The large-scale integration of wind and solar energy has intensified spatiotemporal mismatches among source-side generation, grid-side transmission, load-side demand and storage-side flexibility. To address the separation between energy-system constraints and geographic process analysis, this study proposes a geo-intelligent modeling framework for wind and solar energy planning. The framework integrates resource sensing, potential assessment, power-system dispatch, source-grid-load-storage coordination, and spatial layout optimization, and is demonstrated through a case study of rooftop photovoltaics in China. The results show a spatial mismatch between rooftop-rich regions and areas with high photovoltaic capacity factors. Under a scenario of 90% grid flexibility and a storage capacity equivalent to 8 hours of average load, grid-connected rooftop photovoltaic generation could achieve a carbon reduction of approximately 4 471.2 MtCO₂. The optimization results reveal that rooftop photovoltaic development requires trade-offs among penetration improvement, curtailment control, storage allocation and interregional power transmission, while spatial deployment should optimize county-level combinations to maintain regional supply-demand balance. By integrating geospatial intelligence with source-grid-load-storage coordination, the proposed framework can support the transition of wind and solar potential assessment from static resource evaluation toward spatial decisions that are grid-integrable, carbon-mitigating and implementable.