Abstract
Numerical simulation models are extensively used in the medical device industry to support device design and reduce costs associated with prototype testing. Regulatory agencies such as the US Food and Drug Administration (FDA) require device manufacturers to follow rigorous simulation verification procedures to assess credibility of numerical simulations used to support submissions. Verification techniques are categorized into code and solution verification, the latter employed to ensure that the numerical solution has converged to the exact solution to the mathematical model. These verification techniques are used even though the exact solution cannot be derived. In other words, solution verification is a method to quantify and minimize discretization error, one of the main sources of numerical error.
In this work, we simulate the crimping of a Poly L-lactic Acid (PLLA) stent in the implantation process using an advanced viscoplastic constitutive model to model the PLLA in a quasi-static implicit scheme with Dassault Systèmes Abaqus 2023 using a reference stent geometry from the literature. We monitor and extract the crimping force and peak maximum principal stress in the stent as the quantities of interest for solution verification. We use Richardson extrapolation to estimate the discretization error between the numerical model and the exact solution. Discretization errors for three systematically refined mesh configurations are quantified and we demonstrate that the solution is in the asymptotic range.
