面向新工科的矿物加工工程专业核心课程教学方法研究——以“矿物加工过程检测与控制”课程为例

Research on Teaching Methods for Core Courses in Mineral Processing Engineering under the New Engineering Education Paradigm—A Case Study of the Course “Detection and Control of Mineral Processing Processes”

摘要: 新工科建设背景下，矿物加工工程专业人才培养面临知识体系更新加快、学科交叉融合深化以及人工智能技术快速发展的新要求。作为矿物加工工程专业核心课程，“矿物加工过程检测与控制”承担着培养学生过程检测、自动控制与智能化应用能力的重要任务。然而，课程存在知识体系复杂、理论抽象、实践性强以及学生自主学习能力不足等问题。与此同时，大语言模型技术的快速发展为工程教育教学改革提供了新的技术路径，但其在专业课程学习中的应用仍面临知识准确性不足、专业适配性较差等挑战。针对上述问题，本文以新工科人才培养需求为导向，提出一种基于检索增强生成(Retrieval-Augmented Generation, RAG)的课程辅助学习教学方法。通过构建课程专属知识库，建立“课程知识体系–大模型智能问答–自主学习指导”融合机制，实现课前预习、课中学习和课后巩固全过程支持。教学实践表明，该方法有效提高了学生自主学习能力、工程问题分析能力以及课程学习成效，促进了人工智能技术与专业课程教学的深度融合，为新工科背景下矿物加工工程专业课程教学改革提供了新的思路。

Abstract: Under the background of New Engineering Education, talent cultivation in Mineral Processing Engineering is facing new requirements arising from the rapid evolution of knowledge systems, the deepening integration of interdisciplinary fields, and the fast development of artificial intelligence technologies. As a core course in the Mineral Processing Engineering curriculum, Detection and Control of Mineral Processing Processes plays an important role in developing students’ capabilities in process monitoring, automatic control, and intelligent applications. However, the course is characterized by a complex knowledge system, abstract theoretical concepts, strong practical orientation, and insufficient student self-directed learning ability. Meanwhile, the rapid advancement of large language models (LLMs) has provided new technological pathways for engineering education reform, yet their application in specialized courses still faces challenges such as insufficient knowledge accuracy and limited domain adaptability. To address these issues, this study proposes a course-assisted learning approach based on Retrieval-Augmented Generation (RAG), guided by the requirements of New Engineering Education talent cultivation. By constructing a course-specific knowledge base and establishing an integrated mechanism that combines the course knowledge system, LLM-based intelligent question answering, and self-directed learning guidance, the proposed approach provides comprehensive support for pre-class preparation, in-class learning, and post-class reinforcement. Teaching practice demonstrates that the method effectively enhances students’ self-directed learning ability, engineering problem-solving capability, and overall learning outcomes. It also promotes the deep integration of artificial intelligence technologies with professional course instruction, providing a new perspective for teaching reform in Mineral Processing Engineering under the New Engineering Education framework.