主办：陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊：1976年
汽车实用技术

汽车实用技术 ›› 2025, Vol. 50 ›› Issue (17): 93-99,114.DOI: 10.16638/j.cnki.1671-7988.2025.017.018

• 汽车教育 • 上一篇    

“KG+AI”双引擎驱动的智慧课程改革探索 ——以机械工程控制基础课程为例

师路欢，杨冲，马军磊   

  1. 许昌学院 机电工程学院
  • 发布日期:2025-09-03
  • 通讯作者: 师路欢
  • 作者简介:师路欢（1984－），女，硕士，副教授，研究方向为电机建模仿真与电机驱动控制
  • 基金资助:
    2024 河南省高等教育教学改革研究与实践项目（2024SJGLX0455）；2025 河南省软科学研究项目（2524004 10299）；许昌学院 2024 年度教育教学改革与实践项目（XCU2024-YB-52，XCU2024-PY-06）

Exploration of Intelligent Curriculum Reform Driven by "KG+AI" Dual Engines －Taking the Curriculum of Mechanical Engineering Control Basic as an Example

SHI Luhuan, YANG Chong, MA Junlei   

  1. School of Mechanical and Electrical Engineering, Xuchang University
  • Published:2025-09-03
  • Contact: SHI Luhuan

PDF

摘要： 为贯彻“新工科”建设理念，适应工程教育专业认证的新要求，文章以机械工程控制 基础课程为研究对象，探索在人工智能（AI）、数字资源、知识图谱（KG）等新兴技术赋能 下的智慧课程建设路径。围绕“图谱驱动、AI 赋能、精准教学、持续优化”理念，构建了包 括 KG、智能助学、分层教学和智能反馈在内的智慧教学体系，采用“图谱驱动－AI 赋能－ 闭环实施－数据验证”四阶段方法论推进课程改革。图谱构建包括课程知识点的梳理分类、 关键节点间的语义关系建模、图谱可视化部署与教学资源挂接，实现了知识结构与学习任务 之间的动态联动；AI 技术则应用于学生学习行为采集分析、认知路径跟踪、个性化资源推荐 与错因诊断，为教师提供数据驱动的学情判断与教学策略支持。教学实证研究覆盖 2023－2024 学年共 240 余名机械类专业学生，结果显示，课程满意度达 94%，课堂互动频次提升 35%， 仿真实验完成率达 92%，学生在控制系统建模与项目实战中的能力显著增强。教学实践表明， 该课程改革有效提升了学生工程应用能力与自主学习能力，促进了人才培养质量的整体跃升， 形成了可推广、可复制的智慧课程建设经验，具备良好的推广应用价值。

关键词: AI；KG；智慧课程；智能闭环；能力导向

Abstract: To implement the vision of "New Engineering" construction and meet the new requirements of engineering education professional accreditation, this paper takes the curriculum of Mechanical Engineering Control Basic as the research object and explores a intelligent curriculum development approach empowered by emerging technologies such as artificial intelligence (AI), digital resources, and knowledge graphs (KG). Guided by the vision of "graph-driven, AI-powered, precision teaching, and continuous optimization", a intelligent instructional system that integrates KG, intelligent learning support, differentiated instruction, and adaptive feedback is establishes. The curriculum reform follows a four-stage methodology: graph driven－AI integration－closedloop implementation－data-driven validation. The graph development includes the tease classification of curriculum content, modeling of semantic relations between key nodes, hooking up graph visual deployment and instructional resource, achieving dynamic linkage between knowledge structures and learning tasks. AI technologies are applied to analyze student learning behavior collection, track cognitive pathways, personalized resource recommendations, and identify misconceptions, provide data-driven learning situation judgment and teaching strategy support for teachers. A teaching experiment is conducted during the 2023－2024 academic year involving more than 240 students majoring in mechanical. Results show a 94% curriculum satisfaction rate, a 35% increase in frequency of classroom interaction, and a 92% completion rate of simulation experiments. Students demonstrate significantly enhanced competencies in control system modeling and project practice. Teaching practice shows that this curriculum reform has effectively enhanced students' engineering application ability and autonomous learning ability, promoted an overall leap in the quality of talent cultivation, and formed replicable and scalable experience in the construction of intelligent curriculum, which has good value for promotion and application.

Key words: AI; KG; intelligent curriculum; intelligent closed-loop; ability oriented