieving precise regulation of energy consumption and carbon emissions, thereby promoting sustainable system development.
DOI：
10.1007/s42524-025-4147-6