大学化学 >> 2025, Vol. 40 >> Issue (9): 69-75.doi: 10.12461/PKU.DXHX202411067

专题 上一篇    下一篇

高通量计算与机器学习相结合的综合计算化学实验设计与实践

周佳   

  1. 哈尔滨工业大学(深圳)理学院, 城市水资源与水环境国家重点实验室, 广东 深圳 518055
  • 收稿日期:2024-11-19 录用日期:2025-01-02 发布日期:2025-09-16
  • 通讯作者: 周佳 E-mail:jiazhou@hit.edu.cn
  • 基金资助:
    哈尔滨工业大学深圳校区质量工程项目(高等教育教学改革项目)(HITSZERP22009)；哈尔滨工业大学深圳校区思政课程和课程思政专项课题(HITSZIP22017)

Design and Practice of a Comprehensive Computational Chemistry Experiment Based on High-Throughput Computation and Machine Learning

Jia Zhou   

  1. State Key Laboratory of Urban Water Resource and Environment, School of Science, Harbin Institute of Technology (Shenzhen), Shenzhen 518055, Guangdong Province, China
  • Received:2024-11-19 Accepted:2025-01-02 Published:2025-09-16
  • Contact: Jia Zhou E-mail:jiazhou@hit.edu.cn

PDF

280

赞

摘要： 随着计算化学和人工智能技术的迅猛进步，它们在化学教育领域的应用愈发关键。本实验项目专门为化学高年级本科生及研究生搭建了一个综合性的实验平台，将计算化学与机器学习方法加以融合，对有机化合物的键解离能展开深入探究。课程内容包含量子化学计算方法的基础原理、实操技巧，还有机器学习模型的构建、训练以及验证等多个方面。学生将借由实际操作，学会运用先进的计算工具和算法来预测与分析化学键能，进而增进对化学反应机理的深度认知。课程的目标在于让学生不但能够熟练掌控数据处理和分析的技能，并且能够独立凭借这些技能从事化学问题的研究，为日后的科研工作或者跨学科领域的探索筑牢根基。

关键词: 键解离能, 计算化学, 大数据, 机器学习, SMILES

Abstract: With the rapid advancements in computational chemistry and artificial intelligence technologies, their integration into chemical education has become increasingly vital. This experimental course is specifically tailored for senior undergraduate and graduate chemistry students, providing a comprehensive platform that merges computational chemistry with machine learning methodologies to explore the bond dissociation energies of organic compounds in depth. The curriculum covers fundamental principles and operational techniques of quantum chemical calculation methods, as well as the construction, training, and validation of machine learning models. Through hands-on experience, students will learn to utilize advanced computational tools and algorithms to predict and analyze chemical bond energies, thereby deepening their understanding of chemical reaction mechanisms. The objective of the course is to equip students with proficient data processing and analysis skills, empowering them to independently apply these skills to research chemical problems, thus establishing a strong foundation for future scientific endeavors or interdisciplinary explorations.

Key words: Bond dissociation energy, Computational chemistry, Big data, Machine learning, SMILES