Aerodynamic enhancement of diverterless supersonic inlets via surface sensitivity-based bump optimization

  • Chuang Ma
  • , Daochun Li*
  • , Jiangtao Huang*
  • , Bowen Shu
  • , Chengjun He
  • , Jun Deng
  • , Gang Liu
  • *Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

The Diverterless Supersonic Inlet (DSI) is widely used in modern advanced fighters. As a key component of the DSI, the bump plays a decisive role in determining the aerodynamic efficiency and flow field quality of the inlet. Given that the aerodynamic performance of a DSI is highly sensitive to the shape of its bump, this paper develops a novel method for bump adjoint optimization based on surface sensitivity for gradient calculation. The conventional residual differencing step in traditional adjoint methods is eliminated. This advancement ensures that the efficiency of detailed bump design is entirely independent of the number of design variables. This method was applied to optimize the bump of a research-oriented DSI. The optimization enhanced the stagnation pressure recovery coefficient and lowered the circumferential distortion index. Furthermore, optimizing the bump shape not only improves the compression wave structure but also adjusts the internal duct flow characteristics, thereby reducing stagnation pressure losses caused by viscous dissipation. The optimized bump geometry, with increased height and an upstream peak shift, enhances the spanwise pressure gradient, influencing boundary layer development and improving low-energy gas extraction. Additionally, the study provides insights into the Inlet Capture Curve (ICC) design principles, contributing to the bump shape generation method based on the osculating wave-ride theory.

Original languageEnglish
Article number086135
JournalPhysics of Fluids
Volume37
Issue number8
DOIs
StatePublished - 1 Aug 2025

Fingerprint

Dive into the research topics of 'Aerodynamic enhancement of diverterless supersonic inlets via surface sensitivity-based bump optimization'. Together they form a unique fingerprint.

Cite this