Scholarlink

<p>With the global energy system transitioning to renewable energy, high-efficiency energy storage and conversion technologies have become crucial. However, traditional research paradigms for the research and development (R&amp;D) of energy materials such as batteries and electrocatalysts present the limitations in efficiency. This review systematically summarizes the progress of artificial intelligent (AI) in this field, ranging from classical machine learning (ML) to advanced representation methods such as graph neural networks (GNNs) and transformers that enable precise property prediction and structure generation. It also covers generative models for inverse design and large language models (LLMs) for knowledge extraction, along with key domain databases. Current challenges include limited interpretability and the underutilization of emerging AI technologies. Finally, this review discusses future directions such as the applications of multimodal language models, aiming to provide insights for accelerating high-performance energy materials innovation and advancing the global renewable energy transition.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence for energy materials research: From classical machine learning to large models

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月15日

查看全文
  • Mingxi Jiang,
  • Jie Zhou,
  • Yanggang An,
  • Zhengran Lin,
  • Menghao Yang

摘要

With the global energy system transitioning to renewable energy, high-efficiency energy storage and conversion technologies have become crucial. However, traditional research paradigms for the research and development (R&D) of energy materials such as batteries and electrocatalysts present the limitations in efficiency. This review systematically summarizes the progress of artificial intelligent (AI) in this field, ranging from classical machine learning (ML) to advanced representation methods such as graph neural networks (GNNs) and transformers that enable precise property prediction and structure generation. It also covers generative models for inverse design and large language models (LLMs) for knowledge extraction, along with key domain databases. Current challenges include limited interpretability and the underutilization of emerging AI technologies. Finally, this review discusses future directions such as the applications of multimodal language models, aiming to provide insights for accelerating high-performance energy materials innovation and advancing the global renewable energy transition.