Research on Quality Control Method of Hydrogen Nozzle Based on Machine Learning

With the rapid development of China's aerospace industry, the market demand for launch vehicles has greatly increased. The hydrogen nozzle is the core component of the thrust chamber injector of the hydrogen-oxygen engine. The flow performance of the injector directly affects the combustion efficiency, stability, rocket thrust and service life of the engine. Based on machine learning technology and hydrogen nozzle flow test data, this paper completed the research on the application and methods of hydrogen nozzle quality control. On the basis of the original data, suggestions for using machine learning technology for quality control are put forward from three aspects: feature engineering, model construction and model evaluation.