The team of Associate Professor Yao Zhenpeng of Shanghai Jiaotong University published a review paper on AI-accelerated material discovery in Nature Reviews Materials

Recently, the internationally renowned academic journal Nature Reviews Materials published online the review article "Machine learning for a sustainable energy future" by the team and cooperation team of Associate Professor Yao Zhenpeng of the School of Materials Science and Engineering of Shanghai Jiao Tong University, which provides a forward-looking direction
for the related promotion of machine learning in energy materials, equipment, management and other fields 。 The paper is based on Shanghai Jiao Tong University as the first author, and Associate Professor Yao Zhenpeng from the School of Materials Science and Engineering of Shanghai Jiao Tong University is the first author
.

The transition from fossil to renewable energy is a major global challenge that requires advances across the energy industry at the material, equipment and systems levels to enable efficient collection, storage, conversion and management
of renewable energy.
Researchers in the energy sector have begun to use machine learning techniques to enable these advances (Figure 1).

Figure 1 Traditional and machine learning-accelerated material development paradigms

In the material review, the team highlights the latest advances in energy research driven by machine learning, outlines current challenges and looks ahead, and describes the prerequisites needed to take full advantage of machine learning techniques
.
The team introduced a set of Materials Accelerated Development Performance Indicators (XPIs) to compare the differences and improvement opportunities
of different machine learning paradigms for energy research advancement.
Recent advances
in the application of machine learning to energy harvesting (photovoltaics), storage (batteries), conversion (electrocatalysis), and management (smart grids) were also discussed and evaluated.
Finally, a potential area of research for machine learning in the energy sector is outlined (Figure 2).

Figure 2 Application development direction of machine learning in the renewable field

In addition, the team also published a review "On scientific understanding with artificial intelligence" in Nature Reviews Physics at the same time, which systematically summarized the latest progress
of artificial intelligence in the field of promoting the establishment of scientific theories 。 In recent years, the team has carried out extensive research in the fields of electrochemical energy storage, high-throughput experiments and calculations, and deep machine learning, and has successively conducted research in Science, Nature Energy, Nature Catalysis, Nature Machine Intelligence, Science Advances, Nature Communications, Matter, and Accounts of A series of research papers
have been published in academic journals such as Chemical Research.