Implementation Model of Gradient Optimization Algorithm for FAST Reflector Adjustment and Receiving Ratio

Abstract

In view of the fact that the current automatic position adjustment model of Five-hundred-meter Aperture Spherical radio Telescope (FAST)main cable nodes and actuators is only limited to point-to-point model and lacks the establishment of dynamic change adjustment process, under the condition that the FAST Spherical Telescope meets the optimal reception ratio, aiming at 4 400 triangular reflection panels of FAST, nonlinear programming and gradient descent optimization algorithm were used to establish the network adjustment strategy model of FAST reflector deformation for 2 226 main cable nodes of FAST. The specific position of the telescope after deformation was obtained by using the nonlinear optimization algorithm. According to the Fermat principle, the electromagnetic wave reflection in three-dimensional space was simulated, and the receiving ratio of the feed cabin was calculated. The model was observed and tested. The experimental results show that the receiving ratio is 1.43%, the number of reflection panels in the paraboloid illumination area is adjusted to 1 295, and the number of reflection panels that successfully reflect the signal to the feed cabin is 1 295, with a ratio of 100%. After testing, the mean square obtained in this model is 0.308 9, which shows that FAST can freely adjust the autonomous surface in a certain radio frequency band and obtain a high acceptance ratio.

Keywords

• fast active reflector
• ideal paraboloid
• equation coordinate transformation
• nonlinear programming gradient descent optimization algorithm
• fermat's principle
• grasp simulation