To address the low utilization rate and inaccurate flow prediction of natural gas pressure energy during transmission, as well as the coordinated optimal dispatch problem of battery swapping in centralized charging stations, this paper proposes a distributionally robust optimal dispatch strategy for pressure regulation station microgrids considering dual source-load uncertainties. First, a distributionally robust collaborative optimal dispatch model is developed for a microgrid that integrates pressure differential power generation, a centralized charging station, and the main grid. The linear weighting method is then employed to comprehensively optimize the two objectives: minimizing the microgrid's operating cost and minimizing tie-line power fluctuation. Second, a multi-discrete-scenario probability ambiguity set based on comprehensive norm constraints is introduced to accurately characterize the uncertainty boundaries of pressure energy generation, photovoltaic (PV) output, and load. A two-stage min + maxmin distributionally robust optimization (DRO) model, featuring day-ahead decision-making and intra-day worst-case scenario correction, is formulated and solved using the column-and-constraint generation (C&CG) algorithm. Finally, a case study is conducted based on 16 months of actual operating data from the Qiulin gas gathering station. The results demonstrate that, compared with traditional robust optimization and stochastic optimization, the proposed method significantly enhances the full accommodation level of pressure energy and the overall economic benefits of the microgrid while ensuring high system reliability.