Scholarlink

This paper proposes a novel approach to standardizing Daily Drilling Reports (DDRs) within an oil and gas company. Instead of merely providing example descriptions to prompt Large Language Models, the method enhances model performance by integrating Knowledge Graphs to generate DDR templates. By incorporating structured knowledge, the model is anchored in factual information. The results demonstrate this strategy produces accurate and consistent textual patterns, contributing to improved knowledge management and facilitating the practical application of standardized DDR writing.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

LLM-Based Template Generation with Knowledge Graphs for Daily Drilling Reports

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Luis H. Morelli,
  • Giovani Candido,
  • Jonas Queiroz,
  • Marcelo A. Jaculli,
  • Nelson Jr. Choueri,
  • Arnaldo Cândido Jr.,
  • Bruno E. Penteado,
  • Ivan R. Guilherme,
  • João P. Papa,
  • Stephan R. Perrout,
  • Felipe L. Oliveira

摘要

This paper proposes a novel approach to standardizing Daily Drilling Reports (DDRs) within an oil and gas company. Instead of merely providing example descriptions to prompt Large Language Models, the method enhances model performance by integrating Knowledge Graphs to generate DDR templates. By incorporating structured knowledge, the model is anchored in factual information. The results demonstrate this strategy produces accurate and consistent textual patterns, contributing to improved knowledge management and facilitating the practical application of standardized DDR writing.