Including design sensitivity into Particle Swarm Optimization: Adaptive gradient-based techniques

For most antennas, fabrication errors can significantly degrade the performance of optimized designs. While error sensitivity is important in the engineering design process, it is rarely incorporated in recent optimization techniques such as Particle Swarm Optimization (PSO). In this paper, we propose an algorithm which incorporates the gradient into PSO. The goal is to optimize the design and guarantee that the solution fulfills the sensitivity criterion simultaneously. The particles in the neighboring area of the global optimum detect the gradient of the global optimum and determine the sensitivity of the solution. The proposed optimization with sensitivity analysis is applied to a test function and patch antenna optimization problem.