Abstract: In the context of meteorological operations encompassing weather consultation meetings, support for significant events, and routine service delivery, the increasing complexity of operational systems and functional modules has posed considerable challenges for non-specialist users. These challenges are reflected in difficulties associated with locating specific system functions, navigating complex operational workflows, and interpreting specialized meteorological outputs. To address these practical challenges, an innovative integration of DeepSeek large language model is proposed, establishing a comprehensive framework that incorporates meteorological element analysis, customized graphical product generation, and automated consultation material production. This initiative has led to the development of an intelligent operational support system for meteorological applications based on natural language interaction, which substantially diminishes the existing barriers between users and intricate professional systems. Capitalizing on the formidable capabilities of DeepSeek in intelligent responsiveness and natural language processing, the implemented system facilitates multi-modal human-computer interaction through both vocal and textual input channels. Furthermore, through a systematic interface-oriented transformation of existing meteorological information tools, seamless and efficient coupling with DeepSeek architecture has been successfully accomplished. To address unique demands of professional meteorological visualization, spatial analysis technology has been incorporated to autonomously generate spatial orientation annotations, markedly improving the interpretability and user comprehension of meteorological graphics. The system synthesizes an extensive array of meteorological and geographical data, including 34 administrative divisions and multi-scale geographical units, 89 distinct categories of meteorological products, and a restructured set of 14 functional module interfaces. Empirical validation through rigorous testing across 3 representative operational scenarios, daily forecasting, major event support, and specialized consulting, has demonstrated substantial improvements in operational efficiency and overall user experience. This investigation not only presents a functional and scalable intelligent solution for meteorological services but also provides a transferable innovative framework and practical implementation pathway, offering valuable insights for future intelligentization efforts in domain-specific operational systems.