Session: 12-03-01: Minisymposium on Peridynamic Modeling of Materials’ Behavior
Paper Number: 77326
Start Time: Monday, 11:45 AM
77326 - A Fast Convolution-Based Method for Peridynamic Modelling of Pitting Corrosion
We introduce a fast convolution-based method (FCBM) for peridynamic (PD) models of pitting corrosion damage. Based on passivation, salt-layer, and passive-file rupture mechanisms [1], the model simulates pitting corrosion damage as well as autonomous formation of lacy covers. The meshfree method previously used to discretize the equations in the PD corrosion model scales as O(N^2), where N is the number of nodes in the discretization. For realistic problems involving hundreds or thousands of pits, this becomes too costly. In order to simulate realistic problems (cm to m spatial scales, and months/years in time), different discretization methods are needed. By structuring the integrals in terms of convolutions and using FFT to compute, the computational complexity is reduced to O(NlogN). This approach also has significantly lower memory allocation needs: O(N^2), instead of O(N^2) for other methods.
The model is verified against previous simulations which used the meshfree discretization [2][3] and against experimental results. The efficiency gains are impressive: computation time is reduced from days to hours and from years to days. Together with massive reduction in memory allocation, we are now able to simulate corrosion in a 3D cylinder with hundreds of pits growing simultaneously. This is the first demonstration of predicting pitting corrosion damage in samples of engineering relevance. Future steps include coupling the model with the PD fracture model to simulate stress corrosion cracking (SCC) [4].
Presenting Author: Longzhen Wang University of Nebraska-Lincoln
Authors:
Longzhen Wang University of Nebraska-LincolnSiavash Jafarzadeh University of Nebraska Lincoln
Florin Bobaru University of Nebraska Lincoln
A Fast Convolution-Based Method for Peridynamic Modelling of Pitting Corrosion
Paper Type
Technical Presentation