Design, modeling, and optimization of mechanically reconfigurable aperture antennas

The work in this study develops the framework for placement and actuation of novel reconfigurable dual-offset contour beam reflector antennas (DCBRA). Toward that end, the methodology for the antennas´ design is defined. In addition, two separate optimization problems are stated and solved: actuator position optimization and actuation value optimization. For the former, a method termed as greatest error suppression method is proposed where the position of each actuator is decided one by one after each evaluation of the error between the desired subreflector shape and the actual subreflector shape. For the second problem, a mathematical analysis shows that there exists only one optimal configuration. Two optimization techniques are used for the second problem: the simulated annealing algorithm and a simple univariate optimization technique. The univariate technique always generates the same optimal configuration for different initial configurations and it gives the low bound in the evaluation of the error. The simulated annealing algorithm is a stochastic technique used to search for global optimum point. Finally, as an example the results of the proposed optimization techniques are presented for the generation of a subreflector shape corresponding to the geographical outline of Brazil