Optimized design and experimental study of an ultra-high-pressure air linear flow regulation valve

Abstract

To address the poor linearity between valve core displacement and valve mass flow in conventional air flow control valves, which limits flow regulation accuracy, this study proposes an ultra-high-pressure air linear flow regulation valve (UHPALFRV) designed using a dual-cubic contraction profile, a Laval nozzle, and a quadratic envelope. Multiple numerical models of Laval-nozzle valve seats were established, and their flow characteristics were evaluated via pressure distribution, turbulent kinetic energy (TKE), and Mach number to identify the valve seat geometry most conducive to gas flow. A mathematical model linking valve core displacement to effective flow area was formulated for the novel linear regulation unit, and key parameters were optimized using an improved particle swarm optimization algorithm integrating chaotic mapping, golden sine perturbation, Cauchy dynamic perturbation, and an elite selection strategy, thereby achieving highly linear flow modulation. A high-pressure air valve test platform was constructed, and flow linearity tests were performed. The results indicate that, compared with the other three configurations, the valve seat employing a dual-cubic contraction profile exhibits a more uniform pressure gradient and higher consistency. The mean absolute error (MAE) and root-mean-square error (RMSE) of the pressure distribution on the symmetry plane are 114238.6 and 297583.9, respectively—the lowest among the four groups—and the domain-averaged turbulent kinetic energy (TKE) decreases by approximately 12.1 %. Building on this, the optimized spool–seat combination achieves a maximum TKE reduction of about 38.8 %, while the peak Mach number at the throat increases by roughly 4.26 %. The maximum relative discrepancy between experiments and CFD is around 4.43 %. After optimization, the linearity of the displacement–mass flow rate relationship reaches 99.734 %, with a sum of squared errors (SSE) of 0.00292. These findings substantiate the effectiveness and high-precision characteristics of the proposed structure and optimization methodology.