Dynamic Characteristics Analysis Of Planar Membrane Antenna Structures And Research On Intelligent Vibration Control

Due to the advantages of light weight,large storage ratio and easy deployment,large space thin film antenna structure has become one of the research hotspots of space antenna structure at home and abroad in recent years.In order to grasp the influence of membrane prestressing on its dynamic characteristics and the vibration response law of the structure under the interference load,and to use a more intelligent and autonomous control algorithm to control the vibration of the structure,this paper has carried out the related research and done the following work.1.Based on the idea of modeling from element model to component model and then to system model,the finite element models of membrane surface and support frame are established respectively,and the interaction between them is set as multi-point constraint binding relationship,and the finite element model of membrane antenna structure is established.The membrane antenna structure is further simplified as a cantilever model,and the electromechanical coupling dynamic model and the state space equation of the control system of the piezoelectric cantilever structure in physical space and modal space are established.2.On the basis of the finite element model of the membrane antenna structure,the prestress application analysis and modal analysis of the structure are carried out,and the influence of prestress on the vibration frequency of the frame,the vibration frequency of the membrane surface and the change rule of the flatness of the membrane surface in the vibration process is studied under the inertial interference load caused by the orbit maneuver of the spacecraft.3.Based on the dynamic model of piezoelectric cantilever structure and the state space equation of control system,the classical Q-learning algorithm in reinforcement learning is applied to the vibration control of structure.The selection of variables in state space and action space and the definition of reward function are completed.The state space and action space are discretized,and the influence of learning efficiency parameters and subsequent state value function attenuation parameters on the control effect is analyzed.The simulation results show that the Q-learning algorithm has better control effect than the fuzzy control algorithm under the periodic disturbance load and the inertial disturbance load caused by orbit maneuver.4.The Deep Q Network(DQN)algorithm in deep reinforcement learning is used to study the vibration control of piezoelectric cantilever structure under multiple working conditions.The simulation results show that the control effect of dqn algorithm is better than that of Q-learning algorithm when the structure is disturbed by different orbital maneuvers.However,the control effect of Q-learning algorithm is better than dqn algorithm when the structure is subjected to periodic interference loads of different frequencies.This study provides a certain support for the initial design of thin film antenna structure,and provides a research basis for the follow-up engineering application and intelligent vibration control of thin film antenna structure.