Optimal design and robust analysis of a net of active devices for micro-vibration control of an on-orbit large space antenna

Abstract

Large deployable antennas are required for the advancement of space communications, Earth observation, radio astronomy and deep space exploration. Most contemporary space antennas have exceeded the size of launching vehicles, leading to the necessity of stowed concepts to overcome the limitation. Many structural models have been investigated by different organizations. Generally, mesh deployable reflectors are currently more mature compared to other foldable solutions. This will be the topic of this paper. Orbital disturbances, perturbations originated by on-board sources and thermal deformations affecting the deployed configuration can deteriorate the accuracy of the communications system. Undesired dynamic behaviour of structural components has to be predicted and counteracted. Therefore, vibration control is a key technology to correct the distortions altering the proper functioning of the system. An intelligent active structure is introduced as a structure configured with distributed actuators and sensors and guided by a controller to modify the dynamic response of the system. In this paper, the supporting structure of a very large mesh reflector is described. A FEM formulation is adopted to assemble the frame and it is validated by comparing it with commercial codes. According to the adopted model, the active elements can be embedded in the middle of the truss elements. Of course, active control of all the devices at the same time requires a control effort which could be not affordable in space applications. However, the effectiveness is not the same for all the actuators. In this study, two cascade optimization procedures are performed to assess the best positioning and velocity feedback gains of the actuators that must be controlled to damp vibrations. In particular, a gradient-based technique is used after both a Genetic Algorithm and a Reinforcement Learning methodology to find the best set of gains for the controller. The objective function is set as a weighted sum of power consumption of the actuators. As a study case, the spacecraft implements a velocity feedback strategy when performing a generic attitude manoeuvre to coordinate the simultaneous action of the devices to ensure the damping performance of the system is enhanced. In addition, the impact of damaged actuators and uncertainties on the mechanical properties of the passive structure has been discussed and analysed.