The evolution of smart grid ecosystems has ushered in vast amounts of heterogeneous data emanating from diverse sources and platforms. While these data sets harbor invaluable insights for optimizing energy efficiency, resource allocation, and operational predictability, their effective integration remains a formidable challenge due to inconsistencies in data formats, semantics, and structures. This research introduces a Unified Semantic Modeling (USM) framework aimed at achieving seamless cross-platform data integration in smart grid ecosystems. By harnessing ontological representations and machine learning-driven semantic mappings, the proposed USM framework elucidates inherent relationships within multi-sourced data and fosters unified comprehension. Rigorous evaluations on varied smart grid datasets affirm the efficacy of the USM framework in enhancing data interoperability, reducing integration latency, and augmenting the accuracy of cross-platform data analytics. The findings underline the pivotal role of semantic modeling in navigating the complexities of contemporary smart grid data landscapes and set a benchmark for subsequent integration methodologies.