Abstract: Under the global trend of digital transformation in manufacturing, the mechanical engineering education is confronted with systemic challenges, including an increasingly complex knowledge system, outdated practical training methodologies, and a misalignment between curricular content and technological advancements. This study has examined the prevailing pain points in mechanical engineering pedagogy and then proposed an“AI-mechanical education integration framework”to address these issues through three synergistic approaches. They are dynamic knowledge graphs coupled with adaptive learning mechanisms, digital twin-enabled virtual-physical hybrid training environments, and industrial knowledge distillation for autonomous curriculum evolution. The efficacy of the framework has been preliminarily validated through implementation cases such as the ANSYS knowledge graph system. Projected to mature within five years, this initiative aims to establish an intelligent educational ecosystem that deep integration in industry-academia collaboration, thereby facilitating a paradigm shift in mechanical engineering education from traditional knowledge transmission to competency cultivation. This transformative approach presents viable solution for developing advanced engineering talent capable of meeting the demands of smart manufacturing.

Key words: Artificial intelligence technology, Mechanical engineering courses, Teaching methods