| Since the 1960 s,aerospace technology has developed rapidly,and onboard antenna reflectors have also been studied with the continuous increase of application scenarios.To improve the performance of spaceborne antennas,there are two directions for research:increasing antenna aperture or improving reflector surface accuracy.When the antenna aperture and working band remain unchanged,the size of the antenna gain mainly depends on the accuracy of the reflector surface.The surface accuracy of reflectors is affected by many factors in manufacturing,operation,and other processes.If not actively controlled,it will cause a decrease in the surface accuracy of reflectors,thereby reducing antenna performance.To ensure the accuracy of the reflector surface of spaceborne antennas,active control of the reflector surface has become a key technology that urgently needs to be studied.In this paper,the grid reflector is taken as the research object.The back of the grid reflector is equipped with PZT piezoelectric actuator,which controls its shape and surface by in-plane actuation.The source of shape and surface error of the reflector during the on-orbit operation phase is analyzed.The quasi-static shape and surface control process of the reflector is statics modeled,and the thermal deformation of the reflector under three typical working conditions is analyzed.For the two cases of no power supply number limit and power supply number limit,the specific tasks of studying the related problems encountered in the control process of reflector shape are as follows:(1)The quasi-static adjustment process of the grid reflector is analyzed by statics finite element method,and the thermal deformation of the reflector is analyzed under typical working conditions.Establish a finite element model of the reflector substrate structure,actuator active element,and overall structure for the quasi-static process of reflector surface adjustment;Analyze the sources of surface errors during the on-orbit operation phase of the reflector,and conduct thermal deformation analysis on the reflector under three typical working conditions.From the analysis results,it can be seen that under all three typical operating conditions,significant surface errors of the reflector can be caused,leading to a decrease in surface accuracy and antenna gain.Therefore,it is necessary to use active surface control methods to reduce the surface error of the reflector.(2)A high-precision active control method for reflector surface based on DQN algorithm has been studied.Briefly explain the basic knowledge of DQN algorithm;Using the statics model and thermal deformation analysis knowledge of the reflector overall structure described above,the control optimization model of the reflector surface is derived;Combine the control optimization model with the DQN algorithm for simulation analysis.From the results,it can be seen that using the DQN algorithm with antenna gain as feedback,learning and training the reflector shape control process under different operating conditions can effectively solve the problem of reflector shape control under unknown operating conditions.(3)Based on the assumption of reducing the number of power sources,research has been conducted on the active control of the shape of onboard antenna reflectors.Using the statics model and thermal deformation analysis knowledge of the reflector overall structure described above,the reflector shape control optimization model after reducing the number of power supplies is derived;Provide a detailed introduction to the design process of the genetic simulated annealing algorithm used in this article,and conduct simulation analysis.From the results,it can be seen that the hybrid control method proposed in this article can significantly reduce the reflector surface error,and its convergence speed and accuracy are better than the simulation results of the hybrid control algorithm based on genetic algorithm and least squares method.In summary,this article focuses on the high-precision shape control of spaceborne antenna reflectors.This article conducts mechanical modeling and thermal deformation analysis of reflectors;and propose an active control method for reflector surface based on DQN algorithm;a hybrid optimization method of genetic simulated annealing algorithm and least squares method is proposed to effectively reduce reflector shape error under the condition of limited power supply. |