收藏切换
Construction and application of intelligent question-answering model for accident investigation reports based on DeepSeek and RAG
收藏切换
PDF
Hua LI1, Lizhou WU1, 2, Xinhong LI**, 1, Yue ZHANG1, Yao FENG1, Ziyun QIN1
China Safety Science Journal | 2026, 36(1) : 26 - 34
Less
收藏切换
China Safety Science Journal | 2026, 36(1): 26-34
Safety Science Theories and Methods
Construction and application of intelligent question-answering model for accident investigation reports based on DeepSeek and RAG
Full
Hua LI1, Lizhou WU1, 2, Xinhong LI**, 1, Yue ZHANG1, Yao FENG1, Ziyun QIN1
Affiliations
  • 1School of Resources Engineering, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China
  • 2College of Safety Science and Engineering, Xi'an University of Science and Technology, Xi'an Shaanxi 710054, China
Published: 2026-01-28 doi: 10.16265/j.cnki.issn1003-3033.2026.01.0840
Outline
收藏切换

In order to address the constraints of limited corpus resources, restricted input capacity, and data privacy in applying LLMs to the field of safety engineering, a localized accident question-answering model integrating the DeepSeek with a RAG mechanism was constructed to enable intelligent parsing and knowledge services for complex texts, thereby supporting safety management decision-making. A semantic-feature corpus was built based on accident investigation reports and laws and regulations released by government emergency management systems, and technologies such as PaddleOCR, LayoutLMv3, and YOLOv8 were incorporated to accomplish document structure reconstruction and semantic modeling. The model encompassed four stages—document parsing, semantic alignment, knowledge-base construction, and hybrid retrieval—and was designed with capabilities for causal-chain extraction, regulation matching, and semantic mapping. The results indicated that, compared with the Deepseek-r1:32b model without the RAG mechanism, the enhanced model achieved improvements of 7.7% in automated scoring and 17.6% in human evaluation, and the response-speed and stability metrics presented higher numerical performance than those of the baseline model. The model performance was still influenced by the local parameter scale and the knowledge-updating mechanism, yet the experimental findings demonstrate that it is capable of fulfilling the intended functions in the present study.

DeepSeek  /  large language model (LLM)  /  retrieval augmented generation  /  accident investigation report  /  knowledge base
Hua LI, Lizhou WU, Xinhong LI, Yue ZHANG, Yao FENG, Ziyun QIN. Construction and application of intelligent question-answering model for accident investigation reports based on DeepSeek and RAG[J]. China Safety Science Journal, 2026 , 36 (1) : 26 -34 . DOI: 10.16265/j.cnki.issn1003-3033.2026.01.0840
Year 2026 volume 36 Issue 1
PDF
77
6
Cite this Article
BibTeX
Article Info
doi: 10.16265/j.cnki.issn1003-3033.2026.01.0840
  • Receive Date:2025-09-14
  • Online Date:2026-07-08
  • Published:2026-01-28
Article Data
Affiliations
History
  • Received:2025-09-14
  • Revised:2025-11-21
Funding
Affiliations
    1School of Resources Engineering, Xi'an University of Architecture and Technology, Xi'an Shaanxi 710055, China
    2College of Safety Science and Engineering, Xi'an University of Science and Technology, Xi'an Shaanxi 710054, China
References
Share
https://castjournals.cast.org.cn/joweb/zgaqkxxb/EN/10.16265/j.cnki.issn1003-3033.2026.01.0840
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数

Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)

Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT