A Knowledge-Augmented Assisted Authoring Method for Power Energy Big Data Retrieval in Report Content

To address challenges in power report authoring, such as low data retrieval efficiency, weak knowledge association, and unstable content generation quality, this paper proposes a knowledge-augmented assisted authoring method for power energy big data retrieval in report content by leveraging the text data processing and knowledge reasoning capabilities of large language models (LLMs). By adopting a “small model reasoning and thinking combined with large model retrieval and generation” approach, complex authoring tasks are decomposed into simpler subtasks. The knowledge retrieval and data processing capabilities of LLMs are utilized to assist in generating report content. Experimental results demonstrate that this method effectively achieves complex task decomposition and retrieval-augmented generation (RAG), producing content that aligns more closely with expert preferences, thereby effectively supporting the auxiliary authoring of power report content.