The invention provides an automatic stall wing control method based on deep
reinforcement learning, relates to the field of flow
state control, and solves the problems that a static stall control method cannot cope with unsteady
airflow changes, and an existing closed-
loop control algorithm is limited in response in a complex and high-dimensional flow environment and lacks real-time intelligent adjustment capability. The
system is suitable for aircrafts and the like, the wing flow field state and global parameters are collected in real time through the sensor, control actions are output through the DQN intelligent body, the
dielectric barrier
discharge plasma actuator is driven, and wing stall
delay, lift force lifting and pneumatic vibration suppression are achieved. The problem of dynamic stall control failure is solved, limitation of a traditional method is overcome, full-working-condition intelligent self-
adaptive control is achieved, the application threshold is lowered, the maneuvering
flight safety boundary and control stability of the aircraft are improved, and the advanced control requirement of the high-maneuvering aircraft is met.