A method for suppressing unsteady pressure pulsation in an air intake based on deep reinforcement learning
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-05-28
- Publication Date
- 2026-06-26
AI Technical Summary
Traditional control methods are difficult to effectively cope with nonlinear and unsteady pressure pulsations during supersonic inlet surge, resulting in poor control performance, lag response, and inability to adapt to changing operating conditions, which affects the safety and performance of the aircraft.
By employing a deep reinforcement learning-based approach, and combining the TD3 reinforcement learning model with a high-fidelity CFD simulation environment, a composite adaptive reward function and physical constraint system are constructed to achieve intelligent pressure pulsation suppression.
It achieves efficient and robust pressure pulsation suppression under a wide range of operating conditions, improves the adaptability and response speed of the control, and ensures the safety and engineering feasibility of the control process.
Smart Images

Figure CN122280707A_ABST