Optimal stent elastic recoil design using genetic algorithm and artificial neural network simulation

摘要

This paper presents an integrated approach using artificial neural network (ANN) and genetic algorithm (GA) for elastic recoil optimization of stent.Specifically,ANN is used to predict the elastic recoil responses of different stent design configurations and GA relies on these values to search the optimal design configuration.A multilayer back-propagation neural network is trained with numerical results obtained from finite element analysis to establish the correlation between stent design parameters and elastic recoil responses.Subsequently,GA is employed for arriving at optimum configuration of the stent by minimizing the elastic recoil.By integrating ANN with GA,this paper shows that the computational effort to obtain an optimal stent design can be substantially reduced since ANN simulation simplifies the process of finite element analysis,thus this work enables stent design engineers to work more efficiently.