Mathematical Physics Equations is a fundamental core course for engineering disciplines such as geophysics, with broad applications in seismic wave propagation modeling, electromagnetic field analysis, and fluid dynamics. However, traditional teaching approaches primarily focus on theoretical derivations, with limited emphasis on practical applications. This results in students having insufficient skills in numerical computation and programming, thereby restricting their ability to solve complex engineering problems. With the rapid advancement of artificial intelligence (AI), technologies such as automated programming, intelligent derivation, and visualization have demonstrated immense potential in scientific computing, offering new perspectives and tools for the reform of Mathematical Physics Equations teaching. This paper proposes an AI-driven intelligent teaching reform framework that integrates automated programming demonstrations, problem-driven teaching, and industry-oriented case studies to enhance students’ computational thinking and engineering practice skills. The research findings indicate that this approach not only significantly improves students’ understanding and application of mathematical physics equations but also strengthens their overall innovative capabilities, providing a viable pathway for the modernization of teaching in this domain.