AI赋能地质工程专业教学——以《岩土钻掘工程学》课程为例

AI-Empowered Teaching in Geological Engineering—Taking the Course “Geotechnical Drilling and Excavation Engineering” as an Example

摘要: 《岩土钻掘工程学》是地质工程专业的核心主干课程，具有学科交叉性强、工程实践性突出、实验依赖性高的显著特点。当前，该课程面临教学内容滞后于行业前沿发展、学生学科基础差异显著制约教学深度、实践教学受限于设备场地与安全条件、教学评价方式单一等结构性困境。本文系统梳理了AI赋能地质工程教学改革的既有成果，在精准诊断课程痛点的基础上，提出“知识图谱重构课程内容–AI助教支撑教学模式转型–虚拟仿真突破实践瓶颈–AI驱动多元评价机制”的四维协同改革框架。围绕这一框架，文章阐述了知识图谱以“主题–子主题–知识点–属性”四层结构重塑教学内容体系并融入大陆科学钻探、大洋钻探、极地钻探等前沿应用领域案例的实现路径，分析了AI智能体贯通“课前–课中–课后”全流程个性化学习支持的技术原理，论证了虚拟仿真技术构建“认知–操作–设计”三层次实践教学平台的能力培养价值，并探讨了AI驱动的全过程数据采集、多维度能力画像与即时性反馈预警对教学评价革新的推动作用。同时，AI赋能需警惕技术依赖与算法偏见等潜在风险，应坚持“以学生发展为中心”的价值导向。本文提出的改革方案可为地质工程类课程的数智化转型提供可借鉴的范式参考。

Abstract: Geotechnical Drilling and Excavation Engineering is a core backbone course of the Geological Engineering major, characterized by strong interdisciplinary nature, prominent engineering practicality, and high dependence on experimental teaching. Currently, the course faces structural challenges including curriculum content lagging behind industry frontiers, significant disparities in students’ disciplinary foundations restricting teaching depth, practical teaching constrained by equipment, site and safety conditions, and a unitary teaching evaluation method. This paper systematically reviews the existing research on AI-empowered teaching reform in geological engineering, and based on a precise diagnosis of the course’s pain points, proposes a four-dimensional collaborative reform framework of “knowledge graph reconstructing curriculum content, AI teaching assistant supporting teaching mode transformation, virtual simulation breaking through practical bottlenecks, and AI-driven multi-dimensional evaluation mechanism”. Around this framework, the paper elaborates on the implementation path of knowledge graphs restructuring the teaching content system through a four-layer “topic - subtopic - knowledge point - attribute” structure while integrating cases from frontier application domains such as continental scientific drilling, ocean drilling, and polar drilling; analyzes the technical principles of AI agents providing personalized learning support throughout the “pre-class - in-class - post-class” entire process; demonstrates the competency development value of virtual simulation technology in constructing a three-level “cognition - operation - design” practical teaching platform; and explores the driving role of AI-enabled whole-process data collection, multi-dimensional competency profiling, and real-time feedback and early warning in reforming teaching evaluation. Meanwhile, AI empowerment requires vigilance against potential risks such as technological dependence and algorithmic bias, and should adhere to the value orientation of “student-development-centered”. The reform scheme proposed in this paper can provide a reference paradigm for the digital and intelligent transformation of geological engineering courses.