摘要
量子机器学习融合了量子化学与机器学习的优点,具有比传统密度泛函理论更快的计算速度和更高的准确性.量子机器学习可为复杂、多维、多尺度的催化化学提供更智能和有效的研究方式,通过训练可靠的数据及建立合理的模型和算法,快速、准确地预测最优的催化剂设计参数、最佳的催化剂材料的合成方法和反应条件、以及催化剂结构和性能之间的关系.作者就量子机器学习应用于催化材料的设计、催化反应性能和催化反应机理三方面的发展趋势进行了概述.
Quantum machine learning,which combines the advantages of quantum chemistry and machine learning,possesses faster calculation speed,and higher accuracy than traditional density functional theory.Quantum machine learning provides more intelligent and effective research approaches for complex,multi-dimensional and multi-scale catalytic chemistry.Furthermore,the optimal catalyst design parameters,the optimal synthesis method and reaction conditions of catalyst materials,and the relationship between catalyst structure and performance could be quickly and accurately predicted by training reliable data and establishing reasonable models and algorithms.In this protocol,the application of quantum machine learning to catalytic materials design,catalytic reaction performance and catalytic reaction mechanism are summarized.
作者
钱波
QIAN Bo(State Key Laboratory for Oxo Synthesis and Selective Oxidation,Lanzhou Institute of Chemical Physics,Chinese Academy of Sciences,Lanzhou 730000,China)
出处
《分子催化》
CAS
CSCD
北大核心
2023年第1期63-72,I0003,I0004,共12页
Journal of Molecular Catalysis(China)
基金
中国科学院“西部之光”人才培养引进计划。
关键词
量子机器学习
催化材料设计
催化反应性能
催化反应机理
quantum machine learning
catalytic materials design
catalytic reaction performance
catalytic reaction mechanism
