- title
- Numerical Simulation of Wear of Total Knee Replacements
- Please use the following URL when quoting:
- https://nbn-resolving.org/urn:nbn:de:bsz:14-qucosa2-792165
- publication_date
- 2022
- Date of submission
- 11.01.2022
- Date of defense
- 27.04.2022
- Abstract (EN)
- In this work we discuss the design and implementation of a computer program to simulate the wear on knee implants over a standardized gait cycle. The contact problems during the gait cycle are formulated using linear elasticity with a linear non-penetration condition. Archard's wear law is used to estimate long-term wear. The algebraic problem after discretization is solved using the Truncated Nonsmooth Newton Multigrid (TNNMG) algorithm. The linear correction step of the algorithm leads to indefinite systems in our application for which we describe a novel heuristic damping algorithm. Our problem sizes allow using a direct solver for these systems and with some modifications we are able to reuse the Cholesky factorization of the system matrix over multiple iterations of the TNNMG algorithm. For our application we are able to significantly improve over the performance of IPopt, a state-of-the-art solver using penalty-based methods. The program is then used for in-silico simulations of wear of two different total knee implant models for which in-vitro results are also available. Both wear volume and spatial distribution of wear that are computed compare well with the experimental results.
- Keywords (DE)
- wear, total knee replacement, simulation, contact
- Abrieb, Knieendoprothese, Simulation, Kontaktprobleme
- Classification (DDC)
- 510
- Classification (RVK)
- SK 220
- Examiner
- Prof. Dr. Oliver Sander
- Prof. Dr. Carsten Gräser
- supervisor
- Prof. Dr. Oliver Sander
- Awarding institution
- Technische Universität Dresden, Dresden
- version
- publizierte Version / Verlagsversion
- URN Qucosa
- urn:nbn:de:bsz:14-qucosa2-792165
- Qucosa date of publication
- 18.05.2022
- Document type
- doctoral_thesis
- Document language
- English
- licence
CC BY 4.0