Worst-case sensitivity analysis (WCSA) by particle swarm optimization (PSO): Applications in realistic optimal antenna designs

In practical implementation of engineering designs, the design performances may be subject to tolerances and uncertainties. Relying on a worst-case error tolerance model, this paper proposes the particle swarm optimization (PSO) as an approach of the worst-case sensitivity analysis (WCSA). The proposed method utilizes the absorbing boundary condition in PSO, thereby improving its ability to search the worst-case solution. Numerical tests demonstrate that PSO outperforms GA, vertex prediction, and random search in WCSA problems. The effectiveness of the concepts presented in this paper are examined through the analysis of etching tolerances applied to the design of a 60GHz microstrip patch antenna.