大学化学

上一篇    下一篇

人工智能赋能下的物理化学实验混合式教学实践

沈海云, 邵松雪, 邱丽娟, 于曦, 杜静, 朱莉娜   

  1. 天津大学理学院化学系, 化学化工国家级实验教学示范中心, 天津 300354
  • 收稿日期:2025-12-26 修回日期:2026-02-26
  • 通讯作者: 朱莉娜 E-mail:linazhu@tju.edu.cn Lina Zhu
  • 基金资助:
    2025年天津大学理学院人工智能赋能课程建设项目；天津市普通高等学校本科教学质量与教学改革研究计划(2025)重点项目A251005608；天津市普通高等学校本科教学质量与教学改革研究计划项目B251005611

Hybrid instruction of physical chemistry experiments enabled by artificial intelligence

Haiyun Shen, Songxue Shao, Lijuan Qiu, Xi Yu, Jing Du, Lina Zhu   

  1. National Demonstration Center for Experimental Chemistry and Chemical Engineering Education, Department of Chemistry, College of Science, Tianjin University, Tianjin 300354, China
  • Received:2025-12-26 Revised:2026-02-26
  • Contact: Lina Zhu E-mail:linazhu@tju.edu.cn

PDF

114

赞

摘要： 在教育数字化转型与“新工科”建设的双重驱动下，人工智能技术(AI)与高等教育的融合已成为教学改革的重要方向。本文以物理化学实验课程为例，探索人工智能赋能实验教学的实践路径。具体措施包括：构建中国大学MOOC平台智慧课程，利用人工智能驱动创新实验设计与仪器升级，借助AI技术辅助课程思政建设，以及利用大模型构建多维度综合评价体系。通过搭建知识图谱、部署AI助教、研发实验报告智能评分系统，实现教学资源精准推送、实验过程智能优化、课程思政个性化引导与评价反馈高效化。依托智慧课程资源与AI技术优势，可将智慧教学模式重点融入课前、课后这两个教学相对薄弱的环节，并与课中教学环节联动，形成完整的教学闭环。实践表明，这些措施有效提升了教学效率与评价质量，显著增强了学生的自主探究能力与创新思维，为理工科实验课程的智能化改革提供了可借鉴的实践路径。

关键词: 物理化学实验, 人工智能, 智慧课程, 知识图谱, AI助教, 实验报告智能评分系统

Abstract: Under the dual impetus of educational digital transformation and the “Emerging Engineering Education” initiative, the integration of artificial intelligence (AI) technology with higher education has emerged as a pivotal direction for pedagogical reform. This study examines the Physical Chemistry Experiment course as a case study to explore AI-enhanced experimental teaching methodologies. Key initiatives include: (1) developing smart courses on the Chinese MOOC platform, (2) employing AI to drive innovative experimental design and equipment modernization, (3) utilizing AI-assisted ideological education in curriculum development, and (4) constructing a multidimensional comprehensive evaluation system through large language models. The implementation involves creating knowledge graphs, deploying AI teaching assistants, and developing an intelligent experimental report grading system, enabling precise resource allocation, intelligent optimization of experimental processes, personalized ideological education, and efficient evaluation feedback. By leveraging smart course resources and AI technologies, this teaching model effectively bridges pre-class and post-class learning—traditionally weaker components—with inclass instruction, forming a complete pedagogical cycle. Practical results demonstrate significant improvements in teaching efficiency and assessment quality, along with enhanced student autonomy in inquiry-based learning and innovative thinking. This approach provides a replicable model for intelligent reform in STEM experimental courses.

Key words: Physical chemistry experiments, Artificial intelligence, Smart courses, Knowledge graphs, AI teaching assistants, Intelligent experimental report scoring system