Approved
Implementation and development of a gradient shape optimization method for antennas
Robin Johansson (F07)
Start
2011-06-13
Presentation
2011-12-09 13:15
Location:
E:2311
Finished:
2011-12-12
Master's thesis:
(Contact supervisor)
Abstract
To make antennas as good as possible it requires that the dimensions are optimized for desired properties. Today it’s done manually by changing one dimension at a time. This method is time consuming and it would therefore be more convenient if one could optimize all three dimensions at the same time. To do this three cost functions will be derived and manipulated so that the gradient is obtained by a solution to an adjoint problem. One cost function to minimize the return loss, one to maximize the gain in the desired direction and a third that will minimize the side lobes (side lobe suppression). This will be implemented in a MATLAB script that uses an external program that will do the electromagnetic computations. For each cost function there will be a gradient with respect to the geometry that will be used to optimize the dimensions.
Supervisor: Johan Wettergren (RUAG Space AB)
Examiner: Daniel Sjöberg (EIT)