Research on Process Quality Prediction and Optimization Techniques for Solid Fuel Rocket Engine Integrated with Large Language Models

Solid fuel rocket engines serve as a crucial technological foundation for rapid space launch capabilities and are a significant indicator of national strategic weapon delivery capabilities. With the continuous advancement of industrial manufacturing systems, particularly the explosive growth of generative artificial intelligence technologies in recent years, the ability to collect and analyze vast amounts of industrial big data has been greatly enhanced. Consequently, exploring the deep correlations between production parameters and product quality has become a key aspect of the manufacturing process for solid fuel rocket engines. This paper, set against the backdrop of solid fuel rocket engine manufacturing, leverages machine learning, large model technologies, and other AI algorithms to study the relationship between quality data and quality issues in the production process. By identifying potential quality problems in advance, the research aims to provide timely recommendations for process parameter optimization, thereby reducing material waste, rework, and scrap. This approach holds significant practical value in reducing component quality issues and enhancing the controllability of the production process.