Engineering material failure analysis report generation based on QWen and Llama2

Abstract

Engineering material failure carries risks to the public and personal safety and causes economic loss. Failure analysis report describes the causes of failure and gives the improvement strategy, but requiring a lot of manpower of experts. The powerful text understanding and generation capabilities of large language models make automatic analysis possible. This paper presents an automatic report generation method using large language models, including training on only a few samples and instruction fine-tuning method (prompt engineering). The experiments are executed on two popular large language model bases: QWen and Llama2, involving 127 failure reports in 9 fields such as petroleum and petrochemical, nuclear and thermal power, ocean engineering, energy engineering, infrastructure, aerospace, transportation, equipment manufacturing, and water conservancy. Using different questioning ways, GPT evaluation and analysis based on manual verification are performed on the generated results. The experimental results show that the model can generate valuable case reports after being trained with specific domain data. The introduction of artificial intelligence technology for engineering failure prediction can enhance the safety of engineering development, improve industrial production efficiency, reduce labor costs, provide decision support during data analysis, and foster the emergence of new markets and business models.

Keywords

• Failure analysis report generation
• Large language model
• Prompt engineering
• Llama2
• Qwen
• Gpt evaluation