Inverse optimization of drawing parameters for virtual fabrication of microstructured optical fiber via multi-objective particle swarm optimization algorithm

The microstructured optical fibers (MOFs) fabrication process involves repetitive mapping of the drawing parameters, which is time-consuming, laborious and inefficient, and has become a fundamental obstacle restricting the current design of many MOFs with high excellent performances from being fabricated. This paper adopts a multi-objective particle swarm optimization (MOPSO) algorithm to optimize the drawing parameters inversely according to the target MOFs, to realize fast and precise fabrication. In addition, the quantitative relationship between the drawing parameters is investigated, and guiding suggestions are provided for adjusting the drawing parameters in the experimental fabrication. Ultimately, the MOPSO model is utilized to identify the required drawing parameters for the target MOFs, and an eight-tube node-less hollow core anti-resonant fiber (HC-ARF) is experimentally fabricated with a transmission loss of 0.2 dB/m at the wavelength of 1550 nm and a bending loss of 0.1 dB/m at a bending radius of 5 cm. This study provides an approach to the efficient fabrication of MOFs, accelerating the production of MOFs with novel structures and outstanding performance in a variety of applications.