An optimization-based reformulation of the classical displacement approach for state update of non-linear material models
For researchers in computational mechanics, this work provides a theoretical unification and optimization-based perspective on existing state update algorithms, though it is incremental in nature.
The paper reformulates the classical displacement-based nested approach for incremental state update in non-linear mechanics as a reduced dual optimization problem, eliminating heuristics and providing insights. It demonstrates the unifying nature of a mathematical programming approach by relating to several recent algorithms.
In this paper, we build on recent work using a mathematical programming approach for incremental state update in analysis of non-linear mechanics models. In particular, we consider quasi-static analysis of continuum problems in the linearized kinematics regime, with non-linear material models described using convex energy functions. We find in this case that the classical displacement-based nested approach for incremental state update can be reformulated as solving a reduced dual optimization problem. This reformulation provides insights into the working of the algorithm, and eliminates the need for some heuristics. An important purpose of this paper is to further illustrate the unifying nature of the mathematical programming approach. We therefore present relationships with several of these types of algorithms recently presented in the literature for incremental state update.