Abstract: Traditional lightning overvoltage experiments and electromagnetic transient simulations are constrained by drawbacks such as high equipment costs, significant safety risks, and complex modeling procedures, leading to insufficient practical instruction in the field of power system overvoltage. Given the formidable capabilities of generative artificial intelligence (GAI) in code generation, this study explored the application of GAI technology in virtual simulation experiments for power system overvoltages. By generating hypertext markup language (HTML) code, a comprehensive virtual simulation experimental framework was established, covering the development process of lightning, induced overvoltage simulation, overvoltage calculation for lightning strikes on transmission lines, traveling wave overvoltage calculation in substations, and the operating principles of arresters. These virtual simulation experiments visualize abstract theoretical concepts such as traveling wave propagation, reflection, and refraction, enabling students to gain a deep understanding of the underlying principles through interactive operation, thereby enhancing teaching effectiveness. This study provides valuable references and insights for the application of GAI technology in virtual simulation experiments within the electrical engineering major.