大学化学 >> 2026, Vol. 41 >> Issue (1): 76-84.doi: 10.12461/PKU.DXHX202506023

所属专题： 化学实验数字化设计竞赛获奖作品

专题 上一篇    下一篇

机器学习优化的微色谱柱离子交换色谱法测定痕量砷实验

常灵宇1, 郎艳芳1, 朱玉妍1, 王婕1, 郭莹1, 王蝶2, 丁鹏3, 周跃明1, 龚治湘1, 刘淑娟1   

  1. 1 东华理工大学化学与材料学院, 南昌 330013;
    2 湘西民族职业技术学院现代农业学院, 湖南 湘西 416099;
    3 东华理工大学信息工程学院, 南昌 330013
  • 收稿日期:2025-06-03 录用日期:2025-10-20 发布日期:2025-12-30
  • 通讯作者: 周跃明 E-mail:ymzhou@ecut.edu.cn Yueming Zhou
  • 基金资助:
    江西省高等学校教学改革研究省级课题(JXJG-23-6-34)；东华理工大学实验技术开发项目(DHSY-202313, DHSY-202511)；东华理工大学教学改革研究课题(DHJG-23-33)

Machine Learning-Optimized Microcolumn Ion Exchange Chromatography for Trace Arsenic Determination

Lingyu Chang1, Yanfang Lang1, Yuyan Zhu1, Jie Wang1, Ying Guo1, Die Wang2, Peng Ding3, Yueming Zhou1, Zhixiang Gong1, Shujuan Liu1   

  1. 1 School of Chemistry and Materials Science, East China University of Technology, Nanchang 330013, China;
    2 School of Modern Agriculture, Xiangxi Vocational and Technical College for Nationalities, Xiangxi 416099, Hunan Province, China;
    3 School of Information Engineering, East China University of Technology, Nanchang 330013, China
  • Received:2025-06-03 Accepted:2025-10-20 Published:2025-12-30
  • Contact: Yueming Zhou E-mail:ymzhou@ecut.edu.cn

PDF

129

赞

摘要： 微色谱柱离子交换色谱法对学生操作规范化与熟练度要求较高。本文探索用机器视觉捕捉实验者柱上操作过程中时序行为数据的方法，并构建详尽的实验变量数据库。结合实验直接获取的数据，运用机器学习挖掘数据背后的规律，并基于数字孪生技术实时监测操作过程，以识别误差来源和影响结果稳定性的关键因素。设计成教学实验，培养学生应用数字化技术解决分析化学实验问题的能力。

关键词: 机器学习, 微色谱柱, 痕量砷, 机器视觉

Abstract: Microcolumn ion exchange chromatography demands rigorous standardization and operational proficiency from students. This study investigates a machine vision-based approach to capture time-series behavioral data during on-column operations, establishing a comprehensive experimental variable database. By integrating experimentally acquired data with machine learning algorithms, we elucidate underlying patterns and employ digital twin technology for real-time process monitoring. This methodology enables identification of error sources and critical factors influencing result stability. Implemented as an instructional experiment, this approach cultivates students’ ability to apply digital technologies to address analytical chemistry challenges.

Key words: Machine learning, Microcolumn chromatography, Trace arsenic, Machine vision