Surrogate Modeling of a Nonlinear, Biphasic Model of Articular Cartilage with Artificial Neural Networks

Abstract

Abstract The increasing number of cartilage diseases negatively affects the quality of life for a large part of the population. Understanding the mechanical properties of cartilage is a key component in investigating and mitigating the effects of such diseases. To describe the behavior of articular cartilage, a biphasic fiber‐reinforced numerical model based on the Theory of Porous Media (TPM) has been developed. To possibly provide an alternative for the corresponding time‐consuming computational simulations, the suitability of Artificial Neural Networks (ANNs) as surrogate models is investigated. For this purpose, the simulation results of a compression‐relaxation test are compared with the predictions of the ANN.