Efficient design of Airyprime chain optical bottle beams based on neural networks.

To meet the demand for stable trapping of multiple particles in complex environments and efficient design of multiple optical bottles, Airyprime chain optical bottle (ACOB) beams are proposed, and their structural features are efficiently designed using a neural network. The evolution properties of ACOB beams and the formation mechanism of chain optical bottles are analyzed, revealing the relationships between the number, maximum size, and formation position of the optical bottles and the optical field parameters. The neural network is trained to learn the nonlinear mapping from the optical field parameters to the structural features of the chain optical bottles. Experimental results further verify the ability of the chain optical bottles to stably trap multiple particles. This study provides new insights into the efficient design and intelligent control of complex optical bottle beams.