Regression analysis and parameters optimization of output characteristics for aviation pressure reducing regulator

To study the impact of design parameters and interdependencies on an aviation dual-stage gas pressure reducing regulator’s output characteristics, the AMESim model was established and verified by comparing results with the test. A quadratic regression model of the output characteristics and design parameters was obtained by the response surface method to analyze design parameters’ influence. The quadratic regression model was optimized by using the non-dominated sorting genetic algorithm-Ⅱ (NSGA-Ⅱ) to obtain the optimal design parameters. The response surface results showed that the feedback hole area was the primary factor affecting overshoot, whose interaction with the spool seal’s dynamic leakage area had a significant impact on overshoot. The spool mass was the primary factor affecting output pressure pulsation intensity, whose interaction with the spring stiffness and diaphragm stiffness had a significant influence on pulsation intensity. The optimization indicated that the output characteristics would be optimal when the spool mass, feedback hole area, mainspring stiffness, auxiliary spring stiffness, spool seal’s dynamic leakage area and diaphragm stiffness in Ⅱ-stage structure were 52.26 g, 9.06 mm2, 67.27 N/mm, 10.68 N/mm, 0.64 mm2 and 89.49 N/mm, respectively. The output pressure overshoot was reduced by 28.72%, and the pulsation intensity reduced by 40.63%.