Abstract: To address the problem that the rugged terrain and lighting conditions in the lunar polar region impose severe constraints on the deployment of laser facilities, this paper proposes an optimal deployment design method for a Lunar Laser Power Transmission Station（LPTS）based on a Bayesian algorithm. The proposed method constructs a weighted objective function that incorporates both effective laser power coverage and regional connectivity. It leverages the Bayesian optimization framework to efficiently explore the complex parameter space within a limited number of evaluations, thereby enabling the rapid identification of optimal deployment configurations for Laser Emitter Unit（LEU）. Research shows that when applied to a scenario using actual topographical data of the Shackleton Crater acquired by the Lunar Orbiter Laser Altimeter（LOLA）, the Bayesian-optimized deployment scheme demonstrates a significant enhancement in network coverage performance compared to the initial baseline. Specifically, the total effective coverage increases from 17.80% to 24.08%, while the maximum regional connectivity soars from 38.69% to 99.29%. This study provides an effective solution to the challenge of precisely deploying lunar surface infrastructure and offers a new technical pathway for the fine-grained planning of future lunar missions.