Multi-objective structural optimization of degradable PLLA vascular stents based on surrogate models

Abstract

Due to their material properties, polymer stents have relatively weak mechanical properties compared to metal stents, leading to inadequate support and premature fracture. A multi-parameter structural optimization approach for poly-L-lactic acid (PLLA) vascular stents was developed based on an extensive dataset of finite element simulations and surrogate models. This study investigated the effect of four design parameters—support ring angle (θ₁ and θ₂), link width, and stent thickness—on the support force (F) and maximum equivalent plastic strain (PEEQ). Surrogate models were employed to establish the functional relationship between the design parameters and the optimization objectives, while a genetic algorithm was used to identify the optimal solution. The results showed that all the proposed surrogate models provided improved predictions of stent structural performance, with the radial basis function model providing optimal results. Compared to the initial structure, the optimized stent exhibited a 24.7% increase in F and a 28% reduction in maximum PEEQ. The prediction errors for both F and PEEQ in the optimal structure were below 5%. The proposed optimization framework effectively enhanced the mechanical performance of the stent by improving structural support and reducing localized plastic deformation, offering a systematic approach for the structural optimization of PLLA vascular stents.

The alternative text for this image may have been generated using AI.