Particle swarm optimization for realizing bilayer thermal sensors with bulk isotropic materials

Metamaterial-based devices have been extensively researched on account of their novel functions, such as cloaking, concentrating, rotating, and sensing. However, they are usually achieved by employing metamaterials with extreme parameters, critically restricting engineering preparation. To simplify parametric designs, we propose an optimization model with particle swarm algorithms to realize bilayer thermal sensors composed of bulk isotropic materials (circular structure). For this purpose, the fitness function is defined to evaluate the difference between the actual and expected temperatures. By choosing suitable materials for different regions and treating the radii of sensor, inner shell, and outer shell as design variables, we finally minimize the fitness function via particle swarm optimization. The designed scheme is not only easy to implement in applications, but shows excellent performances in both detective accuracy and thermal invisibility, which are well confirmed by finite-element simulations and laboratory experiments. Optimization model can also be flexibly extended to a square case. This method can not only calculate numerical solutions for difficult analytical theories (circular structure), but also calculate optimal solutions for problems without analytical theories (square structure), providing new inspiration for simplifying the design of metamaterials in a wide range of communities.