Abstract: In this study, the underground cave group lighting of large hydropower stations is targeted, and a comprehensive lighting management system has been developed for its management needs in a special operating environment. The system fully considers the key functions of the cavity group in terms of power transmission, transportation, emergency escape, ventilation gas exchange, and production of production of life. At the same time, the challenges brought by the complexity of the tunnel layout. The stratified partition control strategy is adopted, and intelligent prediction and strengthening learning technology are integrated, and the dynamic adjustment of the lighting layout and realizing brightness is optimized. Through the interconnection of multiple platforms and the data control of the cloud platform, the system not only ensures the effective carrier of the internal work of the cave, but also pays attention to the reasonable allocation of resources and the savings of energy. Effectively solve the problems of large-scale cave group lighting systems in management control and energy consumption. The implementation of the system provides a new perspective for the efficient and intelligent management of the water power station hole group management, and has contributed to improving the stability and security of the power station.